US20190065261A1

US20190065261A1 - Technologies for in-processor workload phase detection

Info

Publication number: US20190065261A1
Application number: US15/859,366
Authority: US
Inventors: Ananth S. Narayan; Sagar V. Dalvi; Mrittika Ganguli; Sergiu D. Ghetie
Original assignee: Intel Corp
Current assignee: Intel Corp
Priority date: 2017-08-30
Filing date: 2017-12-30
Publication date: 2019-02-28
Also published as: US20190068509A1; US20190065281A1; CN109426633A; US20190065212A1; DE112018004798T5; US11748172B2; US20190068696A1; US20190065415A1; US11467885B2; US11030017B2; US20190069433A1; CN109428841A; US20190068464A1; CN109428843A; US20190065172A1; WO2019045929A1; US20190068523A1; US10888016B2; US20200192710A1; US20190068444A1

Abstract

Technologies for providing in-processor workload phase detection include a sled having a compute engine, which itself includes a performance monitor unit. The compute engine obtains telemetry data from the performance monitor unit. The performance monitor unit produces telemetry data indicative of performance metrics of the sled during execution of one or more workloads. The telemetry data is indicative of a resource utilization and workload performance by the sled as the workloads are executed. The compute engine determines, from a lookup table indicative of resource utilization phases, a resource utilization phase based on the obtained telemetry data. A workload fingerprint is updated based on the determined resource utilization phase, and the workload fingerprint is output. Other embodiments are also described and claimed.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

The present application claims the benefit of Indian Provisional Patent Application No. 201741030632, filed Aug. 30, 2017 and U.S. Provisional Patent Application No. 62/584,401, filed Nov. 10, 2017.

BACKGROUND

In systems that distribute workloads among multiple compute devices (e.g., in a data center), a centralized server may compose nodes of compute devices to process the workloads. Each node represents a logical aggregation of resources (e.g., compute, storage, acceleration, and the like) provided by each compute device. In such a system, the centralized server may monitor resource utilization in each node and underlying compute device, which allows the centralized server to adjust the allocation of resources for a workload given the reported resource utilization in the compute devices. Typically, doing so involves each computing device collecting resource utilization metrics and reporting, through a network, the collected metrics to the centralized server, which in turn analyzes the metrics to identify a corresponding phase of the workload being executed. Based on the phase, the centralized server performs some action based on the analysis (e.g., adjusting resources in a compute device, assigning a portion of a workload to another compute device, etc.). Such an approach may consume a considerable amount of network resources in the system.

BRIEF DESCRIPTION OF THE DRAWINGS

The concepts described herein are illustrated by way of example and not by way of limitation in the accompanying figures. For simplicity and clarity of illustration, elements illustrated in the figures are not necessarily drawn to scale. Where considered appropriate, reference labels have been repeated among the figures to indicate corresponding or analogous elements.

FIG. 1 is a simplified diagram of at least one embodiment of a data center for executing workloads with disaggregated resources;

FIG. 2 is a simplified diagram of at least one embodiment of a pod of the data center of FIG. 1;

FIG. 3 is a perspective view of at least one embodiment of a rack that may be included in the pod of FIG. 2;

FIG. 4 is a side plan elevation view of the rack of FIG. 3;

FIG. 5 is a perspective view of the rack of FIG. 3 having a sled mounted therein;

FIG. 6 is a is a simplified block diagram of at least one embodiment of a top side of the sled of FIG. 5;

FIG. 7 is a simplified block diagram of at least one embodiment of a bottom side of the sled of FIG. 6;

FIG. 8 is a simplified block diagram of at least one embodiment of a compute sled usable in the data center of FIG. 1;

FIG. 9 is a top perspective view of at least one embodiment of the compute sled of FIG. 8;

FIG. 10 is a simplified block diagram of at least one embodiment of an accelerator sled usable in the data center of FIG. 1;

FIG. 11 is a top perspective view of at least one embodiment of the accelerator sled of FIG. 10;

FIG. 12 is a simplified block diagram of at least one embodiment of a storage sled usable in the data center of FIG. 1;

FIG. 13 is a top perspective view of at least one embodiment of the storage sled of FIG. 12;

FIG. 14 is a simplified block diagram of at least one embodiment of a memory sled usable in the data center of FIG. 1; and

FIG. 15 is a simplified block diagram of a system that may be established within the data center of FIG. 1 to execute workloads with managed nodes composed of disaggregated resources.

FIG. 16 is a simplified block diagram of at least one embodiment of a system for performing in-processor workload phase detection;

FIG. 17 is a simplified block diagram of at least one embodiment of a compute sled of the system of FIG. 16;

FIG. 18 is a simplified block diagram of at least one embodiment of an environment that may be established by the compute sled of FIGS. 16 and 17; and

FIGS. 19-21 are a simplified flow diagram of at least one embodiment of a method for detecting workload phases via a processor of the compute sled of FIGS. 16 and 17.

