WO2017112265A1 - Commande de gestion d'énergie de serveurs distants - Google Patents

Commande de gestion d'énergie de serveurs distants Download PDF

Info

Publication number
WO2017112265A1
WO2017112265A1 PCT/US2016/063377 US2016063377W WO2017112265A1 WO 2017112265 A1 WO2017112265 A1 WO 2017112265A1 US 2016063377 W US2016063377 W US 2016063377W WO 2017112265 A1 WO2017112265 A1 WO 2017112265A1
Authority
WO
WIPO (PCT)
Prior art keywords
server
cloud
servers
available
power management
Prior art date
Application number
PCT/US2016/063377
Other languages
English (en)
Inventor
Enrique G. Castro-Leon
Original Assignee
Intel Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US14/757,865 external-priority patent/US10067547B2/en
Application filed by Intel Corporation filed Critical Intel Corporation
Priority to DE112016006029.7T priority Critical patent/DE112016006029T5/de
Publication of WO2017112265A1 publication Critical patent/WO2017112265A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26—Power supply means, e.g. regulation thereof
    • G06F1/32—Means for saving power
    • G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
    • G06F1/3206—Monitoring of events, devices or parameters that trigger a change in power modality
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5083—Techniques for rebalancing the load in a distributed system
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5094—Allocation of resources, e.g. of the central processing unit [CPU] where the allocation takes into account power or heat criteria
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • This application relates to the technical field of data processing, more specifically to methods and apparatuses associated with remote control of power management of remote computing servers.
  • IT applications may encompass hundreds of servers in data centers with an annual electricity bill in the order of hundreds of thousands of dollars. Because it is difficult to know server power consumption in advance, servers at data centers are usually deployed with an over-commitment of power, and customers of the data centers often pay the power bill of the over-commitment.
  • the Uptime Institute calculates that the starting data center capital cost for each installed kilowatt is about $10 thousand dollars, and without special measures, that's just enough to power two servers.
  • Figure 1 illustrates an arrangement configured to enable a remote server, that is a customer of a remote computing service, to power manage remote servers that may provide the remote computing service;
  • Figure 2 illustrates a method of the arrangement of Figure 1
  • Figure 3 illustrates an example computer-readable storage medium having instructions configured to practice all or selected aspects of the method of Figure 2;
  • Figure 4 illustrates an alternate arrangement configured to enable a customer of a remote computing service, to power manage remote servers that may provide the remote computing service;
  • Figure 5 illustrates still another alternate arrangement configured to enable a customer of a remote computing service, to power manage remote servers that may provide the remote computing service, all arranged in accordance with embodiments of the present disclosure.
  • an apparatus to control power consumption of computer hardware may comprise a datacenter management control module to receive a request to increase capacity of a server pool having one or more computer servers from a computing device; determine an available server with available capacity to meet or contribute to meeting the requested capacity, and includes a power management controller to collect power consumption data for one or more of a power supply, a memory, or a processor of the available server, wherein the available server may be a remote cloud server in a different subnet than the computer servers of the server pool and the computing device; add or facilitate addition of the available server to the server pool; and transmit power management commands to the server added to the server pool to at least partially control power consumption of the server provided with the power management commands.
  • Figure 1 illustrates an arrangement 100 configured to enable a customer of a remote computing service to extend power management policies of the customer directly to servers that may be providing the remote computing service, i.e., across remote computing service boundaries, in accordance with various embodiments.
  • Remote computing with accompanying support servers, often referred to by those skilled in the art as cloud computing, cloud servers, and/or variants of the like.
  • concept of remote computing and other variants may be referred to as cloud computing, cloud servers, and/or the like in the description to follow. These terms are meant to be synonymous, unless the context clearly indicates otherwise.
  • arrangement 100 may include a cloud customer data center 102 and a cloud data center 104 coupled with each other via network(s) 106.
  • Cloud customer data center 102 may be a recipient of cloud services, such as a company who subscribes to cloud services from a cloud services provider.
  • Cloud customer data center 102 may include a cloud customer server 108 that may be a local server configured to manage various requests for data and/or other services from clients or network users.
  • Cloud customer server 108 may be a server configured to manage other servers 110 organized as one or more server pools, to provide one or more applications to users of cloud customer data center 102.
  • Cloud customer server 108 in combination with other servers 110, may collect, store, and provide data and/or data services, upon request, to users of cloud customer data center 102.
  • Cloud customer server 108 may be configured with data center management (DCM) control module 115 to define, determine, and implement one or more power management policies on other servers 110 forming the one or more server pools.
  • DCM control module 115 may be configured to monitor the working capacity of each of other servers 110 of the one or more server pools and, in response to receiving feedback regarding the capacity of other servers 110, DCM control module 115 may provide commands to other servers 110 to at least partially control power consumption of one or more other servers 110, e.g., to cause one or more other servers 110 to enter a partial or full low-power mode, to power down, to fully or partially hibernate, or the like.
  • DCM control module 115 may also be configured to contact cloud management server 120 of cloud data center 104, query cloud management server 120 e.g., a database of cloud management server 120, regarding availability of cloud servers 122, including their capacity and power management capability, to be added to the one or more server pools in whole or in part (via virtual machines operating on the cloud servers 122) to join other servers 110 in servicing users of cloud customer data center 102.
  • DCM control module 115 may selectively add one or more cloud servers 122 in whole or in part to the one or more server pools to join other servers 110 in servicing users of cloud customer data center 102 (operating user applications).
  • cloud customer server 108 and other servers 110 may share a subnetwork (subnet) and may be connected via a network connection 111.
  • Cloud customer server 108 may include a processor 112, storage 114, peripheral interface 116, and a network interface 118.
  • DCM control module 115 may be located in storage 114 and operated by processor 112.
  • Processor 112 may be any one of a number of single or multi-core processors known in the art, e.g., those available from Intel® Corporation of Santa Clara, CA.
  • Storage 114 may be volatile memory, non-volatile memory, or include both volatile and non-volatile memory.
  • storage 114 may include random access memory (RAM), flash memory, optical disk memory, magnetic field based memory, hard disk drives, or the like.
  • RAM random access memory
  • storage 114 may be configured to store a plurality of instructions which, in response to execution by processor 112, may enable cloud customer server 108 to detect, determine, request, and/or adjust one or more power management policies.
