US20190042126A1

US20190042126A1 - Technologies for storage discovery and reallocation

Info

Publication number: US20190042126A1
Application number: US15/858,569
Authority: US
Inventors: Sujoy Sen; Gamil Cain; Teddy Greer; Anjaneya Reddy Chagam Reddy
Original assignee: Intel Corp
Current assignee: Intel Corp
Priority date: 2017-08-30
Filing date: 2017-12-29
Publication date: 2019-02-07
Also published as: US20190034102A1; CN109426455A; US20190034383A1; EP3451635A1; US20190042090A1; US11050554B2; US10581596B2; US20230050698A1; EP3451196A1; US20190042611A1; US11249816B2; CN109426569A; US20190042408A1; CN113806260A; US20190042136A1; CN115543204A; US10476670B2; CN109428770A; US11888967B2; US20190069434A1

Abstract

Technologies for storage discovery and reallocation include a compute device. The compute device is to receive, from a data storage sled, storage device data from a storage device located on the data storage sled. The storage device data includes storage device self-test data that defines a result of a self-test performed by the storage device. The compute device is also to determine, in response to the storage device self-test data, whether the storage device fails to satisfy a performance threshold. Further, the compute device is to generate, in response to a determination that the storage device fails to satisfy the performance threshold, an adjustment message for the storage device. The adjustment message instructs the storage device to adjust a performance parameter of the storage device. The compute device is also to send the adjustment message to the storage device.

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 large computer networks, storage devices perform important functions such as storing data for use by applications that may operate on other compute devices. Data for one application may be distributed across multiple storage devices. High performance, reliability, and security are essential attributes for these storage devices. Performance capabilities such as size, latency, and throughput are generally specified by the manufacturer of a storage device or unit (e.g., a hard disk drive or solid-state drive). For example, the lower the usable storage capacity for a given amount of media, the better its write performance can be.
However, these capabilities are often recorded on a document (e.g., a performance specification sheet) that is provided with the storage device and may be difficult to access or query. Making useful decisions regarding a large number of storage devices may present obstacles because retrieving performance specification data for each storage device can be a cumbersome and labor-intensive process. Additionally, it is time-consuming to assess a storage device's performance on an ongoing basis. Some known methods involve querying or externally testing a storage device to determine whether it is meeting performance requirements. This, too, becomes untenable at the scale at which many modern computer networks employ storage devices, because performing testing and later collating results can be a tedious and expensive process. Additionally, even after testing, it may be unfeasible to reconfigure a storage device for continued performance, resulting in the storage device being decommissioned while it still has usable life.

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 diagram of a conceptual overview of a data center in which one or more techniques described herein may be implemented according to various embodiments;

FIG. 2 is a diagram of an example embodiment of a logical configuration of a rack of the data center of FIG. 1;

FIG. 3 is a diagram of an example embodiment of another data center in which one or more techniques described herein may be implemented according to various embodiments;

FIG. 4 is a diagram of another example embodiment of a data center in which one or more techniques described herein may be implemented according to various embodiments;

FIG. 5 is a diagram of a connectivity scheme representative of link-layer connectivity that may be established among various sleds of the data centers of FIGS. 1, 3, and 4;

FIG. 6 is a diagram of a rack architecture that may be representative of an architecture of any particular one of the racks depicted in FIGS. 1-4 according to some embodiments;

FIG. 7 is a diagram of an example embodiment of a sled that may be used with the rack architecture of FIG. 6;

FIG. 8 is a diagram of an example embodiment of a rack architecture to provide support for sleds featuring expansion capabilities;

FIG. 9 is a diagram of an example embodiment of a rack implemented according to the rack architecture of FIG. 8;

FIG. 10 is a diagram of an example embodiment of a sled designed for use in conjunction with the rack of FIG. 9;

FIG. 11 is a diagram of an example embodiment of a data center in which one or more techniques described herein may be implemented according to various embodiments;

FIG. 12 is a simplified block diagram of at least one embodiment of a system for storage discovery and automatic repurposing of storage devices;

FIG. 13 is a simplified block diagram of at least one embodiment of a storage device of the system of FIG. 12;

FIG. 14 is a simplified block diagram of at least one embodiment of an environment that may be established by the storage device of FIGS. 12 and 13;

FIG. 15 is a simplified block diagram of at least one embodiment of an environment that may be established by the orchestrator server of FIG. 12;

FIGS. 16 and 17 are simplified flow diagrams of at least one embodiment of a method for providing storage discovery and automatic repurposing of storage devices that may be performed by the storage device of FIGS. 12-14; and

FIGS. 18 and 19 are simplified flow diagrams of at least one embodiment of a method for providing storage discovery and automatic repurposing of storage devices that may be performed by the orchestrator server of FIGS. 12-14.