DETAILED DESCRIPTION OF THE DRAWINGS

While the concepts of the present disclosure are susceptible to various modifications and alternative forms, specific embodiments thereof have been shown by way of example in the drawings and will be described herein in detail. It should be understood, however, that there is no intent to limit the concepts of the present disclosure to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives consistent with the present disclosure and the appended claims.
References in the specification to “one embodiment,” “an embodiment,” “an illustrative embodiment,” etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may or may not necessarily include that particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described. Additionally, it should be appreciated that items included in a list in the form of “at least one A, B, and C” can mean (A); (B); (C); (A and B); (A and C); (B and C); or (A, B, and C). Similarly, items listed in the form of “at least one of A, B, or C” can mean (A); (B); (C); (A and B); (A and C); (B and C); or (A, B, and C).
The disclosed embodiments may be implemented, in some cases, in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried by or stored on a transitory or non-transitory machine-readable (e.g., computer-readable) storage medium, which may be read and executed by one or more processors. A machine-readable storage medium may be embodied as any storage device, mechanism, or other physical structure for storing or transmitting information in a form readable by a machine (e.g., a volatile or non-volatile memory, a media disc, or other media device).
In the drawings, some structural or method features may be shown in specific arrangements and/or orderings. However, it should be appreciated that such specific arrangements and/or orderings may not be required. Rather, in some embodiments, such features may be arranged in a different manner and/or order than shown in the illustrative figures. Additionally, the inclusion of a structural or method feature in a particular figure is not meant to imply that such feature is required in all embodiments and, in some embodiments, may not be included or may be combined with other features.
Referring now to FIG. 1, a data center 100 in which disaggregated resources may cooperatively execute one or more workloads (e.g., applications on behalf of customers) includes multiple pods 110, 120, 130, 140, each of which includes one or more rows of racks. As described in more detail herein, each rack houses multiple sleds, which each may be embodied as a compute device, such as a server, that is primarily equipped with a particular type of resource (e.g., memory devices, data storage devices, accelerator devices, general purpose processors). In the illustrative embodiment, the sleds in each pod 110, 120, 130, 140 are connected to multiple pod switches (e.g., switches that route data communications to and from sleds within the pod). The pod switches, in turn, connect with spine switches 150 that switch communications among pods (e.g., the pods 110, 120, 130, 140) in the data center 100. In some embodiments, the sleds may be connected with a fabric using Intel Omni-Path technology. As described in more detail herein, resources within sleds in the data center 100 may be allocated to a group (referred to herein as a “managed node”) containing resources from one or more other sleds to be collectively utilized in the execution of a workload. The workload can execute as if the resources belonging to the managed node were located on the same sled. The resources in a managed node may even belong to sleds belonging to different racks, and even to different pods 110, 120, 130, 140. Some resources of a single sled may be allocated to one managed node while other resources of the same sled are allocated to a different managed node (e.g., one processor assigned to one managed node and another processor of the same sled assigned to a different managed node). By disaggregating resources to sleds comprised predominantly of a single type of resource (e.g., compute sleds comprising primarily compute resources, memory sleds containing primarily memory resources), and selectively allocating and deallocating the disaggregated resources to form a managed node assigned to execute a workload, the data center 100 provides more efficient resource usage over typical data centers comprised of hyperconverged servers containing compute, memory, storage and perhaps additional resources). As such, the data center 100 may provide greater performance (e.g., throughput, operations per second, latency, etc.) than a typical data center that has the same number of resources.
Referring now to FIG. 2, the pod 110, in the illustrative embodiment, includes a set of rows 200, 210, 220, 230 of racks 240. Each rack 240 may house multiple sleds (e.g., sixteen sleds) and provide power and data connections to the housed sleds, as described in more detail herein. In the illustrative embodiment, the racks in each row 200, 210, 220, 230 are connected to multiple pod switches 250, 260. The pod switch 250 includes a set of ports 252 to which the sleds of the racks of the pod 110 are connected and another set of ports 254 that connect the pod 110 to the spine switches 150 to provide connectivity to other pods in the data center 100. Similarly, the pod switch 260 includes a set of ports 262 to which the sleds of the racks of the pod 110 are connected and a set of ports 264 that connect the pod 110 to the spine switches 150. As such, the use of the pair of switches 250, 260 provides an amount of redundancy to the pod 110. For example, if either of the switches 250, 260 fails, the sleds in the pod 110 may still maintain data communication with the remainder of the data center 100 (e.g., sleds of other pods) through the other switch 250, 260. Furthermore, in the illustrative embodiment, the switches 150, 250, 260 may be embodied as dual-mode optical switches, capable of routing both Ethernet protocol communications carrying Internet Protocol (IP) packets and communications according to a second, high-performance link-layer protocol (e.g., Intel's Omni-Path Architecture's, Infiniband) via optical signaling media of an optical fabric.
It should be appreciated that each of the other pods 120, 130, 140 (as well as any additional pods of the data center 100) may be similarly structured as, and have components similar to, the pod 110 shown in and described in regard to FIG. 2 (e.g., each pod may have rows of racks housing multiple sleds as described above). Additionally, while two pod switches 250, 260 are shown, it should be understood that in other embodiments, each pod 110, 120, 130, 140 may be connected to different number of pod switches (e.g., providing even more failover capacity).
Referring now to FIGS. 3-5, each illustrative rack 240 of the data center 100 includes two elongated support posts 302, 304, which are arranged vertically. For example, the elongated support posts 302, 304 may extend upwardly from a floor of the data center 100 when deployed. The rack 240 also includes one or more horizontal pairs 310 of elongated support arms 312 (identified in FIG. 3 via a dashed ellipse) configured to support a sled of the data center 100 as discussed below. One elongated support arm 312 of the pair of elongated support arms 312 extends outwardly from the elongated support post 302 and the other elongated support arm 312 extends outwardly from the elongated support post 304.
In the illustrative embodiments, each sled of the data center 100 is embodied as a chassis-less sled. That is, each sled has a chassis-less circuit board substrate on which physical resources (e.g., processors, memory, accelerators, storage, etc.) are mounted as discussed in more detail below. As such, the rack 240 is configured to receive the chassis-less sleds. For example, each pair 310 of elongated support arms 312 defines a sled slot 320 of the rack 240, which is configured to receive a corresponding chassis-less sled. To do so, each illustrative elongated support arm 312 includes a circuit board guide 330 configured to receive the chassis-less circuit board substrate of the sled. Each circuit board guide 330 is secured to, or otherwise mounted to, a top side 332 of the corresponding elongated support arm 312. For example, in the illustrative embodiment, each circuit board guide 330 is mounted at a distal end of the corresponding elongated support arm 312 relative to the corresponding elongated support post 302, 304. For clarity of the Figures, not every circuit board guide 330 may be referenced in each Figure.
Each circuit board guide 330 includes an inner wall that defines a circuit board slot 380 configured to receive the chassis-less circuit board substrate of a sled 400 when the sled 400 is received in the corresponding sled slot 320 of the rack 240. To do so, as shown in FIG. 4, a user (or robot) aligns the chassis-less circuit board substrate of an illustrative chassis-less sled 400 to a sled slot 320. The user, or robot, may then slide the chassis-less circuit board substrate forward into the sled slot 320 such that each side edge 414 of the chassis-less circuit board substrate is received in a corresponding circuit board slot 380 of the circuit board guides 330 of the pair 310 of elongated support arms 312 that define the corresponding sled slot 320 as shown in FIG. 4. By having robotically accessible and robotically manipulable sleds comprising disaggregated resources, each type of resource can be upgraded independently of each other and at their own optimized refresh rate. Furthermore, the sleds are configured to blindly mate with power and data communication cables in each rack 240, enhancing their ability to be quickly removed, upgraded, reinstalled, and/or replaced. As such, in some embodiments, the data center 100 may operate (e.g., execute workloads, undergo maintenance and/or upgrades, etc.) without human involvement on the data center floor. In other embodiments, a human may facilitate one or more maintenance or upgrade operations in the data center 100.
It should be appreciated that each circuit board guide 330 is dual sided. That is, each circuit board guide 330 includes an inner wall that defines a circuit board slot 380 on each side of the circuit board guide 330. In this way, each circuit board guide 330 can support a chassis-less circuit board substrate on either side. As such, a single additional elongated support post may be added to the rack 240 to turn the rack 240 into a two-rack solution that can hold twice as many sled slots 320 as shown in FIG. 3. The illustrative rack 240 includes seven pairs 310 of elongated support arms 312 that define a corresponding seven sled slots 320, each configured to receive and support a corresponding sled 400 as discussed above. Of course, in other embodiments, the rack 240 may include additional or fewer pairs 310 of elongated support arms 312 (i.e., additional or fewer sled slots 320). It should be appreciated that because the sled 400 is chassis-less, the sled 400 may have an overall height that is different than typical servers. As such, in some embodiments, the height of each sled slot 320 may be shorter than the height of a typical server (e.g., shorter than a single rank unit, “1U”). That is, the vertical distance between each pair 310 of elongated support arms 312 may be less than a standard rack unit “1U.” Additionally, due to the relative decrease in height of the sled slots 320, the overall height of the rack 240 in some embodiments may be shorter than the height of traditional rack enclosures. For example, in some embodiments, each of the elongated support posts 302, 304 may have a length of six feet or less. Again, in other embodiments, the rack 240 may have different dimensions. Further, it should be appreciated that the rack 240 does not include any walls, enclosures, or the like. Rather, the rack 240 is an enclosure-less rack that is opened to the local environment. Of course, in some cases, an end plate may be attached to one of the elongated support posts 302, 304 in those situations in which the rack 240 forms an end-of-row rack in the data center 100.
In some embodiments, various interconnects may be routed upwardly or downwardly through the elongated support posts 302, 304. To facilitate such routing, each elongated support post 302, 304 includes an inner wall that defines an inner chamber in which the interconnect may be located. The interconnects routed through the elongated support posts 302, 304 may be embodied as any type of interconnects including, but not limited to, data or communication interconnects to provide communication connections to each sled slot 320, power interconnects to provide power to each sled slot 320, and/or other types of interconnects.
The rack 240, in the illustrative embodiment, includes a support platform on which a corresponding optical data connector (not shown) is mounted. Each optical data connector is associated with a corresponding sled slot 320 and is configured to mate with an optical data connector of a corresponding sled 400 when the sled 400 is received in the corresponding sled slot 320. In some embodiments, optical connections between components (e.g., sleds, racks, and switches) in the data center 100 are made with a blind mate optical connection. For example, a door on each cable may prevent dust from contaminating the fiber inside the cable. In the process of connecting to a blind mate optical connector mechanism, the door is pushed open when the end of the cable enters the connector mechanism. Subsequently, the optical fiber inside the cable enters a gel within the connector mechanism and the optical fiber of one cable comes into contact with the optical fiber of another cable within the gel inside the connector mechanism.
The illustrative rack 240 also includes a fan array 370 coupled to the cross-support arms of the rack 240. The fan array 370 includes one or more rows of cooling fans 372, which are aligned in a horizontal line between the elongated support posts 302, 304. In the illustrative embodiment, the fan array 370 includes a row of cooling fans 372 for each sled slot 320 of the rack 240. As discussed above, each sled 400 does not include any on-board cooling system in the illustrative embodiment and, as such, the fan array 370 provides cooling for each sled 400 received in the rack 240. Each rack 240, in the illustrative embodiment, also includes a power supply associated with each sled slot 320. Each power supply is secured to one of the elongated support arms 312 of the pair 310 of elongated support arms 312 that define the corresponding sled slot 320. For example, the rack 240 may include a power supply coupled or secured to each elongated support arm 312 extending from the elongated support post 302. Each power supply includes a power connector configured to mate with a power connector of the sled 400 when the sled 400 is received in the corresponding sled slot 320. In the illustrative embodiment, the sled 400 does not include any on-board power supply and, as such, the power supplies provided in the rack 240 supply power to corresponding sleds 400 when mounted to the rack 240.
Referring now to FIG. 6, the sled 400, in the illustrative embodiment, is configured to be mounted in a corresponding rack 240 of the data center 100 as discussed above. In some embodiments, each sled 400 may be optimized or otherwise configured for performing particular tasks, such as compute tasks, acceleration tasks, data storage tasks, etc. For example, the sled 400 may be embodied as a compute sled 800 as discussed below in regard to FIGS. 8-9, an accelerator sled 1000 as discussed below in regard to FIGS. 10-11, a storage sled 1200 as discussed below in regard to FIGS. 12-13, or as a sled optimized or otherwise configured to perform other specialized tasks, such as a memory sled 1400, discussed below in regard to FIG. 14.
As discussed above, the illustrative sled 400 includes a chassis-less circuit board substrate 602, which supports various physical resources (e.g., electrical components) mounted thereon. It should be appreciated that the circuit board substrate 602 is “chassis-less” in that the sled 400 does not include a housing or enclosure. Rather, the chassis-less circuit board substrate 602 is open to the local environment. The chassis-less circuit board substrate 602 may be formed from any material capable of supporting the various electrical components mounted thereon. For example, in an illustrative embodiment, the chassis-less circuit board substrate 602 is formed from an FR-4 glass-reinforced epoxy laminate material. Of course, other materials may be used to form the chassis-less circuit board substrate 602 in other embodiments.
As discussed in more detail below, the chassis-less circuit board substrate 602 includes multiple features that improve the thermal cooling characteristics of the various electrical components mounted on the chassis-less circuit board substrate 602. As discussed, the chassis-less circuit board substrate 602 does not include a housing or enclosure, which may improve the airflow over the electrical components of the sled 400 by reducing those structures that may inhibit air flow. For example, because the chassis-less circuit board substrate 602 is not positioned in an individual housing or enclosure, there is no backplane (e.g., a backplate of the chassis) to the chassis-less circuit board substrate 602, which could inhibit air flow across the electrical components. Additionally, the chassis-less circuit board substrate 602 has a geometric shape configured to reduce the length of the airflow path across the electrical components mounted to the chassis-less circuit board substrate 602. For example, the illustrative chassis-less circuit board substrate 602 has a width 604 that is greater than a depth 606 of the chassis-less circuit board substrate 602. In one particular embodiment, for example, the chassis-less circuit board substrate 602 has a width of about 21 inches and a depth of about 9 inches, compared to a typical server that has a width of about 17 inches and a depth of about 39 inches. As such, an airflow path 608 that extends from a front edge 610 of the chassis-less circuit board substrate 602 toward a rear edge 612 has a shorter distance relative to typical servers, which may improve the thermal cooling characteristics of the sled 400. Furthermore, although not illustrated in FIG. 6, the various physical resources mounted to the chassis-less circuit board substrate 602 are mounted in corresponding locations such that no two substantively heat-producing electrical components shadow each other as discussed in more detail below. That is, no two electrical components, which produce appreciable heat during operation (i.e., greater than a nominal heat sufficient enough to adversely impact the cooling of another electrical component), are mounted to the chassis-less circuit board substrate 602 linearly in-line with each other along the direction of the airflow path 608 (i.e., along a direction extending from the front edge 610 toward the rear edge 612 of the chassis-less circuit board substrate 602).
As discussed above, the illustrative sled 400 includes one or more physical resources 620 mounted to a top side 650 of the chassis-less circuit board substrate 602. Although two physical resources 620 are shown in FIG. 6, it should be appreciated that the sled 400 may include one, two, or more physical resources 620 in other embodiments. The physical resources 620 may be embodied as any type of processor, controller, or other compute circuit capable of performing various tasks such as compute functions and/or controlling the functions of the sled 400 depending on, for example, the type or intended functionality of the sled 400. For example, as discussed in more detail below, the physical resources 620 may be embodied as high-performance processors in embodiments in which the sled 400 is embodied as a compute sled, as accelerator co-processors or circuits in embodiments in which the sled 400 is embodied as an accelerator sled, storage controllers in embodiments in which the sled 400 is embodied as a storage sled, or a set of memory devices in embodiments in which the sled 400 is embodied as a memory sled.