  • Peripheral interface 116 may include, but is not limited to, universal serial bus (USB) ports, PS/2 ports, interfaces to interact with display devices, wireless devices, and or other types of potential user and/or electronics interfaces.
  • USB universal serial bus
  • Network interface 118 may be configured to selectively connect cloud customer server 108 to other servers 110 and to network(s) 106.
  • Network interface 118 may include, but is not limited to, a wireless local area network interface.
  • network interface 118 may include a wireless personal network interface.
  • An example of a wireless personal network interface may be a Bluetooth® interface.
  • network interface 118 may include a wireless wide area network.
  • An example of a wireless wide area network interface may be a 3G or 4G telecommunication interface. (3G and 4G refer to the 3 rd and 4 th Generation of Mobile Telecommunication Standards as defined by International Telecommunication Union.)
  • Other servers 110 may be any one of a number of servers known in the art. In particular, one or more of other servers 110 may support the Intelligent Platform
  • IPMI Internet Management Interface
  • DCM control module 115 DCM Management Interface
  • DMTF Distributed Management Task Force
  • Redfish API is a RESTful API for the management of scale-out commodity servers, as well as older platforms.
  • the Redfish API deals with resources which are expressed based on Open Data Protocol (OData) or JavaScript Object Notation schema, with one or more representational state transfer constraints.
  • OData Open Data Protocol
  • JavaScript Object Notation schema JavaScript Object Notation schema
  • Resources may be accessed through the use of Hypertext Transfer Protocol (HTTP) operations: GET, PUT, POST, etc., or a set of Actions that go beyond what create, read, update and delete (CRUD) HTTP operations can perform.
  • HTTP Hypertext Transfer Protocol
  • An example of such an action is performing a system reset.
  • API clients can use the schema to discover the semantics of the resource properties.
  • the Redfish API specification provides for three main category of objects:
  • BMC Baseboard Management Controller
  • DCM control module 115 may be configured to communicate via IPMI or Redfish API with other servers 110.
  • cloud data center 104 may be a data center arranged to store and/or operate a plurality of servers configured to provide cloud services to customers, such as to cloud customer data center 102.
  • Cloud data center 104 may include cloud management server 120, cloud servers 122, and a variety of facility services to provide power, heating, and cooling services to the cloud management server 120 and cloud servers 122.
  • a customer of the services provided by cloud data center 104 may be charged for services based in part on the amount of power consumed by each cloud server 122, based on the amount of power used to climate control cloud data center 104, as well as based on the number of cloud servers 122 for which technical support is provided through the cloud data center 104.
  • Cloud data centers may, for various security reasons, block customers and/or other users from accessing certain types of information from cloud servers 122 and may block computer systems that are remote to cloud data center 104 from manipulating cloud servers 122.
  • cloud data center 104 may operate cloud servers 122 so that cloud servers 122 provide as much bandwidth as cloud servers 122 may provide, without concerns for energy efficient usage or operation of cloud servers 122.
  • clients of services provided by cloud data center 104 may be paying for energy costs that may be may be decreased by selective modification to the power policies of cloud servers 122.
  • Cloud management server 120 may be configured with server control module 125 to query cloud servers 122 to determine whether each of cloud servers 122 provide or may provide power management services.
  • cloud management server 120 may be a remote server configured to provide support for or to provide one or more remote computing services, e.g., cloud services.
  • Server control module 125 may be configured to query each of cloud servers 122 and compile a list or a record based on the responses from cloud servers 122.
  • the record created by server control module 125 may include a variety of data. For example, the record may include a list of ones of the cloud servers 122 enabled to provide power management services or capabilities.
  • the record may also include handle identifiers, such as Internet protocol (IP) addresses, and may provide application programming interface (API) calls which may be used to manipulate power management policies of those of cloud servers 122 that are enabled to provide power management services.
  • Service metadata may be information about the capabilities of a service.
  • the record may provide information about the power management services of the cloud servers 122 and may therefore be a service metadata record, or a metadata record.
  • Cloud management server 120 may include a processor or processor core(s) 124, storage 126, peripheral interface 128, and/or network interface 130.
  • Server control module 125 may be located in storage 126 and operated by processor or processor core(s) 124.
  • Processor 124 may be configured to operate server control module 125 to retrieve information from cloud servers 122 and compile/create a record based on the retrieved information by executing one or more instructions stored on storage 126. Similar to processor 1 12, processor 124 may be any one of a number of single or multi-core processors.
  • Storage 126 may be volatile memory, non-volatile memory, and/or a combination of volatile memory and non-volatile memory. Storage 126 may store a plurality of instructions which, when executed, may cause processor 124 to gather information from cloud servers 122 and communicate the gathered information to cloud customer server 108 through network(s) 106. The instructions will be discussed further in connection with Figure 2, according to various embodiments.
  • Peripheral interface 128 may enable a variety of user interfaces, such as mice, keyboards, monitors, and/or audio commands.
  • peripheral interface 128 may enable USB ports, PS/2 ports, Firewire® ports, Bluetooth®, and a like, according to various embodiments.
  • Network interface 130 may be coupled to cloud servers 122 and to network(s) 106 through a network connection 131.
  • Network interface 130 may be a wireless local area network interface, such as a WiFi® interface in compliance with one of the IEEE 802.11 standards.
  • network interface 130 may include, but is not limited to, a wireless personal network interface, such as a Bluetooth® interface.
  • Network interface 130 may include a wireless wide area network interface, such as 3 G or 4G telecommunication interface.
  • cloud management server 120 may be configured to support either IP MI or Redfish API for communication with cloud customer server 108, DCM control module 115 in particular.
  • Cloud servers 122 may be configured to provide a variety of cloud-based services. According to embodiments, from the perspective of cloud customer server 108, cloud servers 122 may be one or more remote servers configured to be added to one or more of the server pools formed with other servers 110 to support or provide one or more remote computing services, e.g., cloud-based services, to users of cloud customer data center 102 (their applications). Cloud servers 122 may provide cloud-based services to users outside of cloud data center 104 based on instructions, commands, and or security policies received from cloud management server 120, according to embodiments of the disclosure. According to embodiments, cloud servers 122 may represent servers associated with a broad range of search engine, social networking and/or e-commerce websites.
  • cloud servers 122 may be configured with power management (PM) control module 137 to receive power management commands from cloud customer server 108 to reduce overall power consumption of cloud servers 122 in order to reduce operating costs.
  • cloud servers 122 may be shared, providing cloud services to a plurality of cloud customer data centers 102. That is, the cloud services may be provided through applications operating on virtual machines operating on cloud servers 122.