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.
FIG. 1 illustrates a conceptual overview of a data center 100 that may generally be representative of a data center or other type of computing network in/for which one or more techniques described herein may be implemented according to various embodiments. As shown in FIG. 1, data center 100 may generally contain a plurality of racks, each of which may house computing equipment comprising a respective set of physical resources. In the particular non-limiting example depicted in FIG. 1, data center 100 contains four racks 102A to 102D, which house computing equipment comprising respective sets of physical resources (PCRs) 105A to 105D. According to this example, a collective set of physical resources 106 of data center 100 includes the various sets of physical resources 105A to 105D that are distributed among racks 102A to 102D. Physical resources 106 may include resources of multiple types, such as—for example—processors, co-processors, accelerators, field programmable gate arrays (FPGAs), memory, and storage. The embodiments are not limited to these examples.
The illustrative data center 100 differs from typical data centers in many ways. For example, in the illustrative embodiment, the circuit boards (“sleds”) on which components such as CPUs, memory, and other components are placed for increased thermal performance In particular, in the illustrative embodiment, the sleds are shallower than typical boards. In other words, the sleds are shorter from the front to the back, where cooling fans are located. This decreases the length of the path that air must to travel across the components on the board. Further, the components on the sled are spaced further apart than in typical circuit boards, and the components are arranged to reduce or eliminate shadowing (i.e., one component in the air flow path of another component). In the illustrative embodiment, processing components such as the processors are located on a top side of a sled while near memory, such as DIMMs, are located on a bottom side of the sled. As a result of the enhanced airflow provided by this design, the components may operate at higher frequencies and power levels than in typical systems, thereby increasing performance Furthermore, the sleds are configured to blindly mate with power and data communication cables in each rack 102A, 102B, 102C, 102D, enhancing their ability to be quickly removed, upgraded, reinstalled, and/or replaced. Similarly, individual components located on the sleds, such as processors, accelerators, memory, and data storage drives, are configured to be easily upgraded due to their increased spacing from each other. In the illustrative embodiment, the components additionally include hardware attestation features to prove their authenticity.
Furthermore, in the illustrative embodiment, the data center 100 utilizes a single network architecture (“fabric”) that supports multiple other network architectures including Ethernet and Omni-Path. The sleds, in the illustrative embodiment, are coupled to switches via optical fibers, which provide higher bandwidth and lower latency than typical twisted pair cabling (e.g., Category 5, Category 5e, Category 6, etc.). Due to the high bandwidth, low latency interconnections and network architecture, the data center 100 may, in use, pool resources, such as memory, accelerators (e.g., graphics accelerators, FPGAs, ASICs, etc.), and data storage drives that are physically disaggregated, and provide them to compute resources (e.g., processors) on an as needed basis, enabling the compute resources to access the pooled resources as if they were local. The illustrative data center 100 additionally receives utilization information for the various resources, predicts resource utilization for different types of workloads based on past resource utilization, and dynamically reallocates the resources based on this information.
The racks 102A, 102B, 102C, 102D of the data center 100 may include physical design features that facilitate the automation of a variety of types of maintenance tasks. For example, data center 100 may be implemented using racks that are designed to be robotically-accessed, and to accept and house robotically-manipulatable resource sleds. Furthermore, in the illustrative embodiment, the racks 102A, 102B, 102C, 102D include integrated power sources that receive a greater voltage than is typical for power sources. The increased voltage enables the power sources to provide additional power to the components on each sled, enabling the components to operate at higher than typical frequencies.
FIG. 2 illustrates an exemplary logical configuration of a rack 202 of the data center 100. As shown in FIG. 2, rack 202 may generally house a plurality of sleds, each of which may comprise a respective set of physical resources. In the particular non-limiting example depicted in FIG. 2, rack 202 houses sleds 204-1 to 204-4 comprising respective sets of physical resources 205-1 to 205-4, each of which constitutes a portion of the collective set of physical resources 206 comprised in rack 202. With respect to FIG. 1, if rack 202 is representative of—for example—rack 102A, then physical resources 206 may correspond to the physical resources 105A comprised in rack 102A. In the context of this example, physical resources 105A may thus be made up of the respective sets of physical resources, including physical storage resources 205-1, physical accelerator resources 205-2, physical memory resources 205-3, and physical compute resources 205-4 comprised in the sleds 204-1 to 204-4 of rack 202. The embodiments are not limited to this example. Each sled may contain a pool of each of the various types of physical resources (e.g., compute, memory, accelerator, storage). By having robotically accessible and robotically manipulatable sleds comprising disaggregated resources, each type of resource can be upgraded independently of each other and at their own optimized refresh rate.
FIG. 3 illustrates an example of a data center 300 that may generally be representative of one in/for which one or more techniques described herein may be implemented according to various embodiments. In the particular non-limiting example depicted in FIG. 3, data center 300 comprises racks 302-1 to 302-32. In various embodiments, the racks of data center 300 may be arranged in such fashion as to define and/or accommodate various access pathways. For example, as shown in FIG. 3, the racks of data center 300 may be arranged in such fashion as to define and/or accommodate access pathways 311A, 311B, 311C, and 311D. In some embodiments, the presence of such access pathways may generally enable automated maintenance equipment, such as robotic maintenance equipment, to physically access the computing equipment housed in the various racks of data center 300 and perform automated maintenance tasks (e.g., replace a failed sled, upgrade a sled). In various embodiments, the dimensions of access pathways 311A, 311B, 311C, and 311D, the dimensions of racks 302-1 to 302-32, and/or one or more other aspects of the physical layout of data center 300 may be selected to facilitate such automated operations. The embodiments are not limited in this context.
FIG. 4 illustrates an example of a data center 400 that may generally be representative of one in/for which one or more techniques described herein may be implemented according to various embodiments. As shown in FIG. 4, data center 400 may feature an optical fabric 412. Optical fabric 412 may generally comprise a combination of optical signaling media (such as optical cabling) and optical switching infrastructure via which any particular sled in data center 400 can send signals to (and receive signals from) each of the other sleds in data center 400. The signaling connectivity that optical fabric 412 provides to any given sled may include connectivity both to other sleds in a same rack and sleds in other racks. In the particular non-limiting example depicted in FIG. 4, data center 400 includes four racks 402A to 402D. Racks 402A to 402D house respective pairs of sleds 404A-1 and 404A-2, 404B-1 and 404B-2, 404C-1 and 404C-2, and 404D-1 and 404D-2. Thus, in this example, data center 400 comprises a total of eight sleds. Via optical fabric 412, each such sled may possess signaling connectivity with each of the seven other sleds in data center 400. For example, via optical fabric 412, sled 404A-1 in rack 402A may possess signaling connectivity with sled 404A-2 in rack 402A, as well as the six other sleds 404B-1, 404B-2, 404C-1, 404C-2, 404D-1, and 404D-2 that are distributed among the other racks 402B, 402C, and 402D of data center 400. The embodiments are not limited to this example.
FIG. 5 illustrates an overview of a connectivity scheme 500 that may generally be representative of link-layer connectivity that may be established in some embodiments among the various sleds of a data center, such as any of example data centers 100, 300, and 400 of FIGS. 1, 3, and 4. Connectivity scheme 500 may be implemented using an optical fabric that features a dual-mode optical switching infrastructure 514. Dual-mode optical switching infrastructure 514 may generally comprise a switching infrastructure that is capable of receiving communications according to multiple link-layer protocols via a same unified set of optical signaling media, and properly switching such communications. In various embodiments, dual-mode optical switching infrastructure 514 may be implemented using one or more dual-mode optical switches 515. In various embodiments, dual-mode optical switches 515 may generally comprise high-radix switches. In some embodiments, dual-mode optical switches 515 may comprise multi-ply switches, such as four-ply switches. In various embodiments, dual-mode optical switches 515 may feature integrated silicon photonics that enable them to switch communications with significantly reduced latency in comparison to conventional switching devices. In some embodiments, dual-mode optical switches 515 may constitute leaf switches 530 in a leaf-spine architecture additionally including one or more dual-mode optical spine switches 520.
In various embodiments, dual-mode optical switches may be capable of receiving both Ethernet protocol communications carrying Internet Protocol (IP packets) and communications according to a second, high-performance computing (HPC) link-layer protocol (e.g., Intel's Omni-Path Architecture's, InfiniBand™) via optical signaling media of an optical fabric. As reflected in FIG. 5, with respect to any particular pair of sleds 504A and 504B possessing optical signaling connectivity to the optical fabric, connectivity scheme 500 may thus provide support for link-layer connectivity via both Ethernet links and HPC links. Thus, both Ethernet and HPC communications can be supported by a single high-bandwidth, low-latency switch fabric. The embodiments are not limited to this example.
FIG. 6 illustrates a general overview of a rack architecture 600 that may be representative of an architecture of any particular one of the racks depicted in FIGS. 1 to 4 according to some embodiments. As reflected in FIG. 6, rack architecture 600 may generally feature a plurality of sled spaces into which sleds may be inserted, each of which may be robotically-accessible via a rack access region 601. In the particular non-limiting example depicted in FIG. 6, rack architecture 600 features five sled spaces 603-1 to 603-5. Sled spaces 603-1 to 603-5 feature respective multi-purpose connector modules (MPCMs) 616-1 to 616-5.
FIG. 7 illustrates an example of a sled 704 that may be representative of a sled of such a type. As shown in FIG. 7, sled 704 may comprise a set of physical resources 705, as well as an MPCM 716 designed to couple with a counterpart MPCM when sled 704 is inserted into a sled space such as any of sled spaces 603-1 to 603-5 of FIG. 6. Sled 704 may also feature an expansion connector 717. Expansion connector 717 may generally comprise a socket, slot, or other type of connection element that is capable of accepting one or more types of expansion modules, such as an expansion sled 718. By coupling with a counterpart connector on expansion sled 718, expansion connector 717 may provide physical resources 705 with access to supplemental computing resources 705B residing on expansion sled 718. The embodiments are not limited in this context.
FIG. 8 illustrates an example of a rack architecture 800 that may be representative of a rack architecture that may be implemented in order to provide support for sleds featuring expansion capabilities, such as sled 704 of FIG. 7. In the particular non-limiting example depicted in FIG. 8, rack architecture 800 includes seven sled spaces 803-1 to 803-7, which feature respective MPCMs 816-1 to 816-7. Sled spaces 803-1 to 803-7 include respective primary regions 803-1A to 803-7A and respective expansion regions 803-1B to 803-7B. With respect to each such sled space, when the corresponding MPCM is coupled with a counterpart MPCM of an inserted sled, the primary region may generally constitute a region of the sled space that physically accommodates the inserted sled. The expansion region may generally constitute a region of the sled space that can physically accommodate an expansion module, such as expansion sled 718 of FIG. 7, in the event that the inserted sled is configured with such a module.