The sled 400 also includes one or more additional physical resources 630 mounted to the top side 650 of the chassis-less circuit board substrate 602. In the illustrative embodiment, the additional physical resources include a network interface controller (NIC) as discussed in more detail below. Of course, depending on the type and functionality of the sled 400, the physical resources 630 may include additional or other electrical components, circuits, and/or devices in other embodiments.
The physical resources 620 are communicatively coupled to the physical resources 630 via an input/output (I/O) subsystem 622. The I/O subsystem 622 may be embodied as circuitry and/or components to facilitate input/output operations with the physical resources 620, the physical resources 630, and/or other components of the sled 400. For example, the I/O subsystem 622 may be embodied as, or otherwise include, memory controller hubs, input/output control hubs, integrated sensor hubs, firmware devices, communication links (e.g., point-to-point links, bus links, wires, cables, light guides, printed circuit board traces, etc.), and/or other components and subsystems to facilitate the input/output operations. In the illustrative embodiment, the I/O subsystem 622 is embodied as, or otherwise includes, a double data rate 4 (DDR4) data bus or a DDR5 data bus.
In some embodiments, the sled 400 may also include a resource-to-resource interconnect 624. The resource-to-resource interconnect 624 may be embodied as any type of communication interconnect capable of facilitating resource-to-resource communications. In the illustrative embodiment, the resource-to-resource interconnect 624 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the resource-to-resource interconnect 624 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to resource-to-resource communications.
The sled 400 also includes a power connector 640 configured to mate with a corresponding power connector of the rack 240 when the sled 400 is mounted in the corresponding rack 240. The sled 400 receives power from a power supply of the rack 240 via the power connector 640 to supply power to the various electrical components of the sled 400. That is, the sled 400 does not include any local power supply (i.e., an on-board power supply) to provide power to the electrical components of the sled 400. The exclusion of a local or on-board power supply facilitates the reduction in the overall footprint of the chassis-less circuit board substrate 602, which may increase the thermal cooling characteristics of the various electrical components mounted on the chassis-less circuit board substrate 602 as discussed above. In some embodiments, power is provided to the processors 820 through vias directly under the processors 820 (e.g., through the bottom side 750 of the chassis-less circuit board substrate 602), providing an increased thermal budget, additional current and/or voltage, and better voltage control over typical boards.
In some embodiments, the sled 400 may also include mounting features 642 configured to mate with a mounting arm, or other structure, of a robot to facilitate the placement of the sled 600 in a rack 240 by the robot. The mounting features 642 may be embodied as any type of physical structures that allow the robot to grasp the sled 400 without damaging the chassis-less circuit board substrate 602 or the electrical components mounted thereto. For example, in some embodiments, the mounting features 642 may be embodied as non-conductive pads attached to the chassis-less circuit board substrate 602. In other embodiments, the mounting features may be embodied as brackets, braces, or other similar structures attached to the chassis-less circuit board substrate 602. The particular number, shape, size, and/or make-up of the mounting feature 642 may depend on the design of the robot configured to manage the sled 400.
Referring now to FIG. 7, in addition to the physical resources 630 mounted on the top side 650 of the chassis-less circuit board substrate 602, the sled 400 also includes one or more memory devices 720 mounted to a bottom side 750 of the chassis-less circuit board substrate 602. That is, the chassis-less circuit board substrate 602 is embodied as a double-sided circuit board. The physical resources 620 are communicatively coupled to the memory devices 720 via the I/O subsystem 622. For example, the physical resources 620 and the memory devices 720 may be communicatively coupled by one or more vias extending through the chassis-less circuit board substrate 602. Each physical resource 620 may be communicatively coupled to a different set of one or more memory devices 720 in some embodiments. Alternatively, in other embodiments, each physical resource 620 may be communicatively coupled to each memory devices 720.
The memory devices 720 may be embodied as any type of memory device capable of storing data for the physical resources 620 during operation of the sled 400, such as any type of volatile (e.g., dynamic random access memory (DRAM), etc.) or non-volatile memory. Volatile memory may be a storage medium that requires power to maintain the state of data stored by the medium. Non-limiting examples of volatile memory may include various types of random access memory (RAM), such as dynamic random access memory (DRAM) or static random access memory (SRAM). One particular type of DRAM that may be used in a memory module is synchronous dynamic random access memory (SDRAM). In particular embodiments, DRAM of a memory component may comply with a standard promulgated by JEDEC, such as JESD79F for DDR SDRAM, JESD79-2F for DDR2 SDRAM, JESD79-3F for DDR3 SDRAM, JESD79-4A for DDR4 SDRAM, JESD209 for Low Power DDR (LPDDR), JESD209-2 for LPDDR2, JESD209-3 for LPDDR3, and JESD209-4 for LPDDR4 (these standards are available at www.jedec.org). Such standards (and similar standards) may be referred to as DDR-based standards and communication interfaces of the storage devices that implement such standards may be referred to as DDR-based interfaces.
In one embodiment, the memory device is a block addressable memory device, such as those based on NAND or NOR technologies. A memory device may also include next-generation nonvolatile devices, such as Intel 3D XPoint™ memory or other byte addressable write-in-place nonvolatile memory devices. In one embodiment, the memory device may be or may include memory devices that use chalcogenide glass, multi-threshold level NAND flash memory, NOR flash memory, single or multi-level Phase Change Memory (PCM), a resistive memory, nanowire memory, ferroelectric transistor random access memory (FeTRAM), anti-ferroelectric memory, magnetoresistive random access memory (MRAM) memory that incorporates memristor technology, resistive memory including the metal oxide base, the oxygen vacancy base and the conductive bridge Random Access Memory (CB-RAM), or spin transfer torque (STT)-MRAM, a spintronic magnetic junction memory based device, a magnetic tunneling junction (MTJ) based device, a DW (Domain Wall) and SOT (Spin Orbit Transfer) based device, a thyristor based memory device, or a combination of any of the above, or other memory. The memory device may refer to the die itself and/or to a packaged memory product. In some embodiments, the memory device may comprise a transistor-less stackable cross point architecture in which memory cells sit at the intersection of word lines and bit lines and are individually addressable and in which bit storage is based on a change in bulk resistance.
Referring now to FIG. 8, in some embodiments, the sled 400 may be embodied as a compute sled 800. The compute sled 800 is optimized, or otherwise configured, to perform compute tasks. Of course, as discussed above, the compute sled 800 may rely on other sleds, such as acceleration sleds and/or storage sleds, to perform such compute tasks. The compute sled 800 includes various physical resources (e.g., electrical components) similar to the physical resources of the sled 400, which have been identified in FIG. 8 using the same reference numbers. The description of such components provided above in regard to FIGS. 6 and 7 applies to the corresponding components of the compute sled 800 and is not repeated herein for clarity of the description of the compute sled 800.
In the illustrative compute sled 800, the physical resources 620 are embodied as processors 820. Although only two processors 820 are shown in FIG. 8, it should be appreciated that the compute sled 800 may include additional processors 820 in other embodiments. Illustratively, the processors 820 are embodied as high-performance processors 820 and may be configured to operate at a relatively high power rating. Although the processors 820 generate additional heat operating at power ratings greater than typical processors (which operate at around 155-230 W), the enhanced thermal cooling characteristics of the chassis-less circuit board substrate 602 discussed above facilitate the higher power operation. For example, in the illustrative embodiment, the processors 820 are configured to operate at a power rating of at least 250 W. In some embodiments, the processors 820 may be configured to operate at a power rating of at least 350 W.
In some embodiments, the compute sled 800 may also include a processor-to-processor interconnect 842. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the processor-to-processor interconnect 842 may be embodied as any type of communication interconnect capable of facilitating processor-to-processor interconnect 842 communications. In the illustrative embodiment, the processor-to-processor interconnect 842 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the processor-to-processor interconnect 842 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications.
The compute sled 800 also includes a communication circuit 830. The illustrative communication circuit 830 includes a network interface controller (NIC) 832, which may also be referred to as a host fabric interface (HFI). The NIC 832 may be embodied as, or otherwise include, any type of integrated circuit, discrete circuits, controller chips, chipsets, add-in-boards, daughtercards, network interface cards, other devices that may be used by the compute sled 800 to connect with another compute device (e.g., with other sleds 400). In some embodiments, the NIC 832 may be embodied as part of a system-on-a-chip (SoC) that includes one or more processors, or included on a multichip package that also contains one or more processors. In some embodiments, the NIC 832 may include a local processor (not shown) and/or a local memory (not shown) that are both local to the NIC 832. In such embodiments, the local processor of the NIC 832 may be capable of performing one or more of the functions of the processors 820. Additionally or alternatively, in such embodiments, the local memory of the NIC 832 may be integrated into one or more components of the compute sled at the board level, socket level, chip level, and/or other levels.
The communication circuit 830 is communicatively coupled to an optical data connector 834. The optical data connector 834 is configured to mate with a corresponding optical data connector of the rack 240 when the compute sled 800 is mounted in the rack 240. Illustratively, the optical data connector 834 includes a plurality of optical fibers which lead from a mating surface of the optical data connector 834 to an optical transceiver 836. The optical transceiver 836 is configured to convert incoming optical signals from the rack-side optical data connector to electrical signals and to convert electrical signals to outgoing optical signals to the rack-side optical data connector. Although shown as forming part of the optical data connector 834 in the illustrative embodiment, the optical transceiver 836 may form a portion of the communication circuit 830 in other embodiments.
In some embodiments, the compute sled 800 may also include an expansion connector 840. In such embodiments, the expansion connector 840 is configured to mate with a corresponding connector of an expansion chassis-less circuit board substrate to provide additional physical resources to the compute sled 800. The additional physical resources may be used, for example, by the processors 820 during operation of the compute sled 800. The expansion chassis-less circuit board substrate may be substantially similar to the chassis-less circuit board substrate 602 discussed above and may include various electrical components mounted thereto. The particular electrical components mounted to the expansion chassis-less circuit board substrate may depend on the intended functionality of the expansion chassis-less circuit board substrate. For example, the expansion chassis-less circuit board substrate may provide additional compute resources, memory resources, and/or storage resources. As such, the additional physical resources of the expansion chassis-less circuit board substrate may include, but is not limited to, processors, memory devices, storage devices, and/or accelerator circuits including, for example, field programmable gate arrays (FPGA), application-specific integrated circuits (ASICs), security co-processors, graphics processing units (GPUs), machine learning circuits, or other specialized processors, controllers, devices, and/or circuits.
Referring now to FIG. 9, an illustrative embodiment of the compute sled 800 is shown. As shown, the processors 820, communication circuit 830, and optical data connector 834 are mounted to the top side 650 of the chassis-less circuit board substrate 602. Any suitable attachment or mounting technology may be used to mount the physical resources of the compute sled 800 to the chassis-less circuit board substrate 602. For example, the various physical resources may be mounted in corresponding sockets (e.g., a processor socket), holders, or brackets. In some cases, some of the electrical components may be directly mounted to the chassis-less circuit board substrate 602 via soldering or similar techniques.
As discussed above, the individual processors 820 and communication circuit 830 are mounted to the top side 650 of the chassis-less circuit board substrate 602 such that no two heat-producing, electrical components shadow each other. In the illustrative embodiment, the processors 820 and communication circuit 830 are mounted in corresponding locations on the top side 650 of the chassis-less circuit board substrate 602 such that no two of those physical resources are linearly in-line with others along the direction of the airflow path 608. It should be appreciated that, although the optical data connector 834 is in-line with the communication circuit 830, the optical data connector 834 produces no or nominal heat during operation.
The memory devices 720 of the compute sled 800 are mounted to the bottom side 750 of the of the chassis-less circuit board substrate 602 as discussed above in regard to the sled 400. Although mounted to the bottom side 750, the memory devices 720 are communicatively coupled to the processors 820 located on the top side 650 via the I/O subsystem 622. Because the chassis-less circuit board substrate 602 is embodied as a double-sided circuit board, the memory devices 720 and the processors 820 may be communicatively coupled by one or more vias, connectors, or other mechanisms extending through the chassis-less circuit board substrate 602. Of course, each processor 820 may be communicatively coupled to a different set of one or more memory devices 720 in some embodiments. Alternatively, in other embodiments, each processor 820 may be communicatively coupled to each memory device 720. In some embodiments, the memory devices 720 may be mounted to one or more memory mezzanines on the bottom side of the chassis-less circuit board substrate 602 and may interconnect with a corresponding processor 820 through a ball-grid array.
Each of the processors 820 includes a heatsink 850 secured thereto. Due to the mounting of the memory devices 720 to the bottom side 750 of the chassis-less circuit board substrate 602 (as well as the vertical spacing of the sleds 400 in the corresponding rack 240), the top side 650 of the chassis-less circuit board substrate 602 includes additional “free” area or space that facilitates the use of heatsinks 850 having a larger size relative to traditional heatsinks used in typical servers. Additionally, due to the improved thermal cooling characteristics of the chassis-less circuit board substrate 602, none of the processor heatsinks 850 include cooling fans attached thereto. That is, each of the heatsinks 850 is embodied as a fan-less heatsinks.
Referring now to FIG. 10, in some embodiments, the sled 400 may be embodied as an accelerator sled 1000. The accelerator sled 1000 is optimized, or otherwise configured, to perform specialized compute tasks, such as machine learning, encryption, hashing, or other computational-intensive task. In some embodiments, for example, a compute sled 800 may offload tasks to the accelerator sled 1000 during operation. The accelerator sled 1000 includes various components similar to components of the sled 400 and/or compute sled 800, which have been identified in FIG. 10 using the same reference numbers. The description of such components provided above in regard to FIGS. 6, 7, and 8 apply to the corresponding components of the accelerator sled 1000 and is not repeated herein for clarity of the description of the accelerator sled 1000.
In the illustrative accelerator sled 1000, the physical resources 620 are embodied as accelerator circuits 1020. Although only two accelerator circuits 1020 are shown in FIG. 10, it should be appreciated that the accelerator sled 1000 may include additional accelerator circuits 1020 in other embodiments. For example, as shown in FIG. 11, the accelerator sled 1000 may include four accelerator circuits 1020 in some embodiments. The accelerator circuits 1020 may be embodied as any type of processor, co-processor, compute circuit, or other device capable of performing compute or processing operations. For example, the accelerator circuits 1020 may be embodied as, for example, field programmable gate arrays (FPGA), application-specific integrated circuits (ASICs), security co-processors, graphics processing units (GPUs), machine learning circuits, or other specialized processors, controllers, devices, and/or circuits.
In some embodiments, the accelerator sled 1000 may also include an accelerator-to-accelerator interconnect 1042. Similar to the resource-to-resource interconnect 624 of the sled 600 discussed above, the accelerator-to-accelerator interconnect 1042 may be embodied as any type of communication interconnect capable of facilitating accelerator-to-accelerator communications. In the illustrative embodiment, the accelerator-to-accelerator interconnect 1042 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the accelerator-to-accelerator interconnect 1042 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications. In some embodiments, the accelerator circuits 1020 may be daisy-chained with a primary accelerator circuit 1020 connected to the NIC 832 and memory 720 through the I/O subsystem 622 and a secondary accelerator circuit 1020 connected to the NIC 832 and memory 720 through a primary accelerator circuit 1020.