  • each of one or more cloud servers 122 may be entirely dedicated to the provision of cloud services to a single cloud customer data center 102, e.g., being added to a server pool that includes all or a subset of the servers 1 10.
  • Cloud servers 122 may include a network interface 132.
  • Network interface 132 may be similar to network interface 130 and network interface 118.
  • Network interface 132 may enable cloud servers 122 to receive instructions and/or commands from cloud management server 120 and may enable cloud servers 122, in particular, PM control module 137, to provide power management-based information or capabilities to cloud management server 120.
  • Network interface 132 may also enable cloud servers 122 to respond to queries and or commands received from cloud customer server 108.
  • Cloud servers 122 may also include storage 138, processor 136, power supply 134, and a power management controller 140.
  • Power management controller 140 may be configured to monitor power consumption of one or more of storage 138, processor 136, and/or power supply 134. Power management controller 140 may determine power consumption based on voltage sensors, current sensors, temperature sensors, and/or other sensors of the like, located proximate to each of storage 138, processor 136, and/or power supply 134. Power management controller 140 may be configured to be responsive to instructions received from PM control module 137, responding to instructions received through network interface 132. For example, power management controller 140 in cooperation with PM control module 137 may provide network interface 132 with data that may be indicative of power management services available through power management controller 140 (for transmission to DCM control module 115).
  • power management controller 140 may be enabled to cause one or more of cloud servers 122 to enter a low-power mode, to hibernate, and/or to power down. According to embodiments, even while cloud server 122 is powered down, external computing devices such as cloud management server 120 and/or cloud customer server 108 may be enabled to access power management controller 140 through network interface 132.
  • Storage 138 may include, but is not limited to, non-volatile memory and/or volatile memory. Storage 138 may include a variety of instructions which, when executed by processor 136, may cause and/or enable processor 136 to provide cloud services to one or more users of network(s) 106.
  • Power supply 134 may provide power to each of the variety of components included in each of cloud servers 122.
  • Network(s) 106 are intended to represent a broad range of wired or wireless, local or wide area networks, private or public, including e.g., the Internet.
  • cloud servers 122 may be configured to support either IPMI or Redfish API for communication with cloud customer server 108, DCM control module 1 15 in particular.
  • FIG. 2 illustrates a method of the arrangement of Figure 1, in accordance with various embodiments.
  • method 200 may begin at block 202.
  • cloud customer server 108 e.g., DCM control module 1 15, may apply a power management policy to a number of other servers 1 10 that may be operated in the same data center as customer server 108.
  • cloud customer server 108 may provide cost savings to the data center by increasing loads on some of servers 1 10 in order to power down others of servers 1 10 for a net power savings.
  • cloud customer server 108 may perform discovery on a cloud service provider by querying cloud management server 120, e.g., via IPMI or Redfish API, for availability of cloud servers 122 and a record of their power management services.
  • cloud management server 120 may query cloud servers 122 to determine which, if any, of cloud servers 122 may be available and/or enabled to provide power management services. For example, cloud management server 120 may query each of cloud servers 122 to determine which, if any, has available capacity and whether the server includes a power management controller.
  • each of cloud servers 122 may execute a diagnostic check or run a self-query to determine their current capacity to take on additional workload, and which power management services are available.
  • each of cloud servers 122 may reply to the query from cloud management server 120 with data or a list indicating their available capacity and which power management services the cloud server 122 may provide.
  • cloud management server 120 may create a record that includes the availability of cloud server 122 and power management services that are available from each of cloud servers 122.
  • Cloud management server 120 may include additional information in the record.
  • cloud management server 120 may include IP addresses associated with each of cloud servers 122. More particularly, the IP addresses may provide access to power management controllers that may be included in the cloud servers 122.
  • Cloud management server 120 may create the record to include other information, such as passwords, other handles to access cloud servers 122, and/or API calls to enable cloud customer server 108 to manipulate power management services of cloud servers 122.
  • cloud management server 120 may provide the record to cloud customer server 108, e.g., in the form meta data, via IPMI or Redfish API, in response to the query of block 204.
  • Cloud management server 120 may encrypt the record prior to transmission to cloud customer server 108, for example, by using public key infrastructure (PKI) certificates.
  • PKI public key infrastructure
  • cloud management server 120 may create a message authentication code (MAC), based on a hash function, to be transmitted with the record to enable cloud customer server 108 to authenticate that the record has not been altered during transmission through network 106.
  • MAC message authentication code
  • cloud customer server 108 may parse the record to discover which, if any, cloud servers 122 can provide the needed capacity, and power management services are available on cloud servers 122, suitable to host the cloud service that the customer may be subscribed to. From the record, cloud customer server 108 may also determine or retrieve handle identifiers, e.g., IP addresses, of the power management enabled cloud servers 122. Before parsing, cloud customer server 108 may authenticate at least part of the information included in the record.
  • handle identifiers e.g., IP addresses
  • cloud customer server 108 may query cloud servers 122, e.g., via IPMI or Redfish API, to determine power consumption status of components, such as processors, memory, and power supplies.
  • Cloud customer server 108 may initiate the query based on handle identifiers, IP addresses, and/or other information included in the record.
  • power management controllers and/or processors of cloud servers 122 may reply, via IPMI or Redfish API, to the query with power consumption data or status.
  • Power consumption data may include indication of the load of each of the processors of cloud servers 122 to enable cloud customer server 108 to determine if one or more cloud servers 122 may be powered down without loss of quality of service.
  • cloud customer server 108 e.g. DCM control module 115
  • DCM control module 1 15 may transmit, e.g., via IPMI or Redfish API, power management commands to cause cloud servers 122 to hibernate, enter standby, enter another low-power mode, power down, or issue power limitation commands.
  • cloud customer server 108 may also reconfigure one or more cloud servers 122 to include cloud servers 122 within the same subnetwork (subnet) as servers 110.
  • cloud customer server 108 may include the outsourced servers as part of a single trusted pool and may extend the enterprise perimeter to increase the size of the trusted compute base of cloud customer server 108.
  • cloud customer server 108 e.g., DCM control module 1 15, may be configured to query, e.g., via IPMI or Redfish API, cloud management server 120, and cloud management server 120 may directly query cloud servers 122.
  • cloud customer server 108 may transmit, e.g., via IPMI or Redfish API, power management commands to cloud servers 122 through cloud management server 120.
  • cloud servers 122 may be configured to reply directly to cloud customer server 108 via IPMI or Redfish API, and through network(s) 106, or cloud servers 122 may be configured to reply to cloud customer server 108 through cloud management server 120.
