# US 11,310,308 B2 · Method for managing resources of a computer cluster by means of historical data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,310,308 B2 - Document ID: US11310308B2 - Status: Granted - Kind: B2 - Application number: 16/230,239 - Filed: 2018-12-21 - Published: 2019-06-27 - Granted: 2022-04-19 - Earliest claimed: 2017-12-22 - Assignees: Bull SAS - Applicants: Bull SAS - Inventors: Thomas Cadeau; Dineshkumar Rajagopal - Web page: https://patentgrep.com/patent/US11310308B2 ## CPC - G06F 9/5061: {Partitioning or combining of resources} - H04L 67/1008: based on parameters of servers, e.g. available memory or workload (monitoring of computer activity G06F11/30) - H04L 67/1034: Reaction to server failures by a load balancer - Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - 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 ## Abstract A method for managing resources of a computer cluster, wherein automated means allocate to a job at least one resource among several resources from the cluster, the automated means selecting the resource based on at least one historical data relative to previous uses of the resources and/or data relative to the arrangement, temperature, power consumption, bandwidth, or maintenance of the cluster or one or more components thereof, the automated mechanism thus determining the wear of the various resources available and choosing the resources with the lowest wear to perform the job, thereby avoiding the over-utilization of resources, one of the main causes of failures in a computer cluster, and increasing the lifetime of the cluster. ## Claims ### Claim 1 A method for managing resources of a computer cluster, comprising: via a computer, allocating a job at least one resource from the computer cluster among several resources from the computer cluster; wherein the allocating is performed by the computer depending on at least one historical data related to at least one previous use of the several resources, wherein said at least one historical data including at least one of data relative to said job and data relative to an arrangement, temperature, power consumption, bandwidth, a log concerning use of a resource, of the computer cluster or one or more components thereof; at least one data relative to a wear of said at least one resource of the computer cluster, wherein wear data is computed from a maintenance date and the at least one historical data, wherein said maintenance date is a date of last maintenance of the computer or one or more components thereof; a cluster topology of the computer cluster; a predetermined resource management strategy selected by an external user for use by the computer; wherein said at least one resource is allocated to said job when said at least one resource comprises a lowest use rate, determined from the at least one historical data, comprises a lowest wear, determined from the at least one data relative to a wear of said at least one resource of the computer cluster, is most suited regarding the cluster topology, is compatible with the predetermined resource management strategy. ### Claim 2 (depends on 1) The method according to claim 1, wherein the computer generates the at least one historical data during a previous use and records the at least one historical data in a database of historical data. ### Claim 3 (depends on 1) The method according to claim 1, wherein the allocating is further dependent on at least one data relative to a maintenance of the computer cluster. ### Claim 4 (depends on 1) The method according to claim 1, wherein the computer determines at least one action to be taken in order to improve management of the resources of the computer cluster based on one or more of said at least one historical data, said data relative to said job, and said data relative to the arrangement, said temperature, said power consumption, said bandwidth, or said maintenance of the computer cluster or said one or more components thereof. ### Claim 5 (depends on 4) The method according to claim 4, wherein the computer further determines the at least one action by means of at least one data relative to at least one effect of at least one action previously taken. ### Claim 6 (depends on 4) The method according to claim 4, wherein the at least one action to be taken is relative to shutting down a component of the computer cluster, substituting a component of the computer cluster, or sending a message. ### Claim 7 (depends on 1) The method according to claim 1, implemented as part of a high-performance computing method. ### Claim 8 (depends on 1) The method according to claim 1, wherein a sensor situated in the computer obtains said data and transmits the data to the computer, the allocating further performed depending on the data. ### Claim 9 (depends on 1) The method according to claim 1, wherein the predetermined resource management strategy comprises prioritizing said at least one resource over other resources. ### Claim 10 A method for managing resources of a computer cluster, comprising: providing a sensor situated in the computer cluster that obtains data and transmits the data to a computer; via the computer, allocating to a job at least one resource from the computer cluster among several resources from the computer cluster; wherein the allocating is performed by the computer depending on at least one historical data related to at least one previous use of the several resources, said at least one historical data including at least one of data relative to said job and data relative to an arrangement, temperature, power consumption, bandwidth, a log concerning use of a resource, or maintenance of the computer cluster or one or more components thereof; at least one data relative to a wear of said at least one resource of the computer cluster, wherein wear data is computed from a maintenance date and the at least one historical data, wherein said maintenance date is a date of last maintenance of the computer or one or more components thereof; a cluster topology of the computer cluster; the data obtained from the sensor; wherein said at least one resource is allocated to said job when said at least one resource comprises a lowest use rate, determined from the at least one historical data, comprises a lowest wear, determined from the at least one data relative to a wear of said at least one resource of the computer cluster, is most suited regarding the cluster topology, is compatible with a predetermined resource management strategy. ### Claim 11 (depends on 10) The method according to claim 10, wherein the data relates to a temperature of the computer and wherein the at least one resource is further allocated to said job when said at least one resource comprises a lowest temperature. ### Claim 12 A method for managing resources of a computer cluster, comprising: via a computer, allocating to a job at least one resource