Referring now to FIG. 11, an illustrative embodiment of the accelerator sled 1000 is shown. As discussed above, the accelerator circuits 1020, communication circuit 830, and optical data connector 834 are mounted to the top side 650 of the chassis-less circuit board substrate 602. Again, the individual accelerator circuits 1020 and communication circuit 830 are mounted to the top side 650 of the chassis-less circuit board substrate 602 such that no two heat-producing, electrical components shadow each other as discussed above. The memory devices 720 of the accelerator sled 1000 are mounted to the bottom side 750 of the of the chassis-less circuit board substrate 602 as discussed above in regard to the sled 600. Although mounted to the bottom side 750, the memory devices 720 are communicatively coupled to the accelerator circuits 1020 located on the top side 650 via the I/O subsystem 622 (e.g., through vias). Further, each of the accelerator circuits 1020 may include a heatsink 1070 that is larger than a traditional heatsink used in a server. As discussed above with reference to the heatsinks 870, the heatsinks 1070 may be larger than tradition heatsinks because of the “free” area provided by the memory devices 750 being located on the bottom side 750 of the chassis-less circuit board substrate 602 rather than on the top side 650.
Referring now to FIG. 12, in some embodiments, the sled 400 may be embodied as a storage sled 1200. The storage sled 1200 is optimized, or otherwise configured, to store data in a data storage 1250 local to the storage sled 1200. For example, during operation, a compute sled 800 or an accelerator sled 1000 may store and retrieve data from the data storage 1250 of the storage sled 1200. The storage sled 1200 includes various components similar to components of the sled 400 and/or the compute sled 800, which have been identified in FIG. 12 using the same reference numbers. The description of such components provided above in regard to FIGS. 6, 7, and 8 apply to the corresponding components of the storage sled 1200 and is not repeated herein for clarity of the description of the storage sled 1200.
In the illustrative storage sled 1200, the physical resources 620 are embodied as storage controllers 1220. Although only two storage controllers 1220 are shown in FIG. 12, it should be appreciated that the storage sled 1200 may include additional storage controllers 1220 in other embodiments. The storage controllers 1220 may be embodied as any type of processor, controller, or control circuit capable of controlling the storage and retrieval of data into the data storage 1250 based on requests received via the communication circuit 830. In the illustrative embodiment, the storage controllers 1220 are embodied as relatively low-power processors or controllers. For example, in some embodiments, the storage controllers 1220 may be configured to operate at a power rating of about 75 watts.
In some embodiments, the storage sled 1200 may also include a controller-to-controller interconnect 1242. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the controller-to-controller interconnect 1242 may be embodied as any type of communication interconnect capable of facilitating controller-to-controller communications. In the illustrative embodiment, the controller-to-controller interconnect 1242 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the controller-to-controller interconnect 1242 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications.
Referring now to FIG. 13, an illustrative embodiment of the storage sled 1200 is shown. In the illustrative embodiment, the data storage 1250 is embodied as, or otherwise includes, a storage cage 1252 configured to house one or more solid state drives (SSDs) 1254. To do so, the storage cage 1252 includes a number of mounting slots 1256, each of which is configured to receive a corresponding solid state drive 1254. Each of the mounting slots 1256 includes a number of drive guides 1258 that cooperate to define an access opening 1260 of the corresponding mounting slot 1256. The storage cage 1252 is secured to the chassis-less circuit board substrate 602 such that the access openings face away from (i.e., toward the front of) the chassis-less circuit board substrate 602. As such, solid state drives 1254 are accessible while the storage sled 1200 is mounted in a corresponding rack 204. For example, a solid state drive 1254 may be swapped out of a rack 240 (e.g., via a robot) while the storage sled 1200 remains mounted in the corresponding rack 240.
The storage cage 1252 illustratively includes sixteen mounting slots 1256 and is capable of mounting and storing sixteen solid state drives 1254. Of course, the storage cage 1252 may be configured to store additional or fewer solid state drives 1254 in other embodiments. Additionally, in the illustrative embodiment, the solid state drivers are mounted vertically in the storage cage 1252, but may be mounted in the storage cage 1252 in a different orientation in other embodiments. Each solid state drive 1254 may be embodied as any type of data storage device capable of storing long term data. To do so, the solid state drives 1254 may include volatile and non-volatile memory devices discussed above.
As shown in FIG. 13, the storage controllers 1220, the communication circuit 830, and the optical data connector 834 are illustratively mounted to the top side 650 of the chassis-less circuit board substrate 602. Again, as discussed above, any suitable attachment or mounting technology may be used to mount the electrical components of the storage sled 1200 to the chassis-less circuit board substrate 602 including, for example, sockets (e.g., a processor socket), holders, brackets, soldered connections, and/or other mounting or securing techniques.
As discussed above, the individual storage controllers 1220 and the communication circuit 830 are mounted to the top side 650 of the chassis-less circuit board substrate 602 such that no two heat-producing, electrical components shadow each other. For example, the storage controllers 1220 and the communication circuit 830 are mounted in corresponding locations on the top side 650 of the chassis-less circuit board substrate 602 such that no two of those electrical components are linearly in-line with other along the direction of the airflow path 608.
The memory devices 720 of the storage sled 1200 are mounted to the bottom side 750 of the of the chassis-less circuit board substrate 602 as discussed above in regard to the sled 400. Although mounted to the bottom side 750, the memory devices 720 are communicatively coupled to the storage controllers 1220 located on the top side 650 via the I/O subsystem 622. Again, because the chassis-less circuit board substrate 602 is embodied as a double-sided circuit board, the memory devices 720 and the storage controllers 1220 may be communicatively coupled by one or more vias, connectors, or other mechanisms extending through the chassis-less circuit board substrate 602. Each of the storage controllers 1220 includes a heatsink 1270 secured thereto. As discussed above, due to the improved thermal cooling characteristics of the chassis-less circuit board substrate 602 of the storage sled 1200, none of the heatsinks 1270 include cooling fans attached thereto. That is, each of the heatsinks 1270 is embodied as a fan-less heatsink.
Referring now to FIG. 14, in some embodiments, the sled 400 may be embodied as a memory sled 1400. The storage sled 1400 is optimized, or otherwise configured, to provide other sleds 400 (e.g., compute sleds 800, accelerator sleds 1000, etc.) with access to a pool of memory (e.g., in two or more sets 1430, 1432 of memory devices 720) local to the memory sled 1200. For example, during operation, a compute sled 800 or an accelerator sled 1000 may remotely write to and/or read from one or more of the memory sets 1430, 1432 of the memory sled 1200 using a logical address space that maps to physical addresses in the memory sets 1430, 1432. The memory sled 1400 includes various components similar to components of the sled 400 and/or the compute sled 800, which have been identified in FIG. 14 using the same reference numbers. The description of such components provided above in regard to FIGS. 6, 7, and 8 apply to the corresponding components of the memory sled 1400 and is not repeated herein for clarity of the description of the memory sled 1400.
In the illustrative memory sled 1400, the physical resources 620 are embodied as memory controllers 1420. Although only two memory controllers 1420 are shown in FIG. 14, it should be appreciated that the memory sled 1400 may include additional memory controllers 1420 in other embodiments. The memory controllers 1420 may be embodied as any type of processor, controller, or control circuit capable of controlling the writing and reading of data into the memory sets 1430, 1432 based on requests received via the communication circuit 830. In the illustrative embodiment, each storage controller 1220 is connected to a corresponding memory set 1430, 1432 to write to and read from memory devices 720 within the corresponding memory set 1430, 1432 and enforce any permissions (e.g., read, write, etc.) associated with sled 400 that has sent a request to the memory sled 1400 to perform a memory access operation (e.g., read or write).
In some embodiments, the memory sled 1400 may also include a controller-to-controller interconnect 1442. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the controller-to-controller interconnect 1442 may be embodied as any type of communication interconnect capable of facilitating controller-to-controller communications. In the illustrative embodiment, the controller-to-controller interconnect 1442 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the controller-to-controller interconnect 1442 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications. As such, in some embodiments, a memory controller 1420 may access, through the controller-to-controller interconnect 1442, memory that is within the memory set 1432 associated with another memory controller 1420. In some embodiments, a scalable memory controller is made of multiple smaller memory controllers, referred to herein as “chiplets”, on a memory sled (e.g., the memory sled 1400). The chiplets may be interconnected (e.g., using EMIB (Embedded Multi-Die Interconnect Bridge)). The combined chiplet memory controller may scale up to a relatively large number of memory controllers and I/O ports, (e.g., up to 16 memory channels). In some embodiments, the memory controllers 1420 may implement a memory interleave (e.g., one memory address is mapped to the memory set 1430, the next memory address is mapped to the memory set 1432, and the third address is mapped to the memory set 1430, etc.). The interleaving may be managed within the memory controllers 1420, or from CPU sockets (e.g., of the compute sled 800) across network links to the memory sets 1430, 1432, and may improve the latency associated with performing memory access operations as compared to accessing contiguous memory addresses from the same memory device.
Further, in some embodiments, the memory sled 1400 may be connected to one or more other sleds 400 (e.g., in the same rack 240 or an adjacent rack 240) through a waveguide, using the waveguide connector 1480. In the illustrative embodiment, the waveguides are 64 millimeter waveguides that provide 16 Rx (i.e., receive) lanes and 16 Rt (i.e., transmit) lanes. Each lane, in the illustrative embodiment, is either 16 Ghz or 32 Ghz. In other embodiments, the frequencies may be different. Using a waveguide may provide high throughput access to the memory pool (e.g., the memory sets 1430, 1432) to another sled (e.g., a sled 400 in the same rack 240 or an adjacent rack 240 as the memory sled 1400) without adding to the load on the optical data connector 834.
Referring now to FIG. 15, a system for executing one or more workloads (e.g., applications) may be implemented in accordance with the data center 100. In the illustrative embodiment, the system 1510 includes an orchestrator server 1520, which may be embodied as a managed node comprising a compute device (e.g., a compute sled 800) executing management software (e.g., a cloud operating environment, such as OpenStack) that is communicatively coupled to multiple sleds 400 including a large number of compute sleds 1530 (e.g., each similar to the compute sled 800), memory sleds 1540 (e.g., each similar to the memory sled 1400), accelerator sleds 1550 (e.g., each similar to the memory sled 1000), and storage sleds 1560 (e.g., each similar to the storage sled 1200). One or more of the sleds 1530, 1540, 1550, 1560 may be grouped into a managed node 1570, such as by the orchestrator server 1520, to collectively perform a workload (e.g., an application 1532 executed in a virtual machine or in a container). The managed node 1570 may be embodied as an assembly of physical resources 620, such as processors 820, memory resources 720, accelerator circuits 1020, or data storage 1250, from the same or different sleds 400. Further, the managed node may be established, defined, or “spun up” by the orchestrator server 1520 at the time a workload is to be assigned to the managed node or at any other time, and may exist regardless of whether any workloads are presently assigned to the managed node. In the illustrative embodiment, the orchestrator server 1520 may selectively allocate and/or deallocate physical resources 620 from the sleds 400 and/or add or remove one or more sleds 400 from the managed node 1570 as a function of quality of service (QoS) targets (e.g., performance targets associated with a throughput, latency, instructions per second, etc.) associated with a service level agreement for the workload (e.g., the application 1532). In doing so, the orchestrator server 1520 may receive telemetry data indicative of performance conditions (e.g., throughput, latency, instructions per second, etc.) in each sled 400 of the managed node 1570 and compare the telemetry data to the quality of service targets to determine whether the quality of service targets are being satisfied. If the so, the orchestrator server 1520 may additionally determine whether one or more physical resources may be deallocated from the managed node 1570 while still satisfying the QoS targets, thereby freeing up those physical resources for use in another managed node (e.g., to execute a different workload). Alternatively, if the QoS targets are not presently satisfied, the orchestrator server 1520 may determine to dynamically allocate additional physical resources to assist in the execution of the workload (e.g., the application 1532) while the workload is executing
Additionally, in some embodiments, the orchestrator server 1520 may identify trends in the resource utilization of the workload (e.g., the application 1532), such as by identifying phases of execution (e.g., time periods in which different operations, each having different resource utilizations characteristics, are performed) of the workload (e.g., the application 1532) and pre-emptively identifying available resources in the data center 100 and allocating them to the managed node 1570 (e.g., within a predefined time period of the associated phase beginning). In some embodiments, the orchestrator server 1520 may model performance based on various latencies and a distribution scheme to place workloads among compute sleds and other resources (e.g., accelerator sleds, memory sleds, storage sleds) in the data center 100. For example, the orchestrator server 1520 may utilize a model that accounts for the performance of resources on the sleds 400 (e.g., FPGA performance, memory access latency, etc.) and the performance (e.g., congestion, latency, bandwidth) of the path through the network to the resource (e.g., FPGA). As such, the orchestrator server 1520 may determine which resource(s) should be used with which workloads based on the total latency associated with each potential resource available in the data center 100 (e.g., the latency associated with the performance of the resource itself in addition to the latency associated with the path through the network between the compute sled executing the workload and the sled 400 on which the resource is located).
In some embodiments, the orchestrator server 1520 may generate a map of heat generation in the data center 100 using telemetry data (e.g., temperatures, fan speeds, etc.) reported from the sleds 400 and allocate resources to managed nodes as a function of the map of heat generation and predicted heat generation associated with different workloads, to maintain a target temperature and heat distribution in the data center 100. Additionally or alternatively, in some embodiments, the orchestrator server 1520 may organize received telemetry data into a hierarchical model that is indicative of a relationship between the managed nodes (e.g., a spatial relationship such as the physical locations of the resources of the managed nodes within the data center 100 and/or a functional relationship, such as groupings of the managed nodes by the customers the managed nodes provide services for, the types of functions typically performed by the managed nodes, managed nodes that typically share or exchange workloads among each other, etc.). Based on differences in the physical locations and resources in the managed nodes, a given workload may exhibit different resource utilizations (e.g., cause a different internal temperature, use a different percentage of processor or memory capacity) across the resources of different managed nodes. The orchestrator server 1520 may determine the differences based on the telemetry data stored in the hierarchical model and factor the differences into a prediction of future resource utilization of a workload if the workload is reassigned from one managed node to another managed node, to accurately balance resource utilization in the data center 100.
To reduce the computational load on the orchestrator server 1520 and the data transfer load on the network, in some embodiments, the orchestrator server 1520 may send self-test information to the sleds 400 to enable each sled 400 to locally (e.g., on the sled 400) determine whether telemetry data generated by the sled 400 satisfies one or more conditions (e.g., an available capacity that satisfies a predefined threshold, a temperature that satisfies a predefined threshold, etc.). Each sled 400 may then report back a simplified result (e.g., yes or no) to the orchestrator server 1520, which the orchestrator server 1520 may utilize in determining the allocation of resources to managed nodes.
Referring now to FIG. 16, a system 1610 for providing in-processor workload phase detection may be implemented in accordance with the data center 100 described above with reference to FIG. 1. In the illustrative embodiment, the system 1610 includes an orchestrator server 1620 communicatively coupled to multiple sleds including a compute sled 1630, a memory sled 1640, a storage sled 1650, and an accelerator sled 1660. One or more of the sleds 1630, 1640, 1650 and 1660 may be grouped into a managed node, such as by the orchestrator server 1620, to collectively perform a workload, such as an application. A managed node may be embodied as an assembly of resources, such as compute resources, memory resources, storage resources, or other resources, from the same or different sleds or racks. Further, a managed node may be established, defined, or “spun up” by the orchestrator server 1620 at the time a workload is to be assigned to the managed node or at any other time, and may exist regardless of whether any workloads are presently assigned to the managed node. The system 1610 may be located in a data center and provide storage and compute services (e.g., cloud services) to a client device 1614 that is in communication with the system 1610 through a network 1612. The orchestrator server 1620 may support a cloud operating environment, such as OpenStack, and managed nodes established by the orchestrator server 1620 may execute one or more applications or processes (i.e., workloads), such as in virtual machines or containers, on behalf of a user of the client device 1614.