  • cloud customer server 108 may be enabled to further manipulate cloud servers 122. For example, by using information from the record, cloud customer server 108 may transplant or install code onto cloud servers 122. According to various embodiments, the code may enable cloud customer server 108, cloud management server 120, and/or other computing devices to determine which power management services are available from cloud servers 122.
  • transplanted or installed code may be encrypted or be subj ect to an attestation scheme to ensure that the code has not been tampered with in transit.
  • cloud customer server 108 may receive power management policies from an administrative management computing device that is remote to both cloud customer data center 102 and cloud data center 104.
  • cloud servers 122 may provide power management service information at lower levels of the open systems interconnect (OSI) stack, e.g., network or physical layers, to enable searchability of the service, various programs of cloud servers 122 may suppress or hide the availability of power management services for security purposes, according to various embodiments.
  • OSI open systems interconnect
  • FIG. 3 illustrates an example computer-readable storage medium having instructions configured to practice all or selected aspects of the method of Figure 2, in accordance with various embodiments of the present disclosure.
  • computer- readable storage medium 302 may include a number of programming instructions 304.
  • Programming instructions 304 may be configured to enable a computing device to perform the cloud customer server operations, such as the DCM control module operations, the cloud management server operations, such as server control module operations, or the cloud server operations, such as PM control module operations of method 200 earlier described with references to Figure 2.
  • programming instructions 304 may include instructions to implement the DCM control module 115.
  • the structure of the DCM control module 115 may include a group of instructions (IG) 312, in response to execution, to apply power management policy, a group of instructions (IG) 314, in response to execution, to query a cloud management server and analyze the answer provided, and a group of instructions (IG) 316, in response to execution, to query and manage a cloud server.
  • Programming instructions 304 may further include instructions to implement the server control module 125.
  • the structure of the server control module 125 may include a group of instructions (IG) 322, in response to execution, to respond to the DCM control module 115, and a group of instructions (IG) 324, in response to execution, to query and manage a cloud server.
  • Programming instructions 304 may also include instructions to implement the PM control module 137.
  • the structure of the server control module 125 may include a group of instructions (IG) 332, in response to execution, to respond to a cloud management server, and a group of instructions (IG) 334, in response to execution, to respond to the DCM control module 115.
  • IG group of instructions
  • programming instructions 304 may be disposed on multiple computer-readable storage media 302 instead.
  • computer-readable storage medium 302 may be a non-transitory storage medium.
  • computer-readable storage medium 302 may be transitory storage medium, such as signals.
  • FIG. 4 illustrates an alternate arrangement configured to enable a customer of a remote computing service, to power manage remote servers that may provide the remote computing service, in accordance with various embodiments of the present disclosure.
  • the arrangement 400 may include cloud customer data center 102, cloud data center 104 and a remote server 402, coupled to each other via one or more networks 106.
  • Cloud customer data center 102 as illustrated and described with reference to Figure 1, may include cloud customer server 108 and other servers 110 configured and perform operations as earlier described, with the exception of DCM control module 1 15.
  • Cloud data center 104 may include cloud management server 120 and cloud servers 122 configured and perform operations as earlier described.
  • cloud servers 122 may be added in whole or in part (its virtual machines), by DCM control module 1 15, to a server pool having servers 1 10 of cloud customer data center 102.
  • DCM control module 115 may be disposed in processor and memory arrangement 404 of remote server 402 in a subnet that is different from the subnets of cloud customer data center 102 and the cloud data center 104, and provide the earlier described data center management operations, such as adding remote cloud servers in cloud data center 104 to one or more server pools of cloud customer data center 102 from remote server 402.
  • DCM control module 115 may query a database associated with a cloud management server 120 of cloud data center 104 for cloud server available capacity and power management capability information, and transmit power control commands to cloud servers 122 of cloud data center 104, out-of-band, using the Redfish communication protocol.
  • FIG. 5 illustrates still another alternate arrangement configured to enable a customer of a remote computing service, to power manage remote servers that may provide the remote computing service, in accordance with various embodiments of the present disclosure.
  • the arrangement 500 may include cloud customer data center 102, a plurality of cloud data centers 104 (e.g., provided by the same or different cloud data center service providers), an aggregation service provider 506, and a remote server 502, coupled to each other via one or more networks 106.
  • Cloud customer data center 102 as illustrated and described with reference to Figure 1 , may include cloud customer server 108 and other servers 1 10 configured and perform operations as earlier described, with the exception of DCM control module 1 15.
  • Each of cloud data centers 104 may respectively include cloud management server 120 and cloud servers 122 configured and perform operations as earlier described.
  • cloud servers 122 of different cloud data center 104 may be added in whole or in part (their virtual machines) to a server pool having servers 110.
  • cloud data centers 104 are abstracted by aggregation service provider 506. That is, aggregation service provider 506 is configured to combine the cloud services provided by cloud data centers 104, as if the combined cloud services are provided by the aggregation service provider 506.
  • DCM control module 1 15 may be disposed in processor and memory arrangement 504 of remote server 502 in a subnet that is different from the subnets of cloud customer data center 102 and the cloud data center 104, and provide the earlier described data center management operations, such as adding remote cloud servers in cloud data center 104 to one or more server pools of cloud customer center 104 from remote server 502.
  • DCM control module 115 would communicate with cloud management servers 120 of cloud data centers 104 via the aggregation service provider 506. That is, the aggregation service provider 506 would appear as a "cloud data center 104" to both cloud customer data center 102 and remote server 402.
  • Aggregation service provider 506 would play the role of cloud management server 120 in Figure 2 (abstracting for all cloud managements 120 of cloud data centers 104 being aggregated). In embodiments, aggregation service provider 506 would pre- query cloud management servers 120 of the cloud data centers 104 being aggregated for the available capacity and power management capability of cloud servers of the cloud data centers 104. Aggregation service provider 506 may aggregate the responsex, and save into a database it maintains. Accordingly, DCM control module 1 15 may query the database associated with aggregation service provider 506 for cloud server capacity and power management capability information. Similarly, DCM control module 115 may transmit power control commands to cloud servers 122 of cloud data centers 104, out-of-band, using the Redfish communication protocol.
  • Various embodiments may include at least one computer readable medium having a number of instructions stored therein and configured to enable a local server, in response to execution by one or more processors of the local server, to query a remote computing service having one or more remote servers and receive a record from the remote computing service in response to the query.