from the computer cluster among several resources from the computer cluster; the allocating is performed by the computer depending on at least one historical data related to at least one previous use of the several resources, said at least one historical data including at least one of data relative to said job and data relative to an arrangement, temperature, power consumption, bandwidth, a log concerning use of a resource, of the computer cluster or one or more components thereof; at least one data relative to a wear of said at least one resource of the computer cluster, wherein wear data is computed from a maintenance date and the at least one historical data, wherein said maintenance date is a date of last maintenance of the computer or one or more components thereof; a cluster topology of the computer cluster; wherein the computer determines relative ages of at least two components of the computer cluster based on the at least one historical data; wherein said at least one resource is allocated to said job when said at least one resource comprises a lowest use rate, determined from the at least one historical data, comprises a lowest wear, determined from the at least one data relative to a wear of said at least one resource of the computer cluster, is most suited regarding the cluster topology, is compatible with a predetermined resource management strategy, is younger than at least one other resource. ### Claim 13 (depends on 12) The method according to claim 12, wherein the computer determines a conclusion concerning the relative ages of the at least two components and emits a message comprising the conclusion. ### Claim 14 (depends on 13) The method according to claim 13, wherein, the conclusion is relative to one of a workload of the computer cluster, a maintenance date of the computer cluster, the temperature of the computer cluster, the power consumption of the computer cluster, and the bandwidth of the computer cluster. ### Claim 15 A non-transitory computer readable program for managing resources of a computer cluster, comprising: code instructions executing on a computer for allocating to a job at least one resource from the computer cluster among several resources from the computer cluster, depending on at least one historical data related to at least one previous use of the several resources, at least one data relative to a wear of said at least one resource of the computer cluster, wherein wear data is computed from a maintenance date and the at least one historical data, wherein said maintenance date is a date of last maintenance of the computer or one or more components thereof, a cluster topology of the computer cluster, and a predetermined resource management strategy by and external user of use by the computer, wherein the predetermined resource management strategy comprises prioritizing said at least one resource over other resources, wherein said at least one resource is allocated to said job when said at least one resource comprises a lowest use rate, determined from the at least one historical data, comprises a lowest wear, determined from the at least one data relative to a wear of said at least one resource of the computer cluster, is most suited regarding the cluster topology, is compatible with the predetermined resource management strategy. ### Claim 16 (depends on 15) The non-transitory computer readable program of claim 15, wherein the non-transitory computer readable program is available for downloading over a telecommunications network. ### Claim 17 A non-transitory computer readable medium comprising program instructions executed on a computer for managing resources of a computer cluster, the computer being configured to allocate to a job at least one resource among several resources of the computer cluster depending on at least one historical data related to at least one previous use of the several resources, and a cluster topology of the computer cluster as determined by the computer wherein the computer determines relative ages of at least two components of the computer cluster based on a maintenance date and said at least one historical data, said at least one historical data including at least one of data relative to said job and data relative an arrangement, temperature, power consumption bandwidth, of the computer cluster or one or more components thereof, and wherein the maintenance date is a date of last maintenance of the computer cluster or one or more component thereof, wherein said at least one resource is allocated to said job when said at least one resource comprises a lowest use rate, determined from the at least one historical data, comprises a lowest wear, determined from at least one data relative to a wear of said at least one resource of the computer cluster, is most suited regarding the cluster topology, is compatible with a predetermined resource management strategy. ### Claim 18 (depends on 17) The non-transitory computer readable medium according to claim 17, wherein the at least one resource is a node, a server, a bandwidth, a power, a memory such as a ram or a flash memory, or a component such as a processor, connector or a data storage medium. ### Claim 19 A system for managing resources of a computer cluster, the system comprising: a non-transitory computer readable medium comprising program instructions executed on a computer and further comprising said computer cluster, the computer being configured to allocate to a job at least one resource among several resources of the computer cluster depending on at least one historical data related to at least one previous use of the several resources, a cluster topology of the computer cluster as determined by the computer, wherein the computer determines relative ages of at least two components of the computer cluster based on a maintenance date and said at least one historical data, said at least one historical data including at least one of data relative to said job and data relative an arrangement, temperature, power consumption bandwidth, of the computer cluster or one or more components thereof, and wherein the maintenance date is a date of last maintenance of the computer cluster or one or more component thereof, wherein said at least one resource is allocated to said job when said at least one resource comprises a lowest use rate, determined from the at least one historical data, comprises a lowest wear, determined from at least one data relative to a wear of said at least one resource of the computer cluster, is most suited regarding the cluster topology, is compatible with a predetermined resource management strategy. ### Claim 20 (depends on 19) The system according to claim 19 further comprising a database including an electronical support comprising, in registered form, said at least one historical data relative to at least one use of said at least one resource of said computer cluster.