In the illustrative embodiment, the compute sled 1630 includes a central processing unit (CPU) 1632 (e.g., a processor or other device or circuitry capable of performing a series of operations) that executes a workload (e.g., the application 1634). The memory sled 1640 includes one or more memory devices 1642 (e.g., non-volatile memory, such as byte-addressable, write in-place non-volatile memory, or volatile memory, such as dynamic random-access memory (DRAM)). The storage sled 1650 includes one or more storage devices 1652, (e.g., hard disk drives (HDDs), solid state drives (SSDs), etc.). The accelerator sled 1660, in the illustrative embodiment, includes one or more accelerator devices 1662. As such, each accelerator device 1662 may be embodied as any device or circuitry (e.g., a specialized processor, an FPGA, an ASIC, a graphics processing unit (GPU), reconfigurable hardware, etc.) capable of accelerating the execution of a function.
Referring now to FIG. 17, the compute sled 1630 may be embodied as any type of compute device capable of performing the functions described herein, including receiving telemetry data from a performance monitor unit executing in the compute sled 1630, determining, from a phase lookup table in the compute sled 1630, a resource utilization phase of a workload based on the telemetry data, updating a workload fingerprint based on the resource utilization phase, and outputting the workload fingerprint to an area of memory in the compute sled 1630 reserved by the performance monitor unit.
As shown in FIG. 17, the illustrative compute sled 1630 includes a compute engine 1702, an input/output (I/O) subsystem 1712, communication circuitry 1714, and one or more data storage devices 1718. Of course, in other embodiments, the compute sled 1630 may include other or additional components, such as those commonly found in a computer (e.g., display, peripheral devices, etc.). Additionally, in some embodiments, one or more of the illustrative components may be incorporated in, or otherwise form a portion of, another component.
The compute engine 1702 may be embodied as any type of device or collection of devices capable of performing various compute functions described below. In some embodiments, the compute engine 1702 may be embodied as a single device such as an integrated circuit, an embedded system, a field-programmable gate array (FPGA), a system-on-a-chip (SoC), or other integrated system or device. Additionally, in some embodiments, the compute engine 1702 includes or is embodied as a processor 1704 (e.g., the CPU 1632) and a memory 1710. The processor 1704 may be embodied as any type of processor capable of performing the functions described herein. For example, the processor 1704 may be embodied as a single or multi-core processor(s), a microcontroller, or other processor or processing/controlling circuit. In some embodiments, the processor 1704 may be embodied as, include, or be coupled to an FPGA, an application specific integrated circuit (ASIC), reconfigurable hardware or hardware circuitry, or other specialized hardware to facilitate performance of the functions described herein. Additionally, in the illustrative embodiment, the processor 1704 includes a phase detection logic unit 1605, which may be embodied as any circuitry or device (e.g., a microcontroller, co-processor, etc.).
As further described herein, the phase detection logic unit 1705 may program components in the processor 1704 to monitor resource utilization during execution of a workload and compare the monitored utilization against a phase lookup table to determine a present workload phase. The processor 1704 also includes one or more model-specific registers (MSRs) 1706. The MSRs 1706 may be embodied as instruction set architecture (ISA) registers (e.g., x86 instruction set registers) that control one or more operations of the processor 1704. Illustratively, the MSRs 1706 include a control MSR 1707 that stores values that the processor 1704 may interpret to perform some operation as a function of the value, such as starting or pausing counters, reserving areas of memory, and the like. The processor 1704 also includes one or more general purpose registers (GPRs) to store data and addresses during operation of the processor 1704. Further, the processor 1704 includes a performance monitor unit (PMU) 1709, which, in operation, monitors various telemetry data indicative of the performance and conditions in the compute sled (e.g., the number of cache hits/misses, cycles per instruction, cache occupancy, core frequency, etc.) using counters during the execution of a workload. The PMU 1709 may be embodied as any circuitry or device that receives performance event signals from the processor 1704.
The memory 1710 may be embodied as any type of volatile or non-volatile memory or data storage capable of performing the functions described herein. Volatile memory may be a storage medium that requires power to maintain the state of data stored by the medium. Non-limiting examples of volatile memory may include various types of random access memory (RAM), such as dynamic random access memory (DRAM) or static random access memory (SRAM). One particular type of DRAM that may be used in a memory module is synchronous dynamic random access memory (SDRAM). In particular embodiments, DRAM of a memory component may comply with a standard promulgated by JEDEC, such as JESD79F for DDR SDRAM, JESD79-2F for DDR2 SDRAM, JESD79-3F for DDR3 SDRAM, JESD79-4A for DDR4 SDRAM, JESD209 for Low Power DDR (LPDDR), JESD209-2 for LPDDR2, JESD209-3 for LPDDR3, and JESD209-4 for LPDDR4 (these standards are available at www.jedec.org). Such standards (and similar standards) may be referred to as DDR-based standards and communication interfaces of the storage devices that implement such standards may be referred to as DDR-based interfaces.
In one embodiment, the memory device is a block addressable memory device, such as those based on NAND or NOR technologies. A memory device may also include future generation nonvolatile devices, such as a three dimensional crosspoint memory device (e.g., Intel 3D XPoint™ memory), or other byte addressable write-in-place nonvolatile memory devices. In one embodiment, the memory device may be or may include memory devices that use chalcogenide glass, multi-threshold level NAND flash memory, NOR flash memory, single or multi-level Phase Change Memory (PCM), a resistive memory, nanowire memory, ferroelectric transistor random access memory (FeTRAM), anti-ferroelectric memory, magnetoresistive random access memory (MRAM) memory that incorporates memristor technology, resistive memory including the metal oxide base, the oxygen vacancy base and the conductive bridge Random Access Memory (CB-RAM), or spin transfer torque (STT)-MRAM, a spintronic magnetic junction memory based device, a magnetic tunneling junction (MTJ) based device, a DW (Domain Wall) and SOT (Spin Orbit Transfer) based device, a thyristor based memory device, or a combination of any of the above, or other memory. The memory device may refer to the die itself and/or to a packaged memory product.
In some embodiments, 3D crosspoint memory (e.g., Intel 3D XPoint™ memory) may comprise a transistor-less stackable cross point architecture in which memory cells sit at the intersection of word lines and bit lines and are individually addressable and in which bit storage is based on a change in bulk resistance. In some embodiments, all or a portion of the memory 1710 may be integrated into the processor 1704. In operation, the memory 1710 may store various software and data used during operation, such as a phase lookup table indicative of an area of the memory 1710 that is reserved for storing resource utilization phases. As further described herein, the phase detection logic unit 1705 may evaluate the present resource utilization of the compute sled 1630 relative to the phase lookup table to determine a present phase of a workload execution.
The compute engine 1702 is communicatively coupled to other components of the compute sled 1630 via the I/O subsystem 1712, which may be embodied as circuitry and/or components to facilitate input/output operations with the compute engine 1702 (e.g., with the processor 1704 and/or the memory 1710) and other components of the compute sled 1702. For example, the I/O subsystem 1712 may be embodied as, or otherwise include, memory controller hubs, input/output control hubs, integrated sensor hubs, firmware devices, communication links (e.g., point-to-point links, bus links, wires, cables, light guides, printed circuit board traces, etc.), and/or other components and subsystems to facilitate the input/output operations. In some embodiments, the I/O subsystem 1712 may form a portion of a SoC and be incorporated, along with one or more of the processor 1704, the memory 1710, and other components of the compute sled 1630, into the compute engine 1702.
The communication circuitry 1714 may be embodied as any communication circuit, device, or collection thereof, capable of enabling communications over the network 1612 between the compute sled 1630 and another compute device (e.g., the memory sled 1640, storage sled 1650, the accelerator sled 1660, etc.). The communication circuitry 1714 may be configured to use any one or more communication technology (e.g., wired or wireless communications) and associated protocols (e.g., Ethernet, Bluetooth®, Wi-Fi®, WiMAX, etc.) to effect such communication.
The illustrative communication circuitry 1714 includes a network interface controller (NIC) 1716, which may also be referred to as a host fabric interface (HFI). The NIC 1716 may be embodied as one or more add-in-boards, daughter cards, network interface cards, controller chips, chipsets, or other devices that may be used by the compute sled 1630 to connect with another compute device (e.g., the memory sled 1640, the storage sled 1650, the accelerator sled 1660, etc.). In some embodiments, the NIC 1716 may be embodied as part of an SoC that includes one or more processors, or included on a multichip package that also contains one or more processors. In some embodiments, the NIC 1716 may include a local processor (not shown) and/or a local memory (not shown) that are both local to the NIC 1716. In such embodiments, the local processor of the NIC 1716 may be capable of performing one or more of the functions of the compute engine 1702 described herein. Additionally or alternatively, in such embodiments, the local memory of the NIC 1716 may be integrated into one or more components of the compute sled 1630 at the board level, socket level, chip level, and/or other levels.
The one or more illustrative data storage devices 1718, may be embodied as any type of devices configured for short-term or long-term storage of data such as, for example, memory devices and circuits, memory cards, hard disk drives, solid-state drives, or other data storage devices. Each data storage device 1718 may include a system partition that stores data and firmware code for the data storage device 1718. Each data storage device 1718 may also include an operating system partition that stores data files and executables for an operating system. Additionally or alternatively, the compute sled 1630 may include one or more peripheral devices 1720. Such peripheral devices 1720 may include any type of peripheral device commonly found in a compute device such as a display, speakers, a mouse, a keyboard, and/or other input/output devices, interface devices, and/or other peripheral devices.
Referring now to FIG. 18, the compute sled 1630 may establish an environment 1800 during operation. The illustrative environment 1800 includes a network communicator 1820, a phase detector 1830, and a remediation engine 1840. Each of the components of the environment 1800 may be embodied as hardware, firmware, software, or a combination thereof. As such, in some embodiments, one or more of the components of the environment 1800 may be embodied as circuitry or a collection of electrical devices (e.g., network communicator circuitry 1820, phase detector circuitry 1830, remediation engine circuitry 1840, etc.). It should be appreciated that, in such embodiments, one or more of the network communicator circuitry 1820, phase detector circuitry 1830, or remediation engine circuitry 1840 may form a portion of one or more of the compute engine 1702, the phase detection logic unit 1705, the communication circuitry 1714, the I/O subsystem 1712, and/or other components of the compute sled 1630.
In the illustrative embodiment, the environment 1800 includes telemetry data 1802, which may be embodied as any data indicative of the performance (e.g., cache occupancy, cache hits/misses, core frequency, instructions retired, bytes read from/written to memory controllers, and the like) and other conditions, such as power usage, of the compute sled 1630. Additionally, in the illustrative embodiment, the environment 1800 includes a phase lookup table 1804 which may be embodied as any data indicative of one or more phases of a workload, determined as a function of the telemetry data 1802. More specifically, the phase lookup table 1804 is a structure that may be populated with phase data (e.g., phase data 1806, described herein) at boot time by the BIOS or after boot, such as by the operating system of the compute sled 1630. The phase lookup table 1804 may be indexed based on various characteristics of a workload phase (e.g., by a time interval, by a threshold of a given metric, etc.). The compute sled 1630 may evaluate the phase lookup table 1804 to determine a present phase during execution of a workload. Additionally, in the illustrative embodiment, the environment 1800 includes phase data 1806, which may be embodied as any data indicative of a resource utilization phase (e.g., period of utilization of a particular type of resource above a threshold amount) and lengths of time of those phases (e.g., phase residencies) for a given workload. Patterns of the phase data 1806 may indicate a “fingerprint” of resource utilization for the workload. In particular, the workload fingerprint includes a phase residency matrix indicative of an amount of time that a workload will likely reside in a present phase. The workload fingerprint also includes a phase transition matrix indicative of an amount of time remaining before the workload will transition to a subsequent phase, and, in the illustrative embodiment, indicators of the likelihood of different possible subsequent phases following the present phase (e.g., a 60% chance that phase B will follow phase A and a 40% chance that phase C will follow phase A). In the illustrative embodiment, the policy data 1808 may be any data indicative of predefined preferences and directives for managing resource utilization in the compute sled 1630. The compute sled 1630 may adjust resources therein based on a workload fingerprint and the policy data 1808.
In the illustrative environment 1800, the network communicator 1820, which may be embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof as discussed above, is configured to facilitate inbound and outbound network communications (e.g., network traffic, network packets, network flows, etc.) to and from the compute sled 1630, respectively. To do so, the network communicator 1820 is configured to receive and process data packets from one system or computing device (e.g., the orchestrator server 1620, memory sled 1640, storage sled 1650, accelerator sled 1660, etc.) and to prepare and send data packets to another computing device or system. Accordingly, in some embodiments, at least a portion of the functionality of the network communicator 1820 may be performed by the communication circuitry 1714, and, in the illustrative embodiment, by the NIC 1716.
The phase detector 1830, which may be embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof, is configured to program the processor 1704 to monitor performance events in the compute sled 1630 during execution of one or more workloads, detect workload phases based on the monitored performance events, update a workload fingerprint, and publish the workload fingerprint. To do so, in the illustrative embodiment, the phase detector 1830 includes a monitor component 1832, a detection component 1834, a fingerprinter component 1836, and a publisher component 1838.
In an embodiment, the monitor component 1832 is configured to perform initialization of phase detection functions in the processor. To do so, the monitor component 1832 may write leaf values (e.g., via a WRMSR instruction) to the control MSR 1707. For example, the monitor component 1832 may be configured to reserve areas of memory used by the phase detector 1830 to store the phase lookup table 1804 and workload fingerprint data by writing a leaf value indicative of an instruction to reserve the areas of memory. In addition, the monitor component 1832 may program components in the compute engine 1702, such as the PMU 1709, to monitor performance events occurring in the compute sled 1630 such as by configuring the PMU 1709 to output counter values for a given event (e.g., cache events relating to cache hits/misses per thousand instructions or cache occupancy, instruction execution events affecting cycles per instruction, etc.) to an MSR 1706 or GPR 1708. Further, the monitor component 1832 may initiate counters in the PMU 1709 by issuing a WRMSR instruction to do so. Once initiated, the PMU 1709, in the illustrative embodiment, executes the counters to monitor the programmed events.
In the illustrative embodiment, the detection component 1834 is configured to retrieve counter values output to the registers by the PMU 1709. In particular, the detection component 1834 may pause counters of the PMU 1709 (e.g., by writing a leaf value indicative of an instruction to pause the counters) at a predefined interval and obtain the counter values representative of the telemetry data 1802 from the MSRs 1706 and GPRs 1708. The fingerprinter component 1836, in the illustrative embodiment, is configured to determine a present phase based on the retrieved telemetry data 1802. In particular, the fingerprinter component 1836 evaluates the telemetry data 1802 relative to the phase lookup table 1804 and identifies a matching tuple (or a relatively close match) to a given entry in the phase lookup table 1804. The matched entry may be indicative of the present phase of the workload execution. Further, the fingerprinter component 1836 may update a workload fingerprint as a function of the determined present phase. The publisher component 1838, in the illustrative embodiment, is configured to export the workload fingerprint to the area of memory reserved by the monitor component 1832. The reserved area of memory is accessible by various consumers, such as the remediation engine 1840, user space analytics applications, the orchestrator server 1620, and the like.
Various components of the phase detector 1830 may write leaf values to and/or read leaf values from the control MSR 1707. Table 1, depicted below, summarizes each leaf value and its corresponding interpretation, in the illustrative embodiment.