  • the record may include information related to power management services available from the one or more remote servers.
  • the instructions may enable the one or more processors to transmit power management commands, based on the information, to the remote computing service to at least partially control power consumption by the one or more remote servers.
  • the information related to the record may include handle identifiers of the power management services or application programming interface (API) calls for the power management services.
  • API application programming interface
  • the instructions may be further configured to enable the local server, in response to execution by the one or more processors of the local server, to transmit the power management commands, using the API calls for the power management services and the handle identifiers, to reduce central processing unit (CPU) power consumption of the one or more remote servers.
  • CPU central processing unit
  • the instructions may be further configured to enable the local server, in response to execution by the one or more processors of the local server, to transmit the power management commands, using the API calls for the power management services, to the remote computing service to receive power consumption data of the one or more remote servers.
  • the power consumption data may indicate a quantity of power consumed by each of the one or more remote servers.
  • the instructions may be further configured to enable the local server, in response to execution by the one or more processors of the local server, to transmit the power management commands to the remote computing service to cause selected ones of the one or more remote servers to enter into at least one of a low-power mode or a power-down mode, based on the power consumption data of the one or more remote servers.
  • the power consumption data may include power supply data, memory data, and processor data.
  • the instructions may be further configured to enable the local server, in response to execution by the one or more processors of the local server, to determine from the record which ones of the one or more remote servers are enabled to be remotely power managed; and configure the remote power manage enabled ones of the one or more remote servers to a subnet shared by the local server to enable power management of the remote power manage enabled ones of the one or more remote servers by the local server.
  • the instructions may be further configured to enable the local server, in response to execution by the one or more processors of the local server, to authenticate at least part of the information included in the record.
  • a method may include querying a remote computing service, by a customer server, having one or more remote servers.
  • the method may include receiving a record, by the customer server, from the remote computing service in response to the query.
  • the record may include information related to power management services available from the one or more remote servers.
  • the method may include transmitting power management commands, based on the information, to the remote computing service to at least partially control power consumption by the one or more remote servers.
  • a method may include receiving a request, by a remote computing management server from a customer server, for a record of power management services available by a number of remote computing servers that are networked to provide remote computing services and that are managed by the remote computing management server.
  • the method may include determining, by the remote computing management server, ones of the number of remote computing servers that provide the power management services and creating, by the remote computing management server, the record of the power management services based on said determining.
  • the method may include transmitting the record of the power management services to the customer server, in response the request for the record, to enable the customer server to access the power management services provided by the ones of the number of remote computing servers.
  • the record may include internet protocol (IP) addresses of power management controllers of each of the ones of the one or more remote computing servers that provide the power management services.
  • the record may include power management service application programming interface (API) calls to enable manipulation of the power management controllers.
  • IP internet protocol
  • API power management service application programming interface
  • the method may also include receiving, by the remote computing management server, instructions from the customer server to transmit data related to power consumed by each of the ones of the one or more remote computing servers that provide the power management services; and transmitting, to the customer server, the data related to the power consumed by each of the one or more remote computing servers.
  • the method may further include, in response to instructions from the customer server, transferring tasks from first partially loaded ones of the one or more remote computing servers that provide the power management services to second partially loaded ones of the one or more remote computing servers that provide power management service; and causing the first partially loaded ones of the one or more remote computing servers to power-down or to enter a low-power mode.
  • the method may further include receiving, by the one or more remote computing servers, one or more programs from the customer server; installing the one or more programs onto the one or more remote computing servers; and enabling control of the one or more remote computing servers by the customer server through the one or more programs.
  • the method may further include generating a message authentication code from the record; and transmitting the message authentication code with the record to the remote server, in response to the request, to enable the remote server to verify the authenticity of the record.
  • one or more computer readable media may include a number of instructions which, if executed by a processor of a customer server, may enable a customer server to operate according to the above described method.
  • a server may include a network interface configured to communicate with a remote computing service through a remote computing management server; memory coupled to the network interface and configured to store a number of instructions; and a processor coupled to the memory and the network interface.
  • the processor may be configured to execute the number of instructions to send a request, via the network interface to the remote computing management server, to receive a record of remote computing servers that are managed by the remote computing management server to provide remote computing service for the server and that include power management controllers.
  • the record may include handles that identify respective ones of the power management controllers.
  • the processor may be further configured to execute the number of instructions to monitor power consumption of the remote computing servers and to selectively cause some of the remote computing servers, by using the handles, to enter a low-power mode based on the monitored power consumption.
  • the processor may be configured to execute the number of instructions to authenticate that the record is unchanged upon receipt by the server.
  • a remote computing server may include a power supply; memory coupled to the power supply to receive power and to store a number of instructions; and one or more processors coupled to the power supply to receive power and coupled to the memory to execute the number of instructions stored by the memory.
  • the remote computing server may include a number of power consumption meters configured to measure power consumption of the power supply the memory, and the one or more processors.
  • the remote computing server may include a power management controller configured to collect data corresponding to the measured power consumption and configured to reduce the power consumption of the one or more processors in response to instructions received by a customer server subscribed to receive remote computing services from the remote computing server.
