Environmentally aware load-balancing

• US 8,954,984 B2
• Filed: 04/19/2012
• Issued: 02/10/2015
• Est. Priority Date: 04/19/2012
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Adinarayan, Geetha, Bernoth, Andrew J., Kannegala, Sateesh S., Sivakumar, Gandhi
• Primary Examiner(s): Kawsar, Abdullah Al
• Assistant Examiner(s): NAM, DAREAUM A
• Application Number:
  US13/450,658
• Publication Number:
  US 20130283289A1
• Time in Patent Office: 1,027 Days
• Field of Search: None
• US Class Current: 718/105
• CPC Class Codes: G06F 9/5094 where the allocation takes into account power or heat criteria power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A method and associated systems for the environmentally aware load-balancing of components of a multi-component power-consuming system. The environmentally aware load-balancer receives continually updated values from at least two environmental sensors that monitor and report the values of environmental metrics that characterize components of the power-consuming system and the environments within which those components are located. When the load-balancer receives a task request directed to the power-consuming system, the load-balancer selects a balanced workload allocation as a function of the values of the received environmental metrics and communicates that balanced workload allocation to a routing mechanism. The routing mechanism then uses the communicated balanced workload allocation to determine which component or components of the power-consuming system should receive the task request.

Claims

1. A method for allocating one or more task workloads of a plurality of task workloads to one or more components of a plurality of components of a multi-component power-consuming system coupled to a routing mechanism, wherein components of the plurality of components that are of a similar class are organized into an associated subsystem, and wherein a memory stores an environmental data table, a support table, and a task-allocation algorithm,

wherein each of the plurality of components of the multi-component power-consuming system is associated with one or more environmental sensors, and with a plurality of non-environmental logging mechanisms,

wherein the one or more environmental sensors identify one or more environmental characteristics associated with a set of operating environments of its associated component and continually update one or more environmental values associated with the one or more environmental characteristics of its associated component in the environmental data table,

wherein each of the plurality of non-environmental logging mechanism identifies one or more non-environmental characteristics of its associated component and continuously updates one or more non-environmental values associated with the one or more non-environmental characteristics in the support table,

the method comprising:

a processor of a computer system receiving a task request that comprises the one or more task workloads,

the processor, in response to receiving the task request, launching the task-allocation algorithm by selecting a balanced workload allocation of a plurality of balanced workload allocations as a function of one or more values stored in the environmental data table and the support table;

wherein the task-allocation algorithm comprises an optimal predefined set of rules that determines how the routing mechanism should allocate the one or more task workloads among at least two components of the plurality of components, wherein the determining comprises identifying one or more classes of components of the plurality of components, wherein the one or more classes of components are candidate components to receive the one or more task workloads, and wherein components or subsystems comprised by a class of the one or more classes of components are each associated with a same class of the one or more classes,

wherein the task-allocation algorithm computes, for each of the candidate components, an allocation weighting of a plurality of allocation weightings as a function of at least one of the one or more environmental values,

wherein the task-allocation algorithm may alter the selected balanced workload allocation, and

wherein the selected balanced workload allocation comprises an optimal predefined set of rules that determine how the routing mechanism should allocate the one or more task workloads among the at least two components, and

wherein the selecting is performed as a further function of a first weighting of the plurality of task-allocation weightings, wherein the first weighting revises the optimal predefined set of rules, wherein the first weighting is associated with one of the at least two components, and wherein a weighting rule of the optimal predefined set of rules is a rule that instructs the routing mechanism to update the value of the first weighting, and

the processor choosing a first communication time, wherein the choosing is a function of at least one previously received value of a tracked environmental factor of the set of environmental factors and is a further function of at least one value previously received from a logging mechanism of the plurality of non-environmental logging mechanisms; and

the processor communicating the selected balanced workload allocation and the task request at the first communication time to the routing mechanism such that the routing mechanism allocates the one or more task workloads among the at least two components,

wherein the first environmental value of the one or more environmental values comprises an outdoor temperature outside a facility in which a component of the at least two components is operating and wherein a first updated non-environmental value of the continuously updated one or more non-environmental values comprises a local time zone at the facility.

2. The method of claim 1, wherein neither of the at least two components is the computer system and wherein neither of the at least two components is a component of a computer system.

3. The method of claim 1, further comprising providing at least one support service for at least one of creating, integrating, hosting, maintaining, and deploying computer-readable program code in the computer system, wherein the computer-readable program code in combination with the computer system is capable of performing the receiving, launching, choosing, and communicating.

4. A computer program product, comprising a memory having a computer readable program code stored therein, said computer readable program code comprising a method for allocating one or more task workloads of a plurality of task workloads to one or more components of a plurality of components of a multi-component power-consuming system coupled to a routing mechanism, wherein components of the plurality of components that are of a similar class are organized into an associated subsystem, and wherein the memory stores an environmental data table, a support table, and a task-allocation algorithm,

wherein each of the plurality of components of the multi-component power-consuming system is associated with one or more environmental sensors, and with a plurality of non-environmental logging mechanisms,

wherein the one or more environmental sensors identify one or more environmental characteristics associated with a set of operating environments of its associated component and continually update one or more environmental values associated with the one or more environmental characteristics of its associated component in the environmental data table,

wherein each of the plurality of non-environmental logging mechanisms identifies one or more non-environmental characteristics of its associated component and continuously updates one or more non-environmental values associated with the one or more non-environmental characteristics in the support table,