FIG. 9 illustrates an example of a rack 902 that may be representative of a rack implemented according to rack architecture 800 of FIG. 8 according to some embodiments. In the particular non-limiting example depicted in FIG. 9, rack 902 features seven sled spaces 903-1 to 903-7, which include respective primary regions 903-1A to 903-7A and respective expansion regions 903-1B to 903-7B. In various embodiments, temperature control in rack 902 may be implemented using an air cooling system. For example, as reflected in FIG. 9, rack 902 may feature a plurality of fans 921 that are generally arranged to provide air cooling within the various sled spaces 903-1 to 903-7. In some embodiments, the height of the sled space is greater than the conventional “1U” server height. In such embodiments, fans 921 may generally comprise relatively slow, large diameter cooling fans as compared to fans used in conventional rack configurations. Running larger diameter cooling fans at lower speeds may increase fan lifetime relative to smaller diameter cooling fans running at higher speeds while still providing the same amount of cooling. The sleds are physically shallower than conventional rack dimensions. Further, components are arranged on each sled to reduce thermal shadowing (i.e., not arranged serially in the direction of air flow). As a result, the wider, shallower sleds allow for an increase in device performance because the devices can be operated at a higher thermal envelope (e.g., 250W) due to improved cooling (i.e., no thermal shadowing, more space between devices, more room for larger heat sinks, etc.).
MPCMs 916-1 to 916-7 may be configured to provide inserted sleds with access to power sourced by respective power modules 920-1 to 920-7, each of which may draw power from an external power source 919. In various embodiments, external power source 919 may deliver alternating current (AC) power to rack 902, and power modules 920-1 to 920-7 may be configured to convert such AC power to direct current (DC) power to be sourced to inserted sleds. In some embodiments, for example, power modules 920-1 to 920-7 may be configured to convert 277-volt AC power into 12-volt DC power for provision to inserted sleds via respective MPCMs 916-1 to 916-7. The embodiments are not limited to this example.
MPCMs 916-1 to 916-7 may also be arranged to provide inserted sleds with optical signaling connectivity to a dual-mode optical switching infrastructure 914, which may be the same as—or similar to—dual-mode optical switching infrastructure 514 of FIG. 5. In various embodiments, optical connectors contained in MPCMs 916-1 to 916-7 may be designed to couple with counterpart optical connectors contained in MPCMs of inserted sleds to provide such sleds with optical signaling connectivity to dual-mode optical switching infrastructure 914 via respective lengths of optical cabling 922-1 to 922-7. In some embodiments, each such length of optical cabling may extend from its corresponding MPCM to an optical interconnect loom 923 that is external to the sled spaces of rack 902. In various embodiments, optical interconnect loom 923 may be arranged to pass through a support post or other type of load-bearing element of rack 902. The embodiments are not limited in this context. Because inserted sleds connect to an optical switching infrastructure via MPCMs, the resources typically spent in manually configuring the rack cabling to accommodate a newly inserted sled can be saved.
FIG. 10 illustrates an example of a sled 1004 that may be representative of a sled designed for use in conjunction with rack 902 of FIG. 9 according to some embodiments. Sled 1004 may feature an MPCM 1016 that comprises an optical connector 1016A and a power connector 1016B, and that is designed to couple with a counterpart MPCM of a sled space in conjunction with insertion of MPCM 1016 into that sled space. Coupling MPCM 1016 with such a counterpart MPCM may cause power connector 1016 to couple with a power connector comprised in the counterpart MPCM. This may generally enable physical resources 1005 of sled 1004 to source power from an external source, via power connector 1016 and power transmission media 1024 that conductively couples power connector 1016 to physical resources 1005.
Sled 1004 may also include dual-mode optical network interface circuitry 1026. Dual-mode optical network interface circuitry 1026 may generally comprise circuitry that is capable of communicating over optical signaling media according to each of multiple link-layer protocols supported by dual-mode optical switching infrastructure 914 of FIG. 9. In some embodiments, dual-mode optical network interface circuitry 1026 may be capable both of Ethernet protocol communications and of communications according to a second, high-performance protocol. In various embodiments, dual-mode optical network interface circuitry 1026 may include one or more optical transceiver modules 1027, each of which may be capable of transmitting and receiving optical signals over each of one or more optical channels. The embodiments are not limited in this context.
Coupling MPCM 1016 with a counterpart MPCM of a sled space in a given rack may cause optical connector 1016A to couple with an optical connector comprised in the counterpart MPCM. This may generally establish optical connectivity between optical cabling of the sled and dual-mode optical network interface circuitry 1026, via each of a set of optical channels 1025. Dual-mode optical network interface circuitry 1026 may communicate with the physical resources 1005 of sled 1004 via electrical signaling media 1028. In addition to the dimensions of the sleds and arrangement of components on the sleds to provide improved cooling and enable operation at a relatively higher thermal envelope (e.g., 250W), as described above with reference to FIG. 9, in some embodiments, a sled may include one or more additional features to facilitate air cooling, such as a heatpipe and/or heat sinks arranged to dissipate heat generated by physical resources 1005. It is worthy of note that although the example sled 1004 depicted in FIG. 10 does not feature an expansion connector, any given sled that features the design elements of sled 1004 may also feature an expansion connector according to some embodiments. The embodiments are not limited in this context.
FIG. 11 illustrates an example of a data center 1100 that may generally be representative of one in/for which one or more techniques described herein may be implemented according to various embodiments. As reflected in FIG. 11, a physical infrastructure management framework 1150A may be implemented to facilitate management of a physical infrastructure 1100A of data center 1100. In various embodiments, one function of physical infrastructure management framework 1150A may be to manage automated maintenance functions within data center 1100, such as the use of robotic maintenance equipment to service computing equipment within physical infrastructure 1100A. In some embodiments, physical infrastructure 1100A may feature an advanced telemetry system that performs telemetry reporting that is sufficiently robust to support remote automated management of physical infrastructure 1100A. In various embodiments, telemetry information provided by such an advanced telemetry system may support features such as failure prediction/prevention capabilities and capacity planning capabilities. In some embodiments, physical infrastructure management framework 1150A may also be configured to manage authentication of physical infrastructure components using hardware attestation techniques. For example, robots may verify the authenticity of components before installation by analyzing information collected from a radio frequency identification (RFID) tag associated with each component to be installed. The embodiments are not limited in this context.
As shown in FIG. 11, the physical infrastructure 1100A of data center 1100 may comprise an optical fabric 1112, which may include a dual-mode optical switching infrastructure 1114. Optical fabric 1112 and dual-mode optical switching infrastructure 1114 may be the same as—or similar to—optical fabric 412 of FIG. 4 and dual-mode optical switching infrastructure 514 of FIG. 5, respectively, and may provide high-bandwidth, low-latency, multi-protocol connectivity among sleds of data center 1100. As discussed above, with reference to FIG. 1, in various embodiments, the availability of such connectivity may make it feasible to disaggregate and dynamically pool resources such as accelerators, memory, and storage. In some embodiments, for example, one or more pooled accelerator sleds 1130 may be included among the physical infrastructure 1100A of data center 1100, each of which may comprise a pool of accelerator resources—such as co-processors and/or FPGAs, for example—that is globally accessible to other sleds via optical fabric 1112 and dual-mode optical switching infrastructure 1114.
In another example, in various embodiments, one or more pooled storage sleds 1132 may be included among the physical infrastructure 1100A of data center 1100, each of which may comprise a pool of storage resources that is globally accessible to other sleds via optical fabric 1112 and dual-mode optical switching infrastructure 1114. In some embodiments, such pooled storage sleds 1132 may comprise pools of solid-state storage devices such as solid-state drives (SSDs). In various embodiments, one or more high-performance processing sleds 1134 may be included among the physical infrastructure 1100A of data center 1100. In some embodiments, high-performance processing sleds 1134 may comprise pools of high-performance processors, as well as cooling features that enhance air cooling to yield a higher thermal envelope of up to 250W or more. In various embodiments, any given high-performance processing sled 1134 may feature an expansion connector 1117 that can accept a far memory expansion sled, such that the far memory that is locally available to that high-performance processing sled 1134 is disaggregated from the processors and near memory comprised on that sled. In some embodiments, such a high-performance processing sled 1134 may be configured with far memory using an expansion sled that comprises low-latency SSD storage. The optical infrastructure allows for compute resources on one sled to utilize remote accelerator/FPGA, memory, and/or SSD resources that are disaggregated on a sled located on the same rack or any other rack in the data center. The remote resources can be located one switch jump away or two-switch jumps away in the spine-leaf network architecture described above with reference to FIG. 5. The embodiments are not limited in this context.
In various embodiments, one or more layers of abstraction may be applied to the physical resources of physical infrastructure 1100A in order to define a virtual infrastructure, such as a software-defined infrastructure 1100B. In some embodiments, virtual computing resources 1136 of software-defined infrastructure 1100B may be allocated to support the provision of cloud services 1140. In various embodiments, particular sets of virtual computing resources 1136 may be grouped for provision to cloud services 1140 in the form of software-defined infrastructure (SDI) services 1138. Examples of cloud services 1140 may include—without limitation—software as a service (SaaS) services 1142, platform as a service (PaaS) services 1144, and infrastructure as a service (IaaS) services 1146.
In some embodiments, management of software-defined infrastructure 1100B may be conducted using a virtual infrastructure management framework 1150B. In various embodiments, virtual infrastructure management framework 1150B may be designed to implement workload fingerprinting techniques and/or machine-learning techniques in conjunction with managing allocation of virtual computing resources 1136 and/or SDI services 1138 to cloud services 1140. In some embodiments, virtual infrastructure management framework 1150B may use/consult telemetry data in conjunction with performing such resource allocation. In various embodiments, an application/service management framework 1150C may be implemented in order to provide QoS management capabilities for cloud services 1140. The embodiments are not limited in this context.
Referring now to FIG. 12, a system 1200 for storage discovery and automatic repurposing of storage devices may be implemented in accordance with the data centers 100, 300, 400, 1100 described above with reference to FIGS. 1, 3, 4, and 11. In the illustrative embodiment, the system 1200 includes an orchestrator server 1202 communicatively coupled to multiple storage sleds including a storage sled 1210 and a storage sled 1220. While two storage sleds 1210, 1220 are shown, it should be understood that in other embodiments, the system 1200 may include a different number of storage sleds. One or more of the sleds 1210, 1220 may be grouped into a managed node, such as by the orchestrator server 1202, to collectively perform a workload (e.g., the application 1208). A managed node may be embodied as an assembly of resources (e.g., physical resources 206), such as compute resources (e.g., physical compute resources 205-4), memory resources (e.g., physical memory resources 205-3), storage resources (e.g., physical storage resources 205-1), or other resources (e.g., physical accelerator resources 205-2), from the same or different sleds (e.g., the sleds 204-1, 204-2, 204-3, 204-4, etc.) or racks (e.g., one or more of racks 302-1 through 302-32). Further, a managed node may be established, defined, or “spun up” by the orchestrator server 1202 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 1200 may be located in a data center and provide storage and compute services (e.g., cloud services) to a compute sled 1204 that is in communication with the system 1200 through a network 1290. The orchestrator server 1202 may support a cloud operating environment, such as OpenStack, and managed nodes established by the orchestrator server 1202 may execute one or more applications or processes (i.e., workloads), such as in virtual machines or containers, on behalf of a user of a client device (not shown).