  • the instructions may be based on power management application programming interface (API) calls made with a handle for the remote computing server.
  • API application programming interface
  • the handle may be managed by a remote computing management server.
  • the remote computing server may include a network interface configured to transfer data between the processor, the power management controller, and a network.
  • the power management controller may be configured to be responsive to the instructions while the remote computing server is powered down or in a low-power mode.
  • the power management controller may be configured to provide a list of power management services available by the remote computing server in response to a power management service query from the remote computing management server to enable the remote computing management server to create a record that includes the list of power management services.
  • a remote computing management server may include a network interface configured to communicatively couple the remote computing management server to one or more remote computing servers; and memory configured to store a number of instructions for transferring information to and from the one or more remote computing servers.
  • the remote computing management server may include one or more processors configured to execute the number of instructions to support or provide remote computing services.
  • the one or more processors may be configured to execute the number of instructions to create a record that includes power management services provided by the one or more remote computing servers, in response to a query from a customer server that receives the remote computing services.
  • the one or more processors may be configured to create the record to include one or more network identifiers of the one or more remote computing servers that provide power management services.
  • the one or more processors may be configured to execute the number of instructions to generate a security code based on the record and to transmit the security code with the record to the customer server to enable the customer server to authenticate the record.
  • At least one computer readable medium may have a number of instructions stored therein and configured to enable a remote computing server, in response to execution by one or more processors of the remote computing server, to provide remote computing services to one or more customer servers and to determine whether the remote computing server is configured to provide remote power management services.
  • the number of instructions may enable the remote computer server to reply to a query from a managing server that is configured manage the remote computer server and other remote computing server. Replying to the query may include indicating which, if any, power management services the remote computing server is configured to provide.
  • the number of instructions may enable the remote computer server to, in response to power management commands received from at least one of the one or more customer servers, decrease power consumption by the remote computing server.
  • the instructions may be further configured to enable the one or more processors of the remote computing server to monitor a power consumption of a power supply of the remote computer server, monitor a power consumption of the one or more processors, and transmit the power consumption of the power supply and the power consumption of the one or more processors to the at least one of the one or more customer servers.
  • decreasing power consumption may include causing the remote computing server to enter a shut down sequence.
  • a method may include providing remote computing services to one or more customer servers; and determining whether the remote computing server is configured to provide remote power management services.
  • the method may include replying to a query from a managing server that is configured manage the remote computer server and other remote computing server, wherein replying to the query includes indicating which, if any, power management services the remote computing server is configured to provide; and, in response to power management commands received from at least one of the one or more customer servers, decrease power consumption by the remote computing server.
  • Example 1 may be an apparatus to control power consumption of computer hardware, comprising: one or more processors; and a datacenter management control module, to be operated on the one or more processors, to: receive a request to increase capacity of a server pool having one or more computer servers from a computing device associated with the server pool;
  • the available server determines an available server with available capacity to meet or contribute to meeting the requested capacity, and includes a power management controller to collect power consumption data for one or more of a power supply, a memory, or a processor of the available server, wherein the available server may be a remote cloud server in a different subnet than the computer servers of the server pool and the computing device; add or facilitate addition of the available server to the server pool; and transmit power management commands to the server added to the server pool to at least partially control power consumption of the server provided with the power management commands.
  • Example 2 may be example 1, where the apparatus may be in the same subnet as the computer servers of the server pool and the computing device.
  • Example 3 may be example 1, where the apparatus may be in a different subnet as the computer servers of the server pool and the computing device.
  • Example 4 may be example 3, wherein the datacenter management control module may transmit the power management commands to the server added to the server pool via an out-of-band communication channel.
  • Example 5 may be example 3, wherein the datacenter management control module may transmit the power management commands to the server added to the server pool using a protocol that conforms to one or more representational state transfer constraints.
  • Example 6 may be example 3, wherein the datacenter management control module may: send a query for available servers and their power management capability to a database; and receive service metadata associated with the available server in response to the query.
  • Example 7 may be example 6, wherein the database may be associated with a cloud management server of a cloud data center, and the available server may be associated with the cloud data center.
  • Example 8 may be example 6, wherein the database may be associated with an aggregator service that aggregate cloud computing services provided by a plurality of cloud data centers; and wherein the available server may be associated with one of a plurality cloud data centers.
  • Example 9 may be any one of examples 3-8, wherein the datacenter management control module may also transmit power management commands to one or more computer servers of the server pool to at least partially control power consumption of the one or more computer servers of the server pool.
  • Example 10 may be example 9, wherein the datacenter management control module may transmit the power management commands to the one or more computer servers of the server pool using an intelligent platform management interface.
  • Example 11 may be a method for controlling power consumption of computer hardware, comprising: receiving, by a datacenter infrastructure management controller, a request to increase capacity of a server pool having one or more computer servers from a computing device associated with the server pool; determining, by the datacenter infrastructure management controller, an available server with available capacity to meet or contribute to meeting the requested capacity, and includes a power management controller to collect power consumption data for one or more of a power supply, a memory, or a processor of the available server, wherein the available server may be a remote cloud server in a different subnet than the computing device; adding or facilitating adding, by the datacenter infrastructure management controller, the available server to the server pool; and transmitting, by the datacenter infrastructure management controller, power management commands to the server added to the server pool to at least partially control power consumption of the server provided with the power management commands.