	TABLE 1

	Leaf Value	Description

	1	Program PMU registers with events
		Program address 1 (e.g., start location of the
		phase lookup table)
		Program address 2 (e.g., location where output
		is stored, where the size of the area depends on
		a total number of hardware threads present in
		the system).
	2	Start counter
		Enable periodic event to log phase data
	4	Stop counter
		Export telemetry data to memory

In the illustrative embodiment, the remediation engine 1840 is configured to evaluate the present workload fingerprint relative to the policy data 1808 and determine whether any remedial actions should be performed within the compute sled 1630. Remedial actions may include reallocating compute resources to a particular workload, assigning a given workload to the compute sled 1630, and the like. For example, the policy data 1808 may indicate that if the cache miss rate is above a certain threshold, then the remediation engine 1840 should reduce the cache size (e.g., by an amount determined as a function of the cache miss rate). As another example, the policy data 1808 may indicate that if the processor capacity exceeds a specified threshold, then the remediation engine 1840 should allocate the excess processor capacity towards a given workload.
Referring now to FIG. 19, the compute sled 1630, in operation, may execute a method 1900 to provide in-processor workload phase detection. As shown, the method 1900 begins at boot of the platform. In block 1902, the BIOS of the compute sled 1630 reserves an area of memory for the phase lookup table. Further, the BIOS reserves an area of memory to store a workload fingerprint output by the processor 1704 in operation. In block 1904, the compute sled 1630 programs the PMU 1709 to monitor telemetry data. In particular, in block 1906, the compute sled 1630 writes, to the control MSR, a leaf value indicative of programming the PMU. In block 1908, the BIOS passes the reserved memory addresses to the processor 1704. The processor 1704 may write the addresses to one of the MSRs 1706. In block 1910, the processor 1704 programs the PMU 1709 on each hardware thread with a predefined set of events to monitor.
In block 1912, the compute sled 1630 initializes the phase lookup table. In particular, in block 1914, the compute sled 1630 executes one or more workloads using a training data set as input, which may be indicative of any metrics data used for a variety of workloads executing in the system 1610. In block 1916, the compute sled 1630 collects telemetry data during the execution of the one or more workloads. In block 1918, the compute sled 1630 generates a resource utilization phase model based on the collected telemetry data. The resource utilization phase model allows the compute sled 1630 to detect a given phase corresponding to a given tuple of training data. For each metric in the telemetry data, the compute sled 1630 identifies the highest possible value or applicable threshold. The compute sled 1630 may use the threshold to discretize the telemetry data metrics into higher granularity values. Discretizing the metrics transforms continuous data into a discrete data point. In block 1920, the compute sled 1630 populates the phase lookup table with the discretized telemetry data.
In block 1922, the compute sled 1630 activates monitoring of telemetry data via the PMU 1709. In doing so, in the illustrative embodiment, the compute sled 1630 writes a leaf value to the control MSR 1707, as indicated in block 1924. The written leaf value is indicative of an instruction for activating event counters in the PMU 1709. In block 1926, the PMU 1709 starts the counters to begin monitoring the telemetry data. The processor 1704 may evaluate the control MSR 1707 (e.g., by performing a RDMSR operation on the control MSR 1707) to obtain the written leaf value and cause the PMU 1709 to start, in response to the written leaf value, the counters.
Referring now to FIG. 20, the PMU 1709 of the processor 1704 may monitor predefined performance events. In the illustrative embodiment, and as in block 1928, the compute sled 1630 executes a workload for the system 1610 (e.g., a workload assigned by the orchestrator server 1620). Event counters in the PMU 1709 may be associated with various resources to monitor, such as cache occupancy, core frequency, cache hit/miss rate per thousand instructions, elapsed core clock ticks, instructions retired, memory controller read/written bytes, data traffic transferred by interconnect links, etc. The compute sled 1630, in the illustrative embodiment, selectively pauses the event counters to retrieve counter values for a given point in time of execution of the workload. In block 1930, the compute sled 1630 determines whether to pause event monitoring in the PMU 1709. For example, the compute sled 1630 may do so at a specified interval or upon receipt of a request to the processor 1704 to do so. If not, then the event counters in the PMU 1709 remain active.
Otherwise, in block 1932, the compute sled 1630 pauses event monitoring by the PMU 1709. For example, to do so, the compute sled 1630 may write a leaf value to the control MSR 1707 indicative of an instruction to pause the event counters, as indicated in block 1934. The processor 1704 reads the control MSR 1707 and causes the PMU 1709 to pause the event counters. In block 1936, the compute sled 1630 retrieves telemetry data from the PMU 1709. In particular, the processor 1704 may retrieve the counter values from the MSRs 1706 and/or GPRs 1708 that store the counter values corresponding to each monitored event.
In block 1938, the compute sled 1630 determines a present phase as a function of the retrieved telemetry data. In particular, the compute sled 1630 may discretize the telemetry data, as indicated in block 1940. For example, the compute sled 1630 may discretize the data in the manner described above. Further, Table 2 below depicts an example discretization of a telemetry data metric for IPC:

TABLE 2

	Mispredicts per thousand	Occupancy - divide current
IPC - Divide current IPC by 2	instructions (MPKI)	occupancy by LLC size

Range	Identifier	Range	Identifier	Range	Identifier

0-0.5	L	0-5	L	0.00-0.33	L
0.5-1.0	M	5-20	M	0.33-0.66	M
>1	H	>20	H	0.66-1.00	H

In Table 2, the IPC range of 0-0.5 is marked ‘L’ (low), the IPC range of 0.15-1.0 is marked ‘M’ (medium), and values higher than 1.0 are marked as ‘H’ (high). The MPKI and cache occupancy values are similarly discretized. In block 1942, the compute sled 1630 queries the phase lookup table 1804 using the discretized telemetry data as input. Table 3 provides an example of a phase lookup table:

TABLE 3

Cache Occupancy	IPC	LLC Miss	Phase Number

L	H	L	0
L	L	H	1
M	H	L	2
L	M	L	3

Table 3 tracks metrics corresponding to cache occupancy, instructions per cycle (IPC), and low-level cache (LLC) misses per thousand instructions. The compute sled 1630 may input the discretized telemetry data as a tuple of metric values, such as (cache occupancy value, IPC value, LLC miss value). In block 1944, the compute sled 1630 retrieves the present phase from the phase lookup table based on the query. Using Table 3 as an example, if the monitored telemetry data, after discretization, corresponds to a low cache occupancy, low IPC, and high LLC miss rate, then the present phase corresponds to phase 1.
In block 1946, the compute sled 1630 updates a workload fingerprint based on the present phase retrieved from the phase lookup table 1804. In doing so, the compute sled 1630, in the illustrative embodiment, recalculates the phase residency and transition matrices based on the present phase, as indicated in block 1948. For example, the compute sled 1630 may extend the phase residency for the present phase if the phase has lasted longer than the time period indicated in the phase residency matrix. Conversely, the compute sled 1630 may shorten the phase residency for the previous phase if the workload transitioned to a subsequent phase earlier than indicated in the phase residency matrix. The compute sled 1630 may also update the probability of the subsequent phase being a particular phase (e.g., phase B) in response to determining that the particular phase (e.g., phase B) occurred after the previous phase. Additionally, the compute sled 1630 outputs the workload fingerprint, as indicated in block 1950. In particular, the compute sled 1630 may write the workload fingerprint to the reserved output location in the memory 1710, as indicated in block 1952. To do so, the compute sled 1630 may determine the memory address reserved by the BIOS as the output location for the workload fingerprint. The compute sled 1630 may then access the memory address and write the workload fingerprint indicative of the present phase at the memory address.
By writing the workload fingerprint to the reserved area of memory, consumers such as user space applications and the orchestrator server 1620 may retrieve the workload fingerprint and analyze the workload fingerprint to perform, in response, some further action. Additionally or alternatively, the remediation engine 1840 may analyze the workload fingerprint and potentially perform, in response to the analysis, a remedial action. Performing the analysis locally (e.g., on the compute sled 1630) offloads this function from the orchestrator server 1620, enabling the orchestrator server 1620 to more efficiently perform other data center wide management operations. Referring now to FIG. 21, in block 1954, the compute sled 1630 (via the remediation engine 1840) retrieves the workload fingerprint from the reserved area of memory.
In block 1956, the compute sled 1630 evaluates the workload fingerprint relative to policy data. The workload fingerprint may satisfy one or more conditions specified in the policy data that cause some remedial action to be performed. In block 1958, the compute sled 1630 determines a remedial action to perform as a function of the policy data. The policy data may specify the remedial action to perform. For example, the policy data may indicate that if cache occupancy is presently low, then the cache size should be reduced. In block 1960, the compute sled 1630 performs the remedial action, if any. Continuing the example, the compute sled 1630 may reduce the cache size for a given workload. Subsequently, the method 1900 loops back to block 1922 of FIG. 19, in which the compute sled 1630 continues monitoring the telemetry data with the PMU 1709.

EXAMPLES

Illustrative examples of the technologies disclosed herein are provided below. An embodiment of the technologies may include any one or more, and any combination of, the examples described below.
Example 1 includes a sled, comprising communication circuitry and a compute engine comprising a performance monitor unit, wherein the compute engine is to (i) obtain telemetry data from the performance monitor unit, wherein the performance monitor unit is to produce telemetry data indicative of resource utilization and workload performance by the sled as one or more workloads are executed, (ii) determine, from a lookup table indicative of a plurality of resource utilization phases, a resource utilization phase based on the obtained telemetry data, (iii) update a workload fingerprint based on the determined resource utilization phase, and (iv) output the workload fingerprint.
Example 2 includes the subject matter of Example 1, and wherein to output the workload fingerprint comprises to output the workload fingerprint to an area of memory in the sled reserved by the performance monitor unit.
Example 3 includes the subject matter of any of Examples 1 and 2, and wherein the compute engine is further to retrieve the workload fingerprint from the area of memory; evaluate the workload fingerprint relative to policy data; determine a remedial action to perform in response to the evaluation; and perform the remedial action.
Example 4 includes the subject matter of any of Examples 1-3, and wherein the compute engine is further to initialize the lookup table with the plurality of resource utilization phases.
Example 5 includes the subject matter of any of Examples 1-4, and wherein to initialize the lookup table with the plurality of resource utilization phases comprises to execute the one or more workloads with training data as input; generate a resource utilization phase model based on telemetry data collected from the execution of the one or more workloads with the training data as input; and populate the lookup table with discretized telemetry data determined based on the resource utilization phase model.
Example 6 includes the subject matter of any of Examples 1-5, and wherein the compute engine is further to program the performance monitor unit to monitor the telemetry data at boot time of the sled.
Example 7 includes the subject matter of any of Examples 1-6, and wherein to program the performance monitor unit comprises to write a value to a register indicative of an instruction to configure, in the performance monitor unit, one or more events to monitor during the execution of the one or more workloads.
Example 8 includes the subject matter of any of Examples 1-7, and wherein to determine the resource utilization phase based on the obtained telemetry data comprises to discretize the telemetry data; query the lookup table using the discretized telemetry data; and receive, in response to the query, the resource utilization phase from the lookup table.
Example 9 includes the subject matter of any of Examples 1-8, and wherein the resource utilization phase is a first resource utilization phase and the workload fingerprint comprises a residency matrix indicative of a time duration that a workload remains in a given resource utilization phase and further comprises a transition probability matrix indicative of a likelihood that the workload transitions to a second resource utilization phase.
Example 10 includes the subject matter of any of Examples 1-9, and wherein to update the workload fingerprint comprises to recalculate the residency matrix and the transition probability matrix based on the determined resource utilization phase.
Example 11 includes the subject matter of any of Examples 1-10, and wherein to obtain the telemetry data comprises to write a value to a register indicative of an instruction to start a counter in the performance monitor unit; and cause, in response to the write of the value to the register, the performance monitor unit to start the counter.
Example 12 includes the subject matter of any of Examples 1-11, and wherein the compute engine is further to write a second value to the register indicative of an instruction to pause the counter in the performance monitor unit.
Example 13 includes the subject matter of any of Examples 1-12, and wherein to obtain the telemetry data comprises to retrieve one or more of a number of cache misses per thousand instructions, a number of cycles per instruction, or cache occupancy data.
Example 14 includes a method, comprising obtaining, by a sled that includes a performance monitor unit, telemetry data produced by the performance monitor unit, wherein the telemetry data is indicative of resource utilization and workload performance by the sled as one or more workloads are executed; determining, by the sled and from a lookup table indicative of a plurality of resource utilization phases, a resource utilization phase based on the obtained telemetry data; updating, by the sled, a workload fingerprint based on the determined resource utilization phase; and outputting, by the sled, the workload fingerprint.
Example 15 includes the subject matter of Example 14, and wherein outputting the workload fingerprint comprises outputting the workload fingerprint to an area of memory in the sled reserved by the performance monitor unit.
Example 16 includes the subject matter of any of Examples 14 and 15, and further including retrieving, by the sled, the workload fingerprint from the area of memory; evaluating, by the sled, the workload fingerprint relative to policy data; determining, by the sled, a remedial action to perform in response to the evaluation; and performing, by the sled, the remedial action.
Example 17 includes the subject matter of any of Examples 14-16, and further including initializing, by the sled, the lookup table with the plurality of resource utilization phases.
Example 18 includes the subject matter of any of Examples 14-17, and wherein initializing the lookup table with the plurality of resource utilization phases comprises executing, by the sled, the one or more workloads with training data as input; generating, by the sled, a resource utilization phase model based on telemetry data collected from the execution of the one or more workloads with the training data as input; and populating, by the sled, the lookup table with discretized telemetry data determined based on the resource utilization phase model.
Example 19 includes the subject matter of any of Examples 14-18, and further including programming, by the sled, the performance monitor unit to monitor the telemetry data at boot time of the sled.
Example 20 includes the subject matter of any of Examples 14-19, and wherein programming the performance monitor unit comprises writing a value to a register indicative of an instruction to configure, in the performance monitor unit, one or more events to monitor during the execution of the one or more workloads.
Example 21 includes the subject matter of any of Examples 14-20, and wherein determining the resource utilization phase based on the obtained telemetry data comprises discretizing the telemetry data; querying the lookup table using the discretized telemetry data; and receiving, in response to the query, the resource utilization phase from the lookup table.
Example 22 includes the subject matter of any of Examples 14-21, and wherein the resource utilization phase is a first resource utilization phase and the workload fingerprint comprises a residency matrix indicative of a time duration that a workload remains in a given resource utilization phase and further comprises a transition probability matrix indicative of a likelihood that the workload transitions to a second resource utilization phase.
Example 23 includes the subject matter of any of Examples 14-22, and wherein updating the workload fingerprint comprises recalculating the residency matrix and the transition probability matrix based on the determined resource utilization phase.
Example 24 includes the subject matter of any of Examples 14-23, and wherein obtaining the telemetry data comprises writing a value to a register indicative of an instruction to start a counter in the performance monitor unit; and causing, in response to the write of the value to the register, the performance monitor unit to start the counter.
Example 25 includes the subject matter of any of Examples 14-24, and further including writing, by the sled, a second value to the register indicative of an instruction to pause the counter in the performance monitor unit.
Example 26 includes the subject matter of any of Examples 14-25, and wherein obtaining the telemetry data comprises retrieving one or more of a number of cache misses per thousand instructions, a number of cycles per instruction, or cache occupancy data.
Example 27 includes one or more machine-readable storage media comprising a plurality of instructions stored thereon that, in response to being executed, cause a sled to perform the method of any of Examples 14-26.
Example 28 includes a sled comprising means for performing the method of any of Examples 14-26.
Example 29 includes a sled comprising a compute engine to perform the method of any of Examples 14-26.
Example 30 includes a sled, comprising a performance monitor unit; and phase detector circuitry to obtain telemetry data from the performance monitor unit, wherein the performance monitor unit is to produce telemetry data indicative of resource utilization and workload performance by the sled as one or more workloads are executed, determine, from a lookup table indicative of a plurality of resource utilization phases, a resource utilization phase based on the obtained telemetry data, update a workload fingerprint based on the determined resource utilization phase, and output the workload fingerprint.
Example 31 includes the subject matter of Example 30, and wherein to output the workload fingerprint comprises to output the workload fingerprint to an area of memory in the sled reserved by the performance monitor unit.
Example 32 includes the subject matter of any of Examples 30 and 31, and wherein the phase detector circuitry is further to retrieve the workload fingerprint from the area of memory; evaluate the workload fingerprint relative to policy data; the sled further comprising remediation engine circuitry to determine a remedial action to perform in response to the evaluation; and perform the remedial action.
Example 33 includes the subject matter of any of Examples 30-32, and wherein the phase detector circuitry is further to initialize the lookup table with the plurality of resource utilization phases.
Example 34 includes the subject matter of any of Examples 30-33, and wherein to initialize the lookup table with the plurality of resource utilization phases comprises to execute the one or more workloads with training data as input; generate a resource utilization phase model based on telemetry data collected from the execution of the one or more workloads with the training data as input; and populate the lookup table with discretized telemetry data determined based on the resource utilization phase model.
Example 35 includes the subject matter of any of Examples 30-34, and wherein the phase detector circuitry is further to program the performance monitor unit to monitor the telemetry data at boot time of the sled.
Example 36 includes the subject matter of any of Examples 30-35, and wherein to program the performance monitor unit comprises to write a value to a register indicative of an instruction to configure, in the performance monitor unit, one or more events to monitor during the execution of the one or more workloads.
Example 37 includes the subject matter of any of Examples 30-36, and wherein to determine the resource utilization phase based on the obtained telemetry data comprises to discretize the telemetry data; query the lookup table using the discretized telemetry data; and receive, in response to the query, the resource utilization phase from the lookup table.
Example 38 includes the subject matter of any of Examples 30-37, and wherein the resource utilization phase is a first resource utilization phase and the workload fingerprint comprises a residency matrix indicative of a time duration that a workload remains in a given resource utilization phase and further comprises a transition probability matrix indicative of a likelihood that the workload transitions to a second resource utilization phase.
Example 39 includes the subject matter of any of Examples 30-38, and wherein to update the workload fingerprint comprises to recalculate the residency matrix and the transition probability matrix based on the determined resource utilization phase.
Example 40 includes the subject matter of any of Examples 30-39, and wherein to obtain the telemetry data comprises to write a value to a register indicative of an instruction to start a counter in the performance monitor unit; and cause, in response to the write of the value to the register, the performance monitor unit to start the counter.
Example 41 includes the subject matter of any of Examples 30-40, and wherein the phase detector circuitry is further to write a second value to the register indicative of an instruction to pause the counter in the performance monitor unit.
Example 42 includes the subject matter of any of Examples 30-41, and wherein to obtain the telemetry data comprises to retrieve one or more of a number of cache misses per thousand instructions, a number of cycles per instruction, or cache occupancy data.
Example 43 includes a sled, comprising circuitry for obtaining telemetry data from a performance monitor unit in the sled, wherein the performance monitor unit is to produce telemetry data indicative of resource utilization and workload performance by the sled as one or more workloads are executed, means for determining, from a lookup table indicative of a plurality of resource utilization phases, a resource utilization phase based on the obtained telemetry data, means for updating a workload fingerprint based on the determined resource utilization phase, and circuitry for outputting the workload fingerprint.
Example 44 includes the subject matter of Example 43, and wherein the circuitry for outputting the workload fingerprint comprises circuitry for outputting the workload fingerprint to an area of memory in the sled reserved by the performance monitor unit.
Example 45 includes the subject matter of any of Examples 43 and 44, and further including circuitry for retrieving the workload fingerprint from the area of memory; means for evaluating the workload fingerprint relative to policy data; means for determining a remedial action to perform in response to the evaluation; and means for performing the remedial action.
Example 46 includes the subject matter of any of Examples 43-45, and further including circuitry for initializing the lookup table with the plurality of resource utilization phases.
Example 47 includes the subject matter of any of Examples 43-46, and wherein the circuitry for initializing the lookup table with the plurality of resource utilization phases comprises circuitry for executing the one or more workloads with training data as input; circuitry for generating a resource utilization phase model based on telemetry data collected from the execution of the one or more workloads with the training data as input; and circuitry for populating the lookup table with discretized telemetry data determined based on the resource utilization phase model.
Example 48 includes the subject matter of any of Examples 43-47, and further including means for programming the performance monitor unit to monitor the telemetry data at boot time of the sled.
Example 49 includes the subject matter of any of Examples 43-48, and wherein the means for programming the performance monitor unit comprises circuitry for writing a value to a register indicative of an instruction to configure, in the performance monitor unit, one or more events to monitor during the execution of the one or more workloads.
Example 50 includes the subject matter of any of Examples 43-49, and wherein the means for determining the resource utilization phase based on the obtained telemetry data comprises circuitry for discretizing the telemetry data; circuitry for querying the lookup table using the discretized telemetry data; and circuitry for receiving, in response to the query, the resource utilization phase from the lookup table.
Example 51 includes the subject matter of any of Examples 43-50, and wherein the resource utilization phase is a first resource utilization phase and the workload fingerprint comprises a residency matrix indicative of a time duration that a workload remains in a given resource utilization phase further comprises a transition probability matrix indicative of a likelihood that the workload transitions to a second resource utilization phase.
Example 52 includes the subject matter of any of Examples 43-51, and wherein the means for updating the workload fingerprint comprises circuitry for recalculating the residency matrix and the transition probability matrix based on the determined resource utilization phase.
Example 53 includes the subject matter of any of Examples 43-52, and wherein the circuitry for obtaining the telemetry data comprises circuitry for writing a value to a register indicative of an instruction to start a counter in the performance monitor unit; and circuitry for causing, in response to the write of the value to the register, the performance monitor unit to start the counter.
Example 54 includes the subject matter of any of Examples 43-53, and wherein the phase detector circuitry is further to write a second value to the register indicative of an instruction to pause the counter in the performance monitor unit.
Example 55 includes the subject matter of any of Examples 43-54, and wherein to obtain the telemetry data comprises to retrieve one or more of a number of cache misses per thousand instructions, a number of cycles per instruction, or cache occupancy data.