In the illustrative embodiment, the storage sled 1210 is similar to the sled 204-1 of FIG. 2, and, in operation, executes an application 1208 (e.g., a workload). The storage sleds 1210, 1220, in the illustrative embodiment, each include one or more storage devices 1230, which in turn include a storage drive controller 1214 and one or more storage units 1218. Each storage drive controller 1214, in the illustrative embodiment, includes a storage health unit 1216. In the illustrative embodiment, the storage health unit 1216 which may be embodied as any device or circuitry (e.g., a processor, an FPGA, an ASIC, etc.) capable of executing the functions described in more detail below, with respect to FIG. 13. In addition, the storage device 1230 may be divided into smaller instances of storage devices, with each instance having its own set of resources (e.g., its own subsidiary storage units 1218).
As stated above, each storage device 1230 also includes a storage unit 1218. In the illustrative embodiment, a storage unit 1218 includes non-volatile memory (e.g., non-volatile write-in-place byte-addressable memory, solid-state drives (SSDs), and/or hard disk devices such as hard disk drives (HDDs). More specifically, a storage unit 1218, in the illustrative embodiment, may be embodied as a set of storage cells that are partitioned into provisionable space 1222 and usable storage capacity 1224. The usable storage capacity 1224 may be embodied as storage space to that is accessible to an application (e.g., to the compute sled 1204 executing the application 1208 to read and/or write persistent data. Over time, storage cells that are continuously used tend to wear out and become less reliable cells for storage. The provisionable space 1222 may be embodied as a sector, partition, or sub-part of the storage unit 1218 that is not presently allocated for access by an application (e.g., by the compute sled 1204 executing the application 1208), but includes spare storage cells that can replace worn-out storage cells from the usable storage capacity 1224.
Referring now to FIG. 13, the storage device 1230 may be embodied as any type of device or circuitry capable of performing the functions described herein, including obtaining performance specification data from performance specification documentation stored within the storage device 1230, performing a self-test, where the self-test is provided by, for example, an orchestrator server (e.g., the orchestrator server 1202) associated with the storage device 1230, reporting a self-test result of the self-test to a compute device (e.g., the orchestrator server 1202), receiving an adjustment message from the compute device (e.g., the orchestrator server 1202), and updating a storage device performance parameter in response to the adjustment message. As shown in FIG. 13, the illustrative storage device 1230 includes a compute engine 1302, an input/output (I/O) subsystem 1308, communication circuitry 1310, and one or more storage units 1218. 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 1302 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 1302 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. In the illustrative embodiment, the compute engine 1302 includes or is embodied as a storage drive controller 1304 (similar to the storage drive controller 1214) and a memory 1306. The storage drive controller 1304 may be embodied as any type of processor capable of performing the functions described herein. For example, the storage drive controller 1304 may be embodied as a microcontroller, a single or multi-core processor(s), or other processor or processing/controlling circuit. In some embodiments, the storage drive controller 1304 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. In the illustrative embodiment, the storage drive controller 1304 includes a storage health unit 1216 which may be embodied as any device or circuitry (e.g., a co-processor, an ASIC, etc.) capable of performing the storage discovery methods described herein.
The memory 1306 may be embodied as any type of volatile (e.g., dynamic random access memory (DRAM), etc.) 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 1306 may be integrated into the storage drive controller 1304. In operation, the memory 1306 may store various software and data used during operation such as performance data, provisionable capacity data, workload data, applications, programs, and libraries.
The compute engine 1302 is communicatively coupled to other components of the storage device 1230 via the I/O subsystem 1308, which may be embodied as circuitry and/or components to facilitate input/output operations with the compute engine 1302 (e.g., with the storage drive controller 1304 and/or the memory 1306) and other components of the storage device 1230. For example, the I/O subsystem 1308 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 1308 may form a portion of a system-on-a-chip (SoC) and be incorporated, along with one or more of the storage drive controller 1304, the memory 1306, and other components of the storage device 1230, into the compute engine 1302.
The communication circuitry 1310 may be embodied as any communication circuit, device, or collection thereof, capable of enabling communications over the network 1290 between the storage device 1230 and another compute device (e.g., the storage sled 1210, the orchestrator server 1202). The communication circuitry 1310 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 communication circuitry 1310 may include a network interface controller (NIC) 1314 (e.g., as an add-in device), which may also be referred to as a host fabric interface (HFI). The NIC 1312 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 storage device 1230 to connect with another compute device (e.g., the compute sled 1204), the orchestrator server 1202, etc.). In some embodiments, the NIC 1312 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 1312 may include a local processor (not shown) and/or a local memory (not shown) that are both local to the NIC 1312. In such embodiments, the local processor of the NIC 1312 may be capable of performing one or more of the functions of the compute engine 1302 described herein. Additionally or alternatively, in such embodiments, the local memory of the NIC 1312 may be integrated into one or more components of the storage device 1230 at the board level, socket level, chip level, and/or other levels. In the illustrative embodiment, the storage device 1230 accesses the network 1290 through a local bus (e.g., a peripheral component interconnect express bus (PCIe)) that is connected to one or more network interface controllers on the corresponding data storage sled 1210.
The storage device 1230 may also include one or more storage units 1218, which 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 storage unit 1218 may include a system partition that stores data and firmware code for the storage unit 1218. Each storage unit 1218 may also include one or more operating system partitions that store data files and executables for operating systems. As described above with respect to FIG. 12, each storage unit 1218 is partitioned into provisionable space 1222 and usable storage capacity 1224.
The orchestrator server 1202, the storage sleds 1210, 1220, and the compute sled 1204 may have components similar to those described in FIG. 13. Further, it should be appreciated that any of the storage device 1230, the storage sleds 1210, 1220, the orchestrator server 1202, or the compute sled 1204 may include other components, sub-components, and devices commonly found in a computing device, which are not discussed above in reference to the storage device 1230 and not discussed herein for clarity of the description.
As described above, the orchestrator server 1202, the sleds 1210, 1220, and the compute sled 1204 are illustratively in communication via the network 1290, which may be embodied as any type of wired or wireless communication network, including global networks (e.g., the Internet), local area networks (LANs) or wide area networks (WANs), cellular networks (e.g., Global System for Mobile Communications (GSM), 3G, Long Term Evolution (LTE), Worldwide Interoperability for Microwave Access (WiMAX), etc.), digital subscriber line (DSL) networks, cable networks (e.g., coaxial networks, fiber networks, etc.), or any combination thereof.
Referring now to FIG. 14, the orchestrator server 1202 may establish an environment 1400 during operation. The illustrative environment 1400 includes a network communicator 1420 and a storage device allocation manager 1430. Each of the components of the environment 1400 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 1400 may be embodied as circuitry or a collection of electrical devices (e.g., network communicator circuitry 1420, storage device allocation circuitry 1430, etc.). In the illustrative embodiment, the environment 1400 includes storage device performance data 1404 which may be embodied as any data indicative of statistics, metrics, scores, reports, or the like, pertaining to the performance of one or more storage devices 1230. For example, the storage device performance data 1404 may be embodied as any data indicative of latency, drive-writes-per-day (DWPD), throughput, input-output operations per second (IOPS), application workload data, or the like.
As used herein, drive writes per day refers to an amount of data that can be written to the storage unit 1218 per day. For example, if a storage unit 1218 is a 100 GB solid state drive and is specified for 1 DWPD, it can withstand 100 GB of data written to it every day for the warranty period (e.g., a period of time, such as three years, that is defined within a performance specification and/or service-level agreement for the storage unit 1218). Alternatively, if a 100 GB SSD is specified for 10 DWPD, it can withstand 1TB of data written to it every day for the warranty period. In addition, the storage device performance data 1404 may include values for the abovementioned data as reported by a storage device (e.g., the storage device 1230) as well as projected, anticipated, expected, or required values for the abovementioned data. For example, storage device performance data 1404 may include a latency threshold value that a storage device 1230 is expected to meet (e.g., pursuant to a service-level agreement).
Additionally, the environment 1400 includes write-intensive storage devices data 1404 and read-intensive storage devices data 1406. Write-intensive storage devices data 1404 may be embodied as any data indicative of identifiers and other data for storage devices that are configured for write-intensive operations and/or have been assigned to a write-intensive storage device pool. As used herein, a write-intensive storage device refers to a storage device 1230 that is configured for a high degree of write operations. For example, a write-intensive storage device is configured for a relatively high number of drive writes per day. Read-intensive storage devices data 1404 may be embodied as any data indicative of identifiers and other data for storage devices that are configured for read-intensive operations and/or have been assigned to a read-intensive storage device pool. As used herein, a read-intensive storage device refers to a storage device 1230 that is configured for a lower degree of write operations. In other words, a read-intensive storage device is mostly used for reading or accessing data. It is to be understood that write-intensive operations cause wear-out and failure on a storage device after a period of time. By contrast, simply reading data from a storage device causes much less impact on the underlying storage units 1218 and may enable the storage device 1230 to be used for longer than its warranty period.
In the illustrative environment 1400, the network communicator 1420, 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 storage device 1230, respectively. To do so, the network communicator 1420 is configured to receive and process data packets from one system or computing device (e.g., the storage sled 1210, the compute sled 1204, etc.) and to prepare and send data packets to a computing device or system (e.g., the storage sled 1210, the compute sled 1204, etc.). Accordingly, in some embodiments, at least a portion of the functionality of the network communicator 1420 may be performed by the communication circuitry 1310, and, in the illustrative embodiment, by the NIC 1312.
The storage device allocation manager 1430, which may be embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof, is configured to determine the health and performance of a connected storage device (e.g., the storage device 1230). If the determined performance fails to satisfy one or more performance thresholds, the storage device allocation manager 1430 is configured to cause the storage device 1230 to adjust its operation in one or more ways. For example, the storage device allocation manager 1430 may cause the storage device 1230 to adjust a performance parameter, or reallocate the storage device 1230 from one storage device pool to another (e.g., from a write-intensive pool to a read-intensive pool).
In the illustrative embodiment, the storage device allocation manager 1430 receives storage device data from the storage device 1230. The storage device data includes results from a storage device self-test. The storage device self-test is performed by the storage device 1230 to test one or more storage device performance parameters (e.g., latency, IOPS, throughput, bandwidth, response time, application workload, or the like). Each storage device performance parameter may adhere to specific threshold values (e.g., pursuant to a service-level agreement may define a performance specification for the storage device 1230). For example, the performance specification may state performance targets, including that the storage device 1230 is configured for a particular write latency (e.g., 80 microseconds). Write latency refers to the elapsed time to execute a data write operation. Accordingly, the storage device 1230 measures its own write latency at certain times. The storage device 1230 may also determine whether the measured write latency meets the threshold defined by the performance specification. The outcome is recorded in a self-test result and submitted to the storage device allocation manager 1430.