  • Example 12 may be example 11, where the datacenter infrastructure management controller may be disposed on a computing apparatus that may be in the same subnet as the computer servers of the server pool and the computing device.
  • Example 13 may be example 11, where the datacenter infrastructure management controller may be disposed on a computing apparatus that may be in a different subnet as the computer servers of the server pool and the computing device.
  • Example 14 may be example 13, wherein the power management commands are sent to the server added to the server pool via an out- of-band communication channel.
  • Example 15 may be example 13, wherein the power management commands are sent to the server added to the server pool using a protocol that conforms to one or more representational state transfer constraints.
  • Example 16 may be example 13, wherein determining an available server includes: sending, by the datacenter infrastructure management controller, a query for available servers and their power management capability to a database ; and receiving, by the datacenter infrastructure management controller, service metadata associated with the available server.
  • Example 17 may be example 16, wherein the database may be associated with a cloud management server of the cloud data center; and wherein the available server may be associated with the cloud data center.
  • Example 18 may be example 16, wherein the database may be associated with an aggregator service that aggregate cloud computing services provided by a plurality of cloud data centers; and wherein the available server may be associated with one of the plurality cloud data centers.
  • Example 19 may be any one of examples 13-18, further comprising transmitting, by the datacenter infrastructure management controller, power management commands to one or more computer servers of the server pool to at least partially control power consumption of the one or more computer servers of the server pool.
  • Example 20 may be example 19, wherein the datacenter management control module may transmit the power management commands to the one or more computer servers of the server pool using an intelligent platform management interface.
  • Example 21 may be at least one computer-readable medium having a plurality of instructions stored therein, and configured to enable a computing apparatus, in response to execution by one or more processors of the computing apparatus, to: receive a request to increase capacity of a server pool having one or more computer servers from a computing device associated with the server pool; determine an available server with available capacity to contribute to the requested capacity, and includes a power management controller to collect power consumption data for one or more of a power supply, a memory, or a processor of the available server, wherein the available server may be a remote cloud server in a different subnet than the computer servers of the server pool and the computing device; add or facilitate addition of the available server to the server pool; and transmit power management commands to the server added to the server pool to at least partially control power consumption of the server.
  • Example 22 may be example 21, where the computing apparatus may be in the same subnet as the computer servers of the server pool and the computing device.
  • Example 23 may be example 21, where the computing apparatus may be in a different subnet as the computer servers of the server pool and the computing device.
  • Example 24 may be example 23, wherein the instructions are to cause the computing apparatus to transmit the power management commands via an out-of-band communication channel.
  • Example 25 may be example 23, wherein the instructions are to cause the computing apparatus to transmit the power management commands using a protocol that conforms to one or more representational state transfer constraints.
  • Example 26 may be example 23, wherein the instructions are also to cause the computing apparatus to: send a query for available servers and their power management capability to a database ; and receive service metadata associated with the available server in response to the query.
  • Example 27 may be example 26, wherein the database may be associated with a cloud management server of a cloud data center; and wherein the available server may be associated with a cloud data center.
  • Example 28 may be example 26, wherein the database may be associated with an aggregator service that aggregate cloud computing services provided by a plurality of cloud data centers; and wherein the available server may be associated with one of a plurality cloud data centers.
  • Example 29 may be any one of the examples 23-28, wherein the instructions are also to cause the computing apparatus to transmit power management commands to one or more computer servers of the server pool to at least partially control power consumption of the one or more computer servers of the server pool.
  • Example 30 may be example 29, wherein the instructions are also to cause the computing apparatus to transmit the power management commands to the one or more servers of the server pool using an intelligent platform management interface.
  • Example 31 may be an apparatus for controlling power consumption of computer hardware, comprising: means for receiving a request to increase capacity of a server pool having one or more computer servers from a computing device associated with the server pool; means for determining an available server with available capacity to contribute to the requested capacity, and includes a power management controller to collect power consumption data for one or more of a power supply, a memory, or a processor of the available server, wherein the available server may be a remote cloud server in a different subnet than the computing device; means for adding or facilitating adding the available server to the server pool; and means for transmitting power management commands to the server added to the server pool to at least partially control power consumption of the server provided with the power management commands.
  • Example 32 may be example 31, where the apparatus may be in the same subnet as the computer servers of the server pool and the computing device.
  • Example 33 may be example 31, where the apparatus may be in a different subnet as the computer servers of the server pool and the computing device.
  • Example 34 may be example 33, wherein the power management commands are sent to the server added to the server pool via an out- of-band communication channel.
  • Example 35 may be example 33, wherein the power management commands are sent to the server added to the server pool using a protocol that conforms to one or more representational state transfer constraints.
  • Example 36 may be example 35, wherein means for determining an available server includes: means for sending a query for available servers and their power management capability to a database; and means for receiving service metadata associated with the available server.
  • Example 37 may be example 36, wherein the database may be associated with a cloud management server of a cloud data center, and the available server may be associated with the cloud data center.
  • Example 38 may be example 36, wherein the database may be associated with an aggregator service that aggregate cloud computing services provided by a plurality of cloud data centers; and wherein the available server may be associated with one of a plurality cloud data centers.
  • Example 39 may be any one of the examples 33-38, further comprising means for transmitting power management commands to one or more servers of the server pool to at least partially control power consumption of the one or more servers of the server pool.
  • Example 40 may be example 39, wherein the means for transmitting power management commands may be further for transmitting the power management commands to the one or more servers of the server pool using an intelligent platform management interface.