Claims

1. A sled, comprising:

communication circuitry;

a compute engine comprising a performance monitor unit, wherein the compute engine is to (i) obtain telemetry data from the performance monitor unit, wherein the performance monitor unit is to produce telemetry data indicative of resource utilization and workload performance by the sled as one or more workloads are executed, (ii) determine, from a lookup table indicative of a plurality of resource utilization phases, a resource utilization phase based on the obtained telemetry data, (iii) update a workload fingerprint based on the determined resource utilization phase, and (iv) output the workload fingerprint.

2. The sled of claim 1, wherein to output the workload fingerprint comprises to output the workload fingerprint to an area of memory in the sled reserved by the performance monitor unit.

3. The sled of claim 2, wherein the compute engine is further to:

retrieve the workload fingerprint from the area of memory;

evaluate the workload fingerprint relative to policy data;

determine a remedial action to perform in response to the evaluation; and

perform the remedial action.

4. The sled of claim 1, wherein the compute engine is further to initialize the lookup table with the plurality of resource utilization phases.

5. The sled of claim 4, wherein to initialize the lookup table with the plurality of resource utilization phases comprises to:

execute the one or more workloads with training data as input;

generate a resource utilization phase model based on telemetry data collected from the execution of the one or more workloads with the training data as input; and

populate the lookup table with discretized telemetry data determined based on the resource utilization phase model.

6. The sled of claim 1, wherein the compute engine is further to program the performance monitor unit to monitor the telemetry data at boot time of the sled.

7. The sled of claim 6, wherein to program the performance monitor unit comprises to write a value to a register indicative of an instruction to configure, in the performance monitor unit, one or more events to monitor during the execution of the one or more workloads.

8. The sled of claim 1, wherein to determine the resource utilization phase based on the obtained telemetry data comprises to:

discretize the telemetry data;

query the lookup table using the discretized telemetry data; and

receive, in response to the query, the resource utilization phase from the lookup table.

9. The sled of claim 1, wherein the resource utilization phase is a first resource utilization phase and the workload fingerprint comprises a residency matrix indicative of a time duration that a workload remains in a given resource utilization phase and further comprises a transition probability matrix indicative of a likelihood that the workload transitions to a second resource utilization phase.

10. The sled of claim 9, wherein to update the workload fingerprint comprises to recalculate the residency matrix and the transition probability matrix based on the determined resource utilization phase.

11. The sled of claim 1, wherein to obtain the telemetry data comprises to:

write a value to a register indicative of an instruction to start a counter in the performance monitor unit; and

cause, in response to the write of the value to the register, the performance monitor unit to start the counter.

12. The sled of claim 11, wherein the compute engine is further to write a second value to the register indicative of an instruction to pause the counter in the performance monitor unit.

13. The sled of claim 1, wherein to obtain the telemetry data comprises to retrieve one or more of a number of cache misses per thousand instructions, a number of cycles per instruction, or cache occupancy data.

14. One or more machine-readable storage media comprising a plurality of instructions stored thereon that, in response to being executed, cause a sled to:

obtain telemetry data produced by a performance monitor unit of the sled, wherein the telemetry data is indicative of resource utilization and workload performance by the sled as one or more workloads are executed;

determine from a lookup table indicative of a plurality of resource utilization phases, a resource utilization phase based on the obtained telemetry data;

update a workload fingerprint based on the determined resource utilization phase; and

output the workload fingerprint.

15. The one or more machine-readable storage media of claim 14, wherein to output the workload fingerprint comprises to output the workload fingerprint to an area of memory in the sled reserved by the performance monitor unit.

16. The one or more machine-readable storage media of claim 15, wherein the plurality of instructions further cause the sled to:

retrieve the workload fingerprint from the area of memory;

evaluate the workload fingerprint relative to policy data;

determine a remedial action to perform in response to the evaluation; and

perform the remedial action.

17. The one or more machine-readable storage media of claim 14, wherein the plurality of instructions further cause the sled to initialize the lookup table with the plurality of resource utilization phases.

18. The one or more machine-readable storage media of claim 17, wherein to initialize the lookup table with the plurality of resource utilization phases comprises to:

execute the one or more workloads with training data as input;

19. The one or more machine-readable storage media of claim 14, wherein the plurality of instructions further cause the sled to program the performance monitor unit to monitor the telemetry data at boot time of the sled.

20. The one or more machine-readable storage media of claim 19, wherein to program the performance monitor unit comprises to write a value to a register indicative of an instruction to configure, in the performance monitor unit, one or more events to monitor during the execution of the one or more workloads.

21. The one or more machine-readable storage media of claim 14, wherein to determine the resource utilization phase based on the obtained telemetry data comprises to:

discretize the telemetry data;

query the lookup table using the discretized telemetry data; and

22. The one or more machine-readable storage media of claim 14, wherein the resource utilization phase is a first resource utilization phase and the workload fingerprint comprises a residency matrix indicative of a time duration that a workload remains in a given resource utilization phase and further comprises a transition probability matrix indicative of a likelihood that the workload transitions to a second resource utilization phase.

23. The one or more machine-readable storage media of claim 22, wherein to update the workload fingerprint comprises recalculating the residency matrix and the transition probability matrix based on the determined resource utilization phase.

24. A sled, comprising:

circuitry for obtaining telemetry data from a performance monitor unit of the sled, wherein the performance monitor unit is to produce telemetry data indicative of resource utilization and workload performance by the sled as one or more workloads are executed;

means for determining, from a lookup table indicative of a plurality of resource utilization phases, a resource utilization phase based on the obtained telemetry data;

means for updating a workload fingerprint based on the determined resource utilization phase; and

circuitry for outputting the workload fingerprint.

25. A method, comprising:

obtaining, by a sled that includes a performance monitor unit, telemetry data produced by the performance monitor unit, wherein the telemetry data is indicative of resource utilization and workload performance by the sled as one or more workloads are executed;

determining, by the sled and from a lookup table indicative of a plurality of resource utilization phases, a resource utilization phase based on the obtained telemetry data;

updating, by the sled, a workload fingerprint based on the determined resource utilization phase; and

outputting, by the sled, the workload fingerprint.

26. The method of claim 25, wherein outputting the workload fingerprint comprises outputting the workload fingerprint to an area of memory in the sled reserved by the performance monitor unit.

27. The method of claim 26, further comprising:

retrieving, by the sled, the workload fingerprint from the area of memory;

evaluating, by the sled, the workload fingerprint relative to policy data;

determining, by the sled, a remedial action to perform in response to the evaluation; and

performing, by the sled, the remedial action.

28. The method of claim 25, further comprising initializing, by the sled, the lookup table with the plurality of resource utilization phases.