If the storage device 1230 satisfies performance thresholds, based on the self-test results, the storage device allocation manager 1430 may determine that the storage device is performing according to specifications. Otherwise (e.g., if the storage device 1230 does not satisfy the performance thresholds), the storage device allocation manager 1430 may generate an adjustment message for the storage device 1230. The adjustment message includes data to cause the storage device 1230 to adjust at least one storage device performance parameter. For example, the adjustment message may include a request to adjust power supply to a component of the storage device 1230 in order to reduce the write latency down to specified thresholds. As another example, the adjustment message may cause the storage device 1230 to select a faster connection to another compute device (e.g., the compute sled 1204) in order to reduce the write latency. As yet another example, the adjustment message may reduce an application workload for the storage device 1230 in order that the write latency for the remaining application workload satisfies performance thresholds.
In the illustrative embodiment, the storage device allocation manager 1430 determines whether a storage device 1230 should be reallocated to a different storage device pool. For example, the storage device 1230 may presently be in a write-intensive storage device pool. However, the storage device allocation manager 1430 may determine, based on one or more self-test results, that the storage device 1230 is unable to provide the write latency specified in the performance specification. The storage device allocation manager 1430 may arrive at this determination after performing one or more preliminary adjustments. For example, the storage device allocation manager 1430 may first adjust other storage device performance parameters (e.g., power, network connectivity) and transmit a request to the storage device 1230 to repeat the self-test. If the self-test results still fail to satisfy performance thresholds, the storage device allocation manager 1430 may determine to reallocate the storage device 1230 to a different storage device pool. As another example, the storage device allocation manager 1430 may immediately reallocate the storage device 1230 to a different storage device pool without first performing any preliminary adjustments.
In the illustrative embodiment, the storage device allocation manager 1430 may reallocate the storage device 1230 from a write-intensive storage device pool to a read-intensive storage device pool. In doing so, the storage device allocation manager 1430 may identify a storage device pool where the performance levels required of the storage devices 1230 match the performance of the storage device 1230. In the illustrative embodiment, the target storage device pool is a more read-intensive storage device pool. For example, a higher write latency may be tolerable in the target device pool, thus making it a more read-intensive storage device pool than the source device pool.
In another embodiment, the storage device allocation manager 1430 may determine, based on self-test results, whether the storage device 1230 is reaching a critical level of drive writes per day (DWPD). For example, a DWPD value of 1 may be specified for the storage device 1230 for a period of one year or 365 calendar days. The storage device allocation manager 1430 may determine that the storage device 1230 has handled at least one DWPD for the past 350 days. Accordingly, the storage device allocation manager 1430 may determine that write performance for the storage device 1230 will soon degrade if it remains in its present device pool. In other words, the storage device allocation manager 1430 may determine that the present device pool of the storage device 1230 will soon be too write-intensive for the storage device 1230. Continuing the example, the storage device allocation manager 1430 may then reallocate the storage device 1230 to a storage device pool where the DWPD level is lower, thereby extending the usable life of the storage device 1230.
Referring now to FIG. 15, the storage device 1230 may establish an environment 1500 during operation. The illustrative environment 1500 includes a network communicator 1520 and a storage health manager 1530. The storage health manager 1530, in the illustrative embodiment, includes a storage capacity allocator 1532, a performance estimator 1534, a self-test execution manager 1536, and a data input/output manager 1538. Each of the components of the environment 1500 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 1500 may be embodied as circuitry or a collection of electrical devices (e.g., network communicator circuitry 1520, storage health manager circuitry 1530, storage capacity allocator circuitry 1532, performance estimator circuitry 1534, self-test execution manager circuitry 1536, and data input/output manager 1538 circuitry, etc.).
In the illustrative embodiment, the environment 1500 includes performance data 1504, capacity data 1506, and workload data 1508. Performance data may be embodied as any data indicative of statistics, metrics, scores, reports or the like pertaining to a storage device's performance For example, performance data may include the warranty period for the storage device 1230, DWPD values for the storage device 1230, the mean time between failures (MTBF) values, uncorrectable bit error rate (UBER) values, or the like. The capacity data 1506 may be embodied as any data indicative of the provisionable and usable capacity (e.g., the provisionable space 1222 and the usable storage capacity 1224) of the storage units 1218 that are included within the storage device 1230. The capacity data 1506 may also include capacity provisioning ratio values for each storage unit 1218 within the storage device 1230. The capacity provisioning ratio may be embodied as the proportion of a storage unit 1218 that is usable storage space (e.g., usable storage capacity 1224) as opposed to provisionable space 1222. This capacity provisioning ratio may be expressed within the storage device 1230 as one or more of a percentage, a ratio, a set of locators, sectors, or cells assigned to be usable versus provisionable, or the like. For example, a storage unit 1218 of total capacity 100 GB may be configured such that 80 GB is usable storage space and 20 GB is provisionable space. Accordingly, the capacity provisioning ratio may be expressed as 80%, 20%, 4:1, the address ranges assigned to each type of space, or the like. The workload data 1508 may be embodied as any data indicative of application data stored in the storage device 1230, data indicative of applications that store data on the storage device 1230, and/or identifiers of the compute sleds or compute devices to which the storage device 1230 provides storage services.
The network communicator 1520 is similar to the network communicator 1420 described above with reference to FIG. 14. The storage health manager 1530, in the illustrative embodiment, is configured to allocate storage within the data storage unit(s) 1218, perform a self test of the performance of the data storage unit(s) 1218 relative to one or more performance thresholds, and manage access to the data stored in the data storage unit(s) 1218 (e.g., in response to access requests from remote compute device(s), such as the compute sled 1204). The storage capacity allocator 1532, in the illustrative embodiment, is configured to allocate storage capacity within the storage device 1230 (e.g., within the storage unit(s) 1218). For example, a storage device 1230, on initial startup, may receive a capacity provisioning ratio from the orchestrator server 1202. In response, the storage device 1230 may allocate space within the storage unit(s) 1218 according to the capacity provisioning ratio.
The performance estimator 1534 is configured to calculate and return performance values for a storage unit 1218 within the storage device 1230. Initially, the performance estimator 1534 may be programmed with performance specification data for a storage unit 1218. Performance specification data may be obtained from documentation or a local or remote database. Using the performance specification data, the performance estimator 1534 may determine initial values for one or more performance parameters for the storage unit 1218. For example, the performance estimator 1534 may determine, for the storage unit 1218, a DWPD value, latency values, reliability values, temperature values, bandwidth values, throughput values, or the like. As the storage unit 1218 operates, the performance estimator 1534 may calculate performance data such as a present DWPD value for the storage unit 1218, return a remaining warranty period for the storage unit 1218, return a provisionable space value and/or return a remaining number of unused storage cells for the storage unit 1218. The performance estimator 1534 is also configured to return a specific identifier or metric to represent a remaining lifespan of a storage unit 1218. The metric is may be a function of various performance data values that are described above with respect to performance data 1504.
The self-test execution manager 1536 is configured to receive a self-test from the orchestrator server 1202, perform the self-test with respect to one or more storage units 1218 of the storage device 1230, and report results to the orchestrator server 1202. In the illustrative embodiment, the storage device allocation manager 1430 of the orchestrator server 1202 analyzes the performance values generated by the performance estimator 1534 to generate a self-test routine tailored to the specifications of a particular storage unit 1218. The orchestrator server 1202 may send the self-test routine to the storage device 1230, which executes the self-test with the self-test execution manager 1536. The self-test execution manager 1536, in the illustrative embodiment, collects and reports the results of the self-test to the orchestrator server 1202.
The data input/output (I/O) manager 1538 is configured to manage inputs and outputs from and to other compute devices respectively. For example, the data I/O manager 1538 manages inputs from an orchestrator server 1202 such as self-tests or from a compute sled 1204 such as application data. As another example, the data I/O manager 1538 manages outputs to the orchestrator server 1202 such as self-test results or to the compute sled 1204 such as application data requested by an application to be retrieved from storage (e.g., from workload data 1508).
It should be appreciated that each of the storage health manager 1530, the storage capacity allocator 1532, the performance estimator 1534, the self-test execution manager 1536, and the data input/output manager 1538 may be separately embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof. For example, the storage health manager 1530 may be embodied as a hardware component, while the storage capacity allocator 1532, the performance estimator 1534, the self-test execution manager 1536, and the data input/output manager 1538 are embodied as virtualized hardware components or as some other combination of hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof.
Referring now to FIG. 16, the storage device 1230, in operation, may execute a method 1600 for storage discovery and automatic repurposing of storage devices. The method 1600 begins with block 1602, in which the storage device 1230 determines to perform the storage discovery methods described herein. In block 1604, the storage device 1230 obtains storage drive performance specification data. As described above with respect to FIG. 15, the performance estimator 1534 may access performance specification data. As an example, and as illustrated in block 1606, the storage device 1230 may read performance specification data from non-volatile memory in the storage device 1230. For example, performance specification data may have been stored at an earlier time in a storage unit 1218. Alternatively, and as illustrated in block 1608, the storage device 1230 may electronically read and parse performance specification documents received from an external database, or read in via an input/output device (e.g., a scanner). In block 1610, the storage device 1230 may store the performance specification data in the storage device 1230 (e.g., if not already stored). In block 1612, the storage device 1230 sends the performance specification data to the orchestrator server 1202 (e.g., through the network 1290).
In block 1614, the storage device 1230 may determine testable performance specification values. Irrespective of a self-test routine to be later received from an orchestrator server 1202, the storage device 1230 may be identify and distinguish testable performance specification values from non-testable values (e.g., drive form factor, weight, size specifications, etc.). More specifically, as shown in block 1616, the storage device 1230 retrieves expected performance specification values from, for example, a performance specification document or service-level agreement.