Abstract

L'invention concerne des serveurs, un support de stockage et des procédés associés à la commande de services de gestion d'énergie des serveurs distants d'un service informatique distant. Dans des modes de réalisation, un appareil destiné à commander la consommation d'énergie du matériel informatique peut comprendre un module de commande de gestion de centre de données pour recevoir une demande d'augmentation de la capacité d'un groupe de serveurs d'un dispositif informatique ; déterminer un serveur disponible ayant une capacité disponible qui comprend un contrôleur de gestion d'énergie pour collecter des données de consommation d'énergie pour un ou plusieurs parmi une alimentation électrique, une mémoire ou un processeur du serveur disponible ; ajouter ou faciliter l'ajout du serveur disponible au groupe de serveurs ; et transmettre des instructions de gestion d'énergie au serveur ajouté au groupe de serveurs pour commander au moins partiellement la consommation d'énergie du serveur disposant des instructions de gestion d'énergie. L'invention concerne également d'autres modes de réalisation.
PCT/US2016/063377 2015-12-24 2016-11-22 Commande de gestion d'énergie de serveurs distants WO2017112265A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
DE112016006029.7T DE112016006029T5 (de) 2015-12-24 2016-11-22 Energiemanagementsteuerung von entfernten servern

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US14/757,865 2015-12-24
US14/757,865 US10067547B2 (en) 2012-06-28 2015-12-24 Power management control of remote servers

Publications (1)

Publication Number Publication Date
WO2017112265A1 true WO2017112265A1 (fr) 2017-06-29

Family

ID=59091149

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2016/063377 WO2017112265A1 (fr) 2015-12-24 2016-11-22 Commande de gestion d'énergie de serveurs distants

Country Status (3)

Country Link
DE (1) DE112016006029T5 (fr)
SG (1) SG10201609781SA (fr)
WO (1) WO2017112265A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040054780A1 (en) * 2002-09-16 2004-03-18 Hewlett-Packard Company Dynamic adaptive server provisioning for blade architectures
US20090106571A1 (en) * 2007-10-21 2009-04-23 Anthony Low Systems and Methods to Adaptively Load Balance User Sessions to Reduce Energy Consumption
KR20120038688A (ko) * 2010-10-14 2012-04-24 주식회사 마스터소프트 원격 전원 관리 시스템 및 방법
US20140006815A1 (en) * 2012-06-28 2014-01-02 Enrique G. Castro-Leon Power management control of remote servers
WO2014159977A1 (fr) * 2013-03-14 2014-10-02 Tso Logic Inc. Système de commande pour commande de puissance

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040054780A1 (en) * 2002-09-16 2004-03-18 Hewlett-Packard Company Dynamic adaptive server provisioning for blade architectures
US20090106571A1 (en) * 2007-10-21 2009-04-23 Anthony Low Systems and Methods to Adaptively Load Balance User Sessions to Reduce Energy Consumption
KR20120038688A (ko) * 2010-10-14 2012-04-24 주식회사 마스터소프트 원격 전원 관리 시스템 및 방법
US20140006815A1 (en) * 2012-06-28 2014-01-02 Enrique G. Castro-Leon Power management control of remote servers
WO2014159977A1 (fr) * 2013-03-14 2014-10-02 Tso Logic Inc. Système de commande pour commande de puissance

Also Published As

Publication number Publication date
DE112016006029T5 (de) 2018-09-06
SG10201609781SA (en) 2017-07-28

Similar Documents

Publication Publication Date Title
US10067547B2 (en) Power management control of remote servers
US9454199B2 (en) Power management control of remote servers
US20210037097A1 (en) Cloud data storage location monitoring
US10547710B2 (en) Device gateway
US11122023B2 (en) Device communication environment
US9202225B2 (en) Aggregate monitoring of utilization data for vendor products in cloud networks
US10958648B2 (en) Device communication environment
US10146607B2 (en) Troubleshooting of cloud-based application delivery
Sharma et al. Energy-efficient resource allocation and migration in private cloud data centre
US8954564B2 (en) Cross-cloud vendor mapping service in cloud marketplace
EP3479249B1 (fr) Technologies destinées à gérer des configurations d'application et des justificatifs d'identité associés
US7962635B2 (en) Systems and methods for single session management in load balanced application server clusters
US20180091391A1 (en) Device State Management
US9973593B2 (en) Device gateway
US20100235887A1 (en) System and method for queuing to a cloud via a queuing proxy
WO2015058569A1 (fr) Procédé et appareil de personnalisation de service de sécurité
Kanapram et al. Exploring the trade-off between performance and energy consumption in cloud infrastructures
WO2017112265A1 (fr) Commande de gestion d'énergie de serveurs distants
US20220326994A1 (en) Computing resource sharing system
WO2017004251A1 (fr) Procédé et système permettant une découverte de fonction et de service
Sehgal et al. Foundations of cloud computing and information security
Shopov An M2M solution for smart metering in electrical power systems
US20230353636A1 (en) Signal processing device, gateway, management server and method
US10135695B1 (en) System and method for managing a remote device
US20230353359A1 (en) Secure onboarding of external compute fabric in an edge horizontal platform

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16879750

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 112016006029

Country of ref document: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16879750

Country of ref document: EP

Kind code of ref document: A1