the method comprising:

the processor receiving a task request that comprises the one or more task workloads,

the processor, in response to receiving the task request, launching the task-allocation algorithm by selecting a balanced workload allocation of a plurality of balanced workload allocations as a function of one or more values stored in the environmental data table and the support table;

wherein the task-allocation algorithm comprises an optimal predefined set of rules that determines how the routing mechanism should allocate the one or more task workloads among at least two components of the plurality of components, wherein the determining comprises identifying one or more classes of components of the plurality of components, wherein the one or more classes of components are candidate components to receive the one or more task workloads, and wherein components or subsystems comprised by a class of the one or more classes of components are each associated with a same class of the one or more classes,

wherein the task-allocation algorithm computes, for each of the candidate components, an allocation weighting of a plurality of allocation weightings as a function of at least one of the one or more environmental values,

wherein the task-allocation algorithm may alter the selected balanced workload allocation, and

wherein the selected balanced workload allocation comprises an optimal predefined set of rules that determine how the routing mechanism should allocate the one or more task workloads among the at least two components, and

wherein the selecting is performed as a further function of a first weighting of the plurality of task-allocation weightings, wherein the first weighting revises the optimal predefined set of rules, wherein the first weighting is associated with one of the at least two components, and wherein a weighting rule of the optimal predefined set of rules is a rule that instructs the routing mechanism to update the value of the first weighting, and

the processor choosing a first communication time, wherein the choosing is a function of at least one previously received value of a tracked environmental factor of the set of environmental factors and is a further function of at least one value previously received from a logging mechanism of the plurality of non-environmental logging mechanisms; and

the processor communicating the selected balanced workload allocation and the task request at the first communication time to the routing mechanism such that the routing mechanism allocates the one or more task workloads among the at least two components,

wherein the first environmental value of the one or more environmental values comprises an outdoor temperature outside a facility in which a component of the at least two components is operating and wherein a first updated non-environmental value of the continuously updated one or more non-environmental values comprises a local time zone at the facility.

5. The computer program product of claim 4, wherein neither of the at least two components is the computer system and wherein neither of the at least two components is a component of a computer system.

6. A computer system comprising a processor, a memory coupled to the processor, and a computer readable hardware storage device coupled to said processor, wherein the storage device stores program code configured to be executed by the processor via the memory to implement a method for allocating one or more task workloads of a plurality of task workloads to one or more components of a plurality of components of a multi-component power-consuming system coupled to a routing mechanism, wherein components of the plurality of components that are of a similar class are organized into an associated subsystem, and wherein the memory stores an environmental data table, a support table, and a task-allocation algorithm,

wherein each of the plurality of components of the multi-component power-consuming system is associated with one or more environmental sensors, and with a plurality of non-environmental logging mechanisms,

wherein the one or more environmental sensors identify one or more environmental characteristics associated with a set of operating environments of its associated component and continually update one or more environmental values associated with the one or more environmental characteristics of its associated component in the environmental data table,

wherein each of the plurality of non-environmental logging mechanisms identifies one or more non-environmental characteristics of its associated component and continuously updates one or more non-environmental values associated with the one or more non-environmental characteristics in the support table,

the method comprising:

the processor receiving a task request that comprises the one or more task workloads,

the processor, in response to receiving the task request, launching the task-allocation algorithm by selecting a balanced workload allocation of a plurality of balanced workload allocations as a function of one or more values stored in the environmental data table and the support table;

wherein the task-allocation algorithm comprises an optimal predefined set of rules that determines how the routing mechanism should allocate the one or more task workloads among at least two components of the plurality of components, wherein the determining comprises identifying one or more classes of components of the plurality of components, wherein the one or more classes of components are candidate components to receive the one or more task workloads, and wherein components or subsystems comprised by a class of the one or more classes of components are each associated with a same class of the one or more classes,

wherein the task-allocation algorithm computes, for each of the candidate components, an allocation weighting of a plurality of allocation weightings as a function of at least one of the one or more environmental values,

wherein the task-allocation algorithm may alter the selected balanced workload allocation, and

wherein the selected balanced workload allocation comprises an optimal predefined set of rules that determine how the routing mechanism should allocate the one or more task workloads among the at least two components, and

wherein the selecting is performed as a further function of a first weighting of the plurality of task-allocation weightings, wherein the first weighting revises the optimal predefined set of rules, wherein the first weighting is associated with one of the at least two components, and wherein a weighting rule of the optimal predefined set of rules is a rule that instructs the routing mechanism to update the value of the first weighting, and

the processor choosing a first communication time, wherein the choosing is a function of at least one previously received value of a tracked environmental factor of the set of environmental factors and is a further function of at least one value previously received from a logging mechanism of the plurality of non-environmental logging mechanisms; and

the processor communicating the selected balanced workload allocation and the task request at the first communication time to the routing mechanism such that the routing mechanism allocates the one or more task workloads among the at least two components,

wherein the first environmental value of the one or more environmental values comprises an outdoor temperature outside a facility in which a component of the at least two components is operating and wherein a first updated non-environmental value of the continuously updated one or more non-environmental values comprises a local time zone at the facility.

7. The computer system of claim 6, wherein neither of the at least two components is the computer system and wherein neither of the at least two components is a component of a computer system.