In block 1618, the storage device 1230 performs a self-test of one or more storage units 1218. In the illustrative embodiment, a self-test is designed to measure the performance of the storage unit 1218 using one or more performance parameters (e.g., latency, IOPS, throughput, etc.). As described above, the orchestrator server 1202 may generate a self-test routine tailored to the storage device 1230 and/or a storage unit 1218 and transmit the self-test routine to the storage device 1230. In another embodiment, the storage device 1230 is configured with a self-test that is programmed within the storage drive controller 1304. As illustrated in block 1620, the storage device 1230 may test the IOPS values for a storage unit 1218 of the data storage device 1230. As another example, shown in block 1622, the storage device 1230 may test drive latency for a storage unit 1218 (e.g., by measuring the time taken to write a test data set to the data storage unit 1218 the data to be written).
As yet another example, in block 1624, the storage device 1230 may determines the current DWPD value. More specifically, the current DWPD value may be a calculation of the total drive writes to the storage unit 1218 (e.g., in gigabytes) divided by the total number of days of operation. The storage device 1230 may also determine the possible future drive writes per day value (e.g., by identifying a historical trend in DWPD over multiple days and projecting the trend out to a particular date), as indicated in block 1626. As still another example, shown in block 1628, the storage device 1230 determines a temperature of the storage unit 1230. As a still further example, shown in block 1630, the storage device 1230 determines network metrics of the storage unit 1230 (such as throughput, bandwidth, etc.). In block 1632, the storage device 1230 reports results of the self-test to the orchestrator server 1202. In some embodiments, in reporting the self-test results, the storage device 1230 sends the values determined in one or more of the blocks 1620, 1622, 1624, 1626, 1628, 1630 to the orchestrator server 1202. In other embodiments, the storage device 1230 sends a result of a comparison of one or more of the values determined in blocks 1620, 1622, 1624, 1626, 1628, 1630 to a corresponding target value (e.g., a value defined in the specification data, such as a target latency or IOPS that is to be satisfied). For example the storage device 1230 may send an indication that the measured IOPS is 95% of the target IOPS defined in the specification data or that the measured latency is 103% of the latency defined in the specification data.
Referring now to FIG. 17, in block 1634, the storage device 1230 determines whether an adjustment message was received from the orchestrator server 1202. If no adjustment message is received, the method loops back to block 1618, where the storage device 1230 continues to perform self-testing (e.g., on a predetermined schedule). If an adjustment message is received, the method 1600 branches to block 1636. In block 1636, the storage device 1230 implements the adjustment instructions included within the adjustment message. For example, and as shown in block 1638, the storage device 1230 adjusts the usable storage space of a storage unit 1218. In the illustrative embodiment, the sum of usable storage space and provisionable space does not exceed the total available space within a storage unit 1218. Accordingly, the adjustment instruction may be to reconfigure the storage unit 1218 with more usable storage space (and thus less provisionable space), thereby rendering the storage unit 1218 relatively more read-intensive than before. As noted earlier, this is because a reduction in provisionable space reduces a number of available unused storage cells that can replace aged or worn-out cells within the usable storage space, thereby reducing the number of times the storage unit 1218 can be written to. However, reducing provisionable space does not affect the ability of a storage unit 1218 to fulfill data read requests. By contrast, the adjustment instruction may be to reconfigure the storage unit 1218 with less usable storage space (and thus more provisionable space), and make the storage unit 1218 relatively more write-intensive. Similarly, and as shown in block 1640, the storage device 1230 may adjust the provisionable space of the storage unit 1218 in response to the adjustment instruction. Increasing the provisionable space will bring about a reduction in usable storage space, and vice versa.
As another example of adjustment, shown in block 1642, the storage device 1230 may change its logical connection with one or more compute devices (e.g., to discontinue data access services for a sled (e.g., a compute sled) executing a workload (e.g., an application) that is write-intensive and begin data access services for a sled executing a workload that is more read-intensive than write intensive). For instance, the adjustment may be to have the storage device 1230 discontinue data access operations for a compute sled or accelerator sled whose performance requirements are no longer being met by the storage device 1230. The adjustment instruction may be to increase or decrease a connectivity speed (e.g., by using a different port or connection methodology and/or allocating a larger or smaller portion of the total communication bandwidth to servicing data access requests from a particular sled). As yet another example of adjustment, shown in block 1644, the storage device 1230 updates one or more configuration values in response to a new allocation by the orchestrator server 1202. The adjustment message includes information on the reallocation. For example, the adjustment message may inform the storage device 1230 that it has been reassigned to a different storage device pool (e.g., a read-intensive storage device pool).
Referring now to FIG. 18, the orchestrator server 1202 may execute a method 1800 for storage discovery and automatic repurposing of storage devices. In block 1802, the orchestrator server 1202 determines whether to monitor a storage device (e.g., a storage device 1230). If so, the method branches to block 1804, where the orchestrator server 1202 sets a capacity provisioning ratio for the storage device 1230. As described above, the capacity provisioning ratio refers to the proportion of a storage unit 1218 that is usable storage space as opposed to provisionable space. This capacity provisioning ratio may be expressed within the storage device 1230 as one or more of a percentage, a ratio, a set of locators, sectors, cells assigned to be usable versus provisionable, or the like. The orchestrator server 1202 sets the capacity provisioning ratio as a function of a number of variables, including, for example, the anticipated pool assignment (e.g., to a write-intensive pool or to a read-intensive pool) for the storage device 1230, the expected performance requirements of other compute devices that will request data access services from the storage device 1230, the reported performance specifications of the storage device 1230, and the like. In block 1806, the orchestrator server 1202 transmits the capacity provisioning ratio to the storage device 1230.
In block 1808, the orchestrator server 1202 monitors storage device self-test results. As described above with respect to block 1632 of FIG. 16, the storage device 1230 sends self-test results to the orchestrator server 1202. In the illustrative embodiment, the self-tests may be performed periodically, on demand by the orchestrator server 1202, in response to a stimulus by the storage device 1230 or other connected compute device, or in response to an event (e.g., initial startup, boot time, an error, a failure, a power surge, an unexpected data read or write event, etc.). In block 1810, the orchestrator server 1202 analyzes the received self-test results and determines if the storage device 1230 is meeting one or more performance targets or thresholds (e.g., pursuant to performance specification data, pursuant to a service level agreement, etc.). Accordingly, and as shown in block 1812, the orchestrator server 1202 may query compute devices connected to or reliant on the storage device 1230 (e.g., to the storage sled 1210, 1220 in which the storage device 1230 is located) for performance targets or may access service-level agreement documentation for performance target data.
The method advances to block 1814, where the orchestrator server 1202 compares self-test results received from the storage device 1230 to the service-level agreement targets and also any other performance targets received from compute devices that have been assigned to use the storage device 1230. As an example, in block 1816, the orchestrator server 1202 compares latency values from the self-test to latency targets. More specifically, the orchestrator server 1202 compares, for example, a write latency value returned from the storage device 1230 to write latency targets provided from performance specification data within the service-level agreement and/or write latency requirements requested by individual compute sleds, accelerator sleds, or the like. As another example, in block 1818, the orchestrator server 1202 compares IOPS values to IOPS targets. As yet another example, in block 1820, the orchestrator server 1202 compares drive writes per day values to present drive writes per day targets. As a still further example, in block 1822, the orchestrator server 1202 compares other health metrics (e.g., network connectivity speeds, temperature, noise levels, etc.).
In block 1824, the orchestrator server 1202 determines whether the self-test results meet the expected or required performance levels. If the results meet expectations, the method loops back to block 1802 and continues to monitor the storage device 1230. If the results do not meet expectations, the method branches to block 1826 of FIG. 19. Referring now to FIG. 19, in block 1826, the orchestrator server 1202 determines an adjustment for the storage device 1230. For example, the orchestrator server 1202 determines a type of adjustment (e.g., reallocation, configuration change, update connections to compute devices (e.g., sleds) that are to remotely access the storage device 1230, etc.) or a degree of adjustment (adjustment to capacity provisioning ratio, etc.).
For example, and as illustrated in block 1828, the orchestrator server 1202 determines whether the capacity provisioning ratio assigned to the storage device 1230 needs to be updated. The capacity provisioning ratio may have been assigned to the entire storage device 1230 or to individual storage units 1218 of the storage device 1230. As described above, in response to a determination that the write latency of a storage device 1230 is increasing, the orchestrator server 1202 may determine that the write performance of the storage device 1230 is declining. Accordingly, the orchestrator server 1202 may determine to increase the provisionable space for one or more storage units 1218 of the storage device 1230 in order to make available greater numbers or fresh storage cells that can more quickly replace aged storage cells in the usable space. The orchestrator server 1202 may determine that such an adjustment to the capacity provisioning ratio will bring the write latency of the storage device 1230 closer to expected performance levels.
As another example, in block 1830, the orchestrator server 1202 determines whether a drive writes per day value exceeds a threshold. If the DWPD value exceeds a performance threshold (e.g., a threshold based on the age of the storage device 1230), the orchestrator server 1202 may determine that the storage device 1230 will soon (e.g., within a predefined number of days) reach an end of life and become unusable at least as a write-intensive storage device. As a result of the determinations in block 1826, the orchestrator server 1202 generates an adjustment message in block 1832. More specifically, in block 1834, the orchestrator server 1202 may set an updated capacity provisioning ratio for the storage device 1230 as described above. As another example, and as illustrated in block 1836, the orchestrator server 1202 may update the storage device allocation. More specifically, the storage device 1230 may be reallocated to a different storage device pool (e.g., to a more read-intensive storage device pool or to a more write-intensive storage device pool). The storage device 1230 may also be decommissioned. As illustrated in block 1838, the orchestrator server 1202 may implement a progressive or tiered reallocation of the storage device 1230. For example, the orchestrator server 1202 may execute a planned de-escalation of write intensity for the storage device 1230. In doing so, the orchestrator server 1202 may continually reassign the storage device 1230 to progressively more read-intensive storage device pools (e.g., on a predetermined schedule). Doing so may extend the usable lifetime of the storage device 1230.
As another example, shown in block 1840, the orchestrator server 1202 may adjust other storage device configurations. For example, the orchestrator server 1202 may adjust execution times of certain types of software (e.g., security software in the storage device 1230 that may reduce performance) to increase the availability of the storage device 1230 to respond to data access requests from other sleds (e.g., from the compute sled 1204) during certain times of day. The orchestrator server 1202 may adjust a boot schedule for the storage device 1230. Additionally or alternatively, the orchestrator server 1202 may adjust a connectivity or component usage for the storage device 1230, as another example. In block 1842, the orchestrator server 1202 transmits the adjustment message bearing one or more of the aforementioned adjustment instructions to the storage device 1230. As described above with respect to block 1636, this causes the storage device 1230 to implement the one or more adjustment instructions.

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 compute device to manage storage device performance, the compute device comprising a compute engine to receive, from a data storage sled, storage device data from a storage device located on the data storage sled, wherein the storage device data includes storage device self-test data that defines a result of a self-test performed by the storage device; determine, in response to the storage device self-test data, whether the storage device fails to satisfy a performance threshold; generate, in response to a determination that the storage device fails to satisfy the performance threshold, an adjustment message for the storage device, wherein the adjustment message instructs the storage device to adjust a performance parameter of the storage device; and send the adjustment message to the storage device.
Example 2 includes the subject matter of Example 1, and wherein to generate the adjustment message comprises to generate an adjustment message that instructs the storage device to adjust a capacity provisioning ratio for the storage device, wherein the capacity provisioning ratio defines a ratio of usable storage space to provisionable space for the storage device.
Example 3 includes the subject matter of any of Examples 1 and 2, and wherein the compute engine is to set the capacity provisioning ratio for the storage device, and wherein the compute engine is further to receive performance specification data from the storage device on a first startup of the storage device, wherein the performance specification data includes a performance target for the storage device; generate the capacity provisioning ratio for the storage device as a function of at least one of an anticipated device pool allocation for the storage device and the performance target; and send the capacity provisioning ratio to the storage device to cause the storage device to apportion a usable storage space and a provisionable space within a storage unit of the storage device as a function of the capacity provisioning ratio.
Example 4 includes the subject matter of any of Examples 1-3, and wherein the storage device is associated with a first device pool, and wherein the compute engine is further to determine, in response to receipt of the storage device self-test data, whether the storage device fails to satisfy a performance threshold for the first device pool; determine whether the storage device satisfies a performance threshold for a second device pool, different from the first device pool; and reallocate, in response to to determination that the storage device satisfies the performance threshold for the second device pool, the storage device to the second device pool.
Example 5 includes the subject matter of any of Examples 1-4, and wherein to generate the adjustment message comprises to generate an adjustment message to reallocate the storage device from a write-intensive storage device pool to a read-intensive storage device pool.
Example 6 includes the subject matter of any of Examples 1-5, and wherein a first drive writes per day value for the first device pool exceeds a second drive writes per day value for the second device pool.
Example 7 includes the subject matter of any of Examples 1-6, and wherein the storage device performance parameter includes an identifier for a compute sled connected to the storage device, an amount of usable storage space, an amount of provisionable space, or a storage device configuration value.
Example 8 includes the subject matter of any of Examples 1-7, and wherein the compute engine is further to query a compute sled connected to the storage device for a performance target of the storage device, wherein the performance target is associated with a workload to be executed by the compute sled; determine whether the storage device fails to satisfy the performance target; and generate, in response to a determination that the storage device fails to satisfy the performance target, a second adjustment message for the storage device, wherein the second adjustment message instructs the storage device to satisfy the performance target associated with the compute sled.
Example 9 includes the subject matter of any of Examples 1-8, and wherein to determine whether the storage device fails to satisfy a performance threshold comprises to retrieve at least one of a latency value, an input-output per second (IOPS) value, or a drive writes per day (DWPD) value from the storage device self-test data; determine, as a function of performance specification data that includes a performance target for the storage device, at least one of a latency threshold, an IOPS threshold, or a DWPD threshold; and compare one or more of the latency value to the latency threshold, the IOPS value to the IOPS threshold, or the DWPD value to the DWPD threshold.
Example 10 includes the subject matter of any of Examples 1-9, and wherein the compute engine is further to identify the one or more performance thresholds based on performance specification data that includes a performance target for the storage device; generate, as a function of the one or more performance thresholds, a self-test for the storage device to perform; and transmit the self-test to the storage device.
Example 11 includes the subject matter of any of Examples 1-10, and wherein the compute engine is to compose a managed node that includes the storage device.
Example 12 includes a method to manage storage device performance, the method comprising receive, from a data storage sled, storage device data from a storage device located on the data storage sled, wherein the storage device data includes storage device self-test data that defines a result of a self-test performed by the storage device; determine, in response to the storage device self-test data, whether the storage device fails to satisfy a performance threshold; generate, in response to a determination that the storage device fails to satisfy the performance threshold, an adjustment message for the storage device, wherein the adjustment message instructs the storage device to adjust a performance parameter of the storage device; and send the adjustment message to the storage device.
Example 13 includes the subject matter of Example 12, and wherein generating the adjustment message comprises generating an adjustment message that instructs the storage device to adjust a capacity provisioning ratio for the storage device, wherein the capacity provisioning ratio defines a ratio of usable storage space to provisionable space for the storage device.
Example 14 includes the subject matter of any of Examples 12 and 13, and further including receiving, by the compute device, performance specification data from the storage device on a first startup of the storage device, wherein the performance specification data includes a performance target for the storage device; generating, by the compute device, the capacity provisioning ratio for the storage device as a function of at least one of an anticipated device pool allocation for the storage device and the performance target; and sending, by the compute device, the capacity provisioning ratio to the storage device to cause the storage device to apportion a usable storage space and a provisionable space within a storage unit of the storage device as a function of the capacity provisioning ratio.
Example 15 includes the subject matter of any of Examples 12-14, and wherein the storage device is associated with a first device pool, the method further comprising determining, by the compute device and in response to receipt of the storage device self-test data, whether the storage device fails to satisfy a performance threshold for the first device pool; determining, by the compute device, whether the storage device satisfies a performance threshold for a second device pool, different from the first device pool; and reallocating, by the compute device and in response to a determination that the storage device satisfies the performance threshold for the second device pool, the storage device to the second device pool.
Example 16 includes the subject matter of any of Examples 12-15, and wherein generating the adjustment message comprises generating an adjustment message to reallocate the storage device from a write-intensive storage device pool to a read-intensive storage device pool.
Example 17 includes the subject matter of any of Examples 12-16, and wherein a first drive writes per day value for the first device pool exceeds a second drive writes per day value for the second device pool.
Example 18 includes the subject matter of any of Examples 12-17, and wherein the storage device performance parameter includes an identifier for a compute sled connected to the storage device, an amount of usable storage space, an amount of provisionable space, or a storage device configuration value.
Example 19 includes the subject matter of any of Examples 12-18, and further including querying, by the compute device, a compute sled connected to the storage device for a performance target of the storage device, wherein the performance target is associated with a workload to be executed by the compute sled; determining, by the compute device, whether the storage device fails to satisfy the performance target; and generating, by the compute device and in response to a determination that the storage device fails to satisfy the performance target, a second adjustment message for the storage device, wherein the second adjustment message instructs the storage device to satisfy the performance target associated with the compute sled.
Example 20 includes the subject matter of any of Examples 12-19, and wherein determining whether the storage device fails to satisfy a performance threshold comprises retrieving at least one of a latency value, an input-output per second (IOPS) value, or a drive writes per day (DWPD) value from the storage device self-test data; determining, as a function of performance specification data that includes a performance target for the storage device, at least one of a latency threshold, an IOPS threshold, or a DWPD threshold; and comparing one or more of the latency value to the latency threshold, the IOPS value to the IOPS threshold, or the DWPD value to the DWPD threshold.
Example 21 includes the subject matter of any of Examples 12-20, and further including identifying, by the compute device, the one or more performance thresholds based on performance specification data that includes a performance target for the storage device; generating, by the compute device and as a function of the one or more performance thresholds, a self-test for the storage device to perform; and transmitting, by the compute device, the self-test to the storage device.
Example 22 includes the subject matter of any of Examples 12-21, and further including composing, by the compute device, a managed node that includes the storage device.
Example 23 includes a compute device comprising a compute engine to perform the method of any of Examples 12-22.
Example 24 includes a compute device comprising means for performing the method of any of Examples 12-22.
Example 25 includes one or more machine-readable storage media comprising a plurality of instructions stored thereon that, in response to being executed, cause a compute device to perform the method of any of Examples 12-22.
Example 26 includes a storage device to manage storage performance, the storage device comprising a non-volatile memory; and a storage drive controller to perform a self-test on the non-volatile memory to produce self-test data indicative of a present performance of the storage device relative to a performance target associated with specification data for the storage device; report, through a network, the self-test data to a compute device; receive, through the network, an adjustment message from the compute device; and update, in response to the adjustment message, a storage device performance parameter.
Example 27 includes the subject matter of Example 26, and wherein to perform the self-test comprises to select at least one storage device performance parameter for the storage device; identify a present value for the at least one storage device performance parameter; and generate self-test data that includes the present value for the at least one storage device performance parameter.
Example 28 includes the subject matter of any of Examples 26 and 27, and wherein the storage drive controller is further to identify the performance target from the performance specification data; compare the present performance parameter to the performance target; determine whether the storage device fails to satisfy the performance target; and report, to the compute device, the determination as part of the self-test.
Example 29 includes the subject matter of any of Examples 26-28, and wherein to update the storage device performance parameter comprises to update at least one of a usable storage space parameter, a provisionable space parameter, or a drive writes per day value for the storage device.
Example 30 includes the subject matter of any of Examples 26-29, and wherein to perform the self-test comprises to determine a present number of input/output instructions per second executable by the storage device.
Example 31 includes the subject matter of any of Examples 26-30, and wherein to perform the self-test comprises to determine a present latency of the storage device.
Example 32 includes the subject matter of any of Examples 26-31, and wherein to perform the self-test comprises to determine, as a function of a historical trend in drive writes per day (DWPD), an expected number of drive writes per day (DWPD) to be performed by the storage device; and determine a change to a present DWPD value caused by an update to a provisionable space parameter of the storage device.
Example 33 includes a method for managing storage performance, the method comprising performing, by a storage device, a self-test on non-volatile memory of the storage device to produce self-test data indicative of a present performance of the storage device relative to a performance target associated with specification data for the storage device; reporting, by the storage device and through a network, the self-test data to a compute device; receiving, by the storage device and through the network, an adjustment message from the compute device; and updating, by the storage device and in response to the adjustment message, a storage device performance parameter.
Example 34 includes the subject matter of Example 33, and wherein performing the self-test comprises selecting at least one storage device performance parameter for the storage device; identifying a present value for the at least one storage device performance parameter; and generating self-test data that includes the present value for the at least one storage device performance parameter.
Example 35 includes the subject matter of any of Examples 33 and 34, and further including identifying, by the storage device, the performance target from the performance specification data; comparing, by the storage device, the present performance parameter to the performance target; determining, by the storage device, whether the storage device fails to satisfy the performance target; and reporting, by the storage device and to the compute device, the determination as part of the self-test.
Example 36 includes the subject matter of any of Examples 33-35, and wherein updating the storage device performance parameter comprises updating at least one of a usable storage space parameter, a provisionable space parameter, or a drive writes per day value for the storage device.
Example 37 includes the subject matter of any of Examples 33-36, and wherein performing the self-test comprises determining a present number of input/output instructions per second executable by the storage device.
Example 38 includes the subject matter of any of Examples 33-37, and wherein performing the self-test comprises determining a present latency of the storage device.
Example 39 includes the subject matter of any of Examples 33-38, and wherein performing the self-test comprises determining, as a function of a historical trend in drive writes per day (DWPD), an expected number of drive writes per day (DWPD) to be performed by the storage device; and determining a change to a present DWPD value caused by an update to a provisionable space parameter of the storage device.
Example 40 includes a storage device comprising a storage drive controller to perform the method of any of Examples 33-39.
Example 41 includes a storage device comprising means for performing the method of any of Examples 33-39.
Example 42 includes one or more machine-readable storage media comprising a plurality of instructions stored thereon that, in response to being executed, cause a compute device to perform the method of any of Examples 33-39.
Example 43 includes a compute device comprising means for receiving, from a data storage sled, storage device data from a storage device located on the data storage sled, wherein the storage device data includes storage device self-test data that defines a result of a self-test performed by the storage device; means for determining, in response to the storage device self-test data, whether the storage device fails to satisfy a performance threshold; means for generating, in response to a determination that the storage device fails to satisfy the performance threshold, an adjustment message for the storage device, wherein the adjustment message instructs the storage device to adjust a performance parameter of the storage device; and means for sending the adjustment message to the storage device.
Example 44 includes the subject matter of Example 43, and wherein the means for generating the adjustment message comprises means for generating an adjustment message that instructs the storage device to adjust a capacity provisioning ratio for the storage device, wherein the capacity provisioning ratio defines a ratio of usable storage space to provisionable space for the storage device.
Example 45 includes the subject matter of any of Examples 43 and 44, and wherein the compute engine is to set the capacity provisioning ratio for the storage device, the compute device further comprising means for receiving performance specification data from the storage device on a first startup of the storage device, wherein the performance specification data includes a performance target for the storage device; means for generating the capacity provisioning ratio for the storage device as a function of at least one of an anticipated device pool allocation for the storage device and the performance target; and means for sending the capacity provisioning ratio to the storage device to cause the storage device to apportion a usable storage space and a provisionable space within a storage unit of the storage device as a function of the capacity provisioning ratio.
Example 46 includes the subject matter of any of Examples 43-45, and wherein the storage device is associated with a first device pool, the compute device further comprising means for determining, in response to receipt of the storage device self-test data, whether the storage device fails to satisfy a performance threshold for the first device pool; means for determining whether the storage device satisfies a performance threshold for a second device pool, different from the first device pool; and means for reallocating, in response to to determination that the storage device satisfies the performance threshold for the second device pool, the storage device to the second device pool.
Example 47 includes the subject matter of any of Examples 43-46, and wherein the means for generating the adjustment message comprise means for generating an adjustment message to reallocate the storage device from a write-intensive storage device pool to a read-intensive storage device pool.
Example 48 includes the subject matter of any of Examples 43-47, and wherein a first drive writes per day value for the first device pool exceeds a second drive writes per day value for the second device pool.
Example 49 includes the subject matter of any of Examples 43-48, and wherein the storage device performance parameter includes an identifier for a compute sled connected to the storage device, an amount of usable storage space, an amount of provisionable space, or a storage device configuration value.
Example 50 includes the subject matter of any of Examples 43-49, and further including means for querying a compute sled connected to the storage device for a performance target of the storage device, wherein the performance target is associated with a workload to be executed by the compute sled; means for determining whether the storage device fails to satisfy the performance target; and means for generating, in response to a determination that the storage device fails to satisfy the performance target, a second adjustment message for the storage device, wherein the second adjustment message instructs the storage device to satisfy the performance target associated with the compute sled.
Example 51 includes the subject matter of any of Examples 43-50, and wherein the means for determining whether the storage device fails to satisfy a performance threshold comprises means for retrieving at least one of a latency value, an input-output per second (IOPS) value, or a drive writes per day (DWPD) value from the storage device self-test data; means for determining, as a function of performance specification data that includes a performance target for the storage device, at least one of a latency threshold, an IOPS threshold, or a DWPD threshold; and means for comparing one or more of the latency value to the latency threshold, the IOPS value to the IOPS threshold, or the DWPD value to the DWPD threshold.
Example 52 includes the subject matter of any of Examples 43-51, and further including means for identifying the one or more performance thresholds based on performance specification data that includes a performance target for the storage device; means for generating, as a function of the one or more performance thresholds, a self-test for the storage device to perform; and means for transmitting the self-test to the storage device.
Example 53 includes the subject matter of any of Examples 43-52, and further including means for composing a managed node that includes the storage device.
Example 54 includes a storage device to manage storage performance, the storage device comprising a non-volatile memory; means for performing a self-test on the non-volatile memory to produce self-test data indicative of a present performance of the storage device relative to a performance target associated with specification data for the storage device; means for reporting, through a network, the self-test data to a compute device; means for receiving, through the network, an adjustment message from the compute device; and means for updating, in response to the adjustment message, a storage device performance parameter.
Example 55 includes the subject matter of Example 54, and wherein the means for performing the self-test comprises means for selecting at least one storage device performance parameter for the storage device; means for identifying a present value for the at least one storage device performance parameter; and means for generating self-test data that includes the present value for the at least one storage device performance parameter.
Example 56 includes the subject matter of any of Examples 54 and 55, and further including means for identifying the performance target from the performance specification data; means for comparing the present performance parameter to the performance target; means for determining whether the storage device fails to satisfy the performance target; and means for reporting, to the compute device, the determination as part of the self-test.
Example 57 includes the subject matter of any of Examples 54-56, and wherein the means for updating the storage device performance parameter comprises means for updating at least one of a usable storage space parameter, a provisionable space parameter, or a drive writes per day value for the storage device.
Example 58 includes the subject matter of any of Examples 54-57, and wherein the means for performing the self-test comprises means for determining a present number of input/output instructions per second executable by the storage device.
Example 59 includes the subject matter of any of Examples 54-58, and wherein the means for performing the self-test comprises means for determining a present latency of the storage device.
Example 60 includes the subject matter of any of Examples 54-59, and wherein the means for performing the self-test comprise means for determining, as a function of a historical trend in drive writes per day (DWPD), an expected number of drive writes per day (DWPD) to be performed by the storage device; and means for determining a change to a present DWPD value caused by an update to a provisionable space parameter of the storage device.

Claims

1. A compute device to manage storage device performance, the compute device comprising:

a compute engine to:

receive, from a data storage sled, storage device data from a storage device located on the data storage sled, wherein the storage device data includes storage device self-test data that defines a result of a self-test performed by the storage device;

determine, in response to the storage device self-test data, whether the storage device fails to satisfy a performance threshold;

generate, in response to a determination that the storage device fails to satisfy the performance threshold, an adjustment message for the storage device, wherein the adjustment message instructs the storage device to adjust a performance parameter of the storage device; and

send the adjustment message to the storage device.

2. The compute device of claim 1, wherein to generate the adjustment message comprises to generate an adjustment message that instructs the storage device to adjust a capacity provisioning ratio for the storage device, wherein the capacity provisioning ratio defines a ratio of usable storage space to provisionable space for the storage device.

3. The compute device of claim 2, wherein the compute engine is to set the capacity provisioning ratio for the storage device, and wherein the compute engine is further to:

receive performance specification data from the storage device on a first startup of the storage device, wherein the performance specification data includes a performance target for the storage device;

generate the capacity provisioning ratio for the storage device as a function of at least one of an anticipated device pool allocation for the storage device and the performance target; and

send the capacity provisioning ratio to the storage device to cause the storage device to apportion a usable storage space and a provisionable space within a storage unit of the storage device as a function of the capacity provisioning ratio.

4. The compute device of claim 1, wherein the storage device is associated with a first device pool, and wherein the compute engine is further to:

determine, in response to receipt of the storage device self-test data, whether the storage device fails to satisfy a performance threshold for the first device pool;

determine whether the storage device satisfies a performance threshold for a second device pool, different from the first device pool; and

reallocate, in response to to determination that the storage device satisfies the performance threshold for the second device pool, the storage device to the second device pool.

5. The compute device of claim 4, wherein a first drive writes per day value for the first device pool exceeds a second drive writes per day value for the second device pool.

6. The compute device of claim 1, wherein the compute engine is further to:

query a compute sled connected to the storage device for a performance target of the storage device, wherein the performance target is associated with a workload to be executed by the compute sled;

determine whether the storage device fails to satisfy the performance target; and

generate, in response to a determination that the storage device fails to satisfy the performance target, a second adjustment message for the storage device, wherein the second adjustment message instructs the storage device to satisfy the performance target associated with the compute sled.

7. The compute device of claim 1, wherein to determine whether the storage device fails to satisfy a performance threshold comprises to:

retrieve at least one of a latency value, an input-output per second (IOPS) value, or a drive writes per day (DWPD) value from the storage device self-test data;

determine, as a function of performance specification data that includes a performance target for the storage device, at least one of a latency threshold, an IOPS threshold, or a DWPD threshold; and

compare one or more of the latency value to the latency threshold, the IOPS value to the IOPS threshold, or the DWPD value to the DWPD threshold.

8. The compute device of claim 1, wherein the compute engine is further to:

identify the one or more performance thresholds based on performance specification data that includes a performance target for the storage device;

generate, as a function of the one or more performance thresholds, a self-test for the storage device to perform; and

transmit the self-test to the storage device.

9. The compute device of claim 1, wherein to determine whether the storage device fails to satisfy a performance threshold comprises to:

10. The compute device of claim 1, wherein the compute engine is further to:

transmit the self-test to the storage device.

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

send the adjustment message to the storage device.

12. The one or more machine-readable storage media of claim 11, wherein the plurality of instructions further cause compute device to generate an adjustment message that instructs the storage device to adjust a capacity provisioning ratio for the storage device, wherein the capacity provisioning ratio defines a ratio of usable storage space to provisionable space for the storage device.

13. The one or more machine-readable storage media of claim 11, wherein the plurality of instructions further cause the compute device to:

set the capacity provisioning ratio for the storage device;

14. The one or more machine-readable storage media of claim 11, wherein the storage device is associated with a first device pool, and wherein the plurality of instructions further cause the compute device to:

15. The one or more machine-readable storage media of claim 11, wherein a first drive writes per day value for the first device pool exceeds a second drive writes per day value for the second device pool.

16. The one or more machine-readable storage media of claim 11, wherein the plurality of instructions further cause the compute device to:

17. The one or more machine-readable storage media of claim 11, wherein the plurality of instructions further cause the compute device to:

18. The one or more machine-readable storage media of claim 11, wherein the plurality of instructions further cause the compute device to:

transmit the self-test to the storage device.

19. A method to manage storage device performance, the method comprising:

send the adjustment message to the storage device.

20. The method of claim 19, wherein generating the adjustment message comprises generating an adjustment message that instructs the storage device to adjust a capacity provisioning ratio for the storage device, wherein the capacity provisioning ratio defines a ratio of usable storage space to provisionable space for the storage device.

21. The method of claim 19, further comprising:

receiving, by the compute device, performance specification data from the storage device on a first startup of the storage device, wherein the performance specification data includes a performance target for the storage device;

generating, by the compute device, the capacity provisioning ratio for the storage device as a function of at least one of an anticipated device pool allocation for the storage device and the performance target; and

sending, by the compute device, the capacity provisioning ratio to the storage device to cause the storage device to apportion a usable storage space and a provisionable space within a storage unit of the storage device as a function of the capacity provisioning ratio.

22. The method of claim 19, wherein the storage device is associated with a first device pool, the method further comprising:

determining, by the compute device and in response to receipt of the storage device self-test data, whether the storage device fails to satisfy a performance threshold for the first device pool;

determining, by the compute device, whether the storage device satisfies a performance threshold for a second device pool, different from the first device pool; and

reallocating, by the compute device and in response to a determination that the storage device satisfies the performance threshold for the second device pool, the storage device to the second device pool.

23. The method of claim 19, wherein generating the adjustment message comprises generating an adjustment message to reallocate the storage device from a write-intensive storage device pool to a read-intensive storage device pool.

24. The method of claim 23, wherein a first drive writes per day value for the first device pool exceeds a second drive writes per day value for the second device pool.

25. The method of claim 19, further comprising:

querying, by the compute device, a compute sled connected to the storage device for a performance target of the storage device, wherein the performance target is associated with a workload to be executed by the compute sled;

determining, by the compute device, whether the storage device fails to satisfy the performance target; and

generating, by the compute device and in response to a determination that the storage device fails to satisfy the performance target, a second adjustment message for the storage device, wherein the second adjustment message instructs the storage device to satisfy the performance target associated with the compute sled.