Performance optimization of workloads in virtualized information handling systems

Patent metadata

Field	Value
Publication number	US-2016306644-A1
Application number	US-201514686429-A
Country	US
Kind code	A1
Filing date	Apr 14, 2015
Priority date	Apr 14, 2015
Publication date	Oct 20, 2016
Grant date	—

Abstract

Methods and systems for performance optimization of workloads in virtualized information handling systems may analyze a workload at a virtual machine executing on a hypervisor. A workload policy specifying hardware resources for allocation to the virtual machine may be generated. After allocation of the hardware resources to the virtual machine, the workload may be executed on the virtual machine. The hypervisor may prioritize hardware resources among a plurality of virtual machines.

First claim

What is claimed is:

1 . A method for optimizing workloads on virtualized information handling systems, the method comprising:
receiving a workload request for a first virtual machine executing on a hypervisor executing on a virtualized information handling system, wherein the workload request specifies a computing task for execution at the first virtual machine, and wherein the hypervisor executes a plurality of virtual machines, including the first virtual machine; based on the workload request, determining workload attributes of the computing task, the workload attributes including dependency of the computing task on hardware resources accessible to the hypervisor; identifying the hardware resources available for allocation to the first virtual machine; based on the workload attributes determined and the hardware resources identified, determining a workload policy for the computing task, wherein the workload policy specifies the hardware resources used to execute the computing task at the first virtual machine; implementing the workload policy at the first virtual machine, including allocating the hardware resources specified in the workload policy to the first virtual machine; and executing the computing task at the first virtual machine in accordance with the workload policy.

2 . The method of claim 1 , wherein the allocating the hardware resources includes at least one of:
accessing a basic input/output system of the virtualized information handling system; and accessing respective hardware drivers for the hardware resources, the hardware drivers installed in the hypervisor.

3 . The method of claim 1 , wherein the workload request further specifies an application executing on the first virtual machine associated with the computing task, and wherein the workload attributes are selected from at least one of:
a degree of multithreading used by the computing task; a thread priority used by the computing task; an instruction set architecture used by the computing task; a memory capacity used by the computing task; a memory data throughput used by the computing task; a memory cache capacity used by the computing task; a memory cache architecture used by the computing task; a degree of non-uniform memory access permitted by the computing task; a degree of background execution permitted by the computing task; a network latency permitted by the computing task; a storage capacity used by the computing task; a storage data throughput used by the computing task; a storage latency permitted by the computing task; a degree of acceleration by a graphics processing unit used by the computing task; a vertical synchronization setting permitted by the computing task; a degree of digital signal processing used by the computing task; a degree of integer processing used by the computing task; a degree of background execution permitted by the computing task; and an operating system used by the computing task.

4 . The method of claim 1 , further comprising:
monitoring an operational state of at least some of the hardware resources, including monitoring which hardware resources are allocated to each of the plurality of virtual machines, wherein the allocating the hardware resources is performed based on the operational state.

5 . The method of claim 4 , further comprising:
when the operational state of at least some of the hardware resources indicates that the hardware resources specified in the workload policy are not available to the first virtual machine, modifying the workload policy according to the operational state.

6 . The method of claim 5 , wherein modifying the workload policy is based on at least one of:
a priority assigned to the first virtual machine; and a tolerance of the first virtual machine to variances in the dependency of the computing task on the hardware resources.

7 . The method of claim 1 , wherein allocating the hardware resources includes modifying at least some of the hardware resources, and wherein the hardware resources are selected from at least one of:
a number of processor cores available; a degree of exclusive access to processor cores available; a degree of processor multithreading available; a degree of background execution available; a degree of user mode execution available; a processor frequency; a processor architecture; a processor cache architecture; a processor power state; a processor cache capacity available; a processor power consumption available; an instruction set architecture available; a graphics processing unit available; a memory capacity available; a memory throughput available; a degree of non-uniform memory access available; a storage capacity available; a storage throughput available; a storage latency available; a network capacity available; a network throughput available; a network latency available; a system reboot feature; and a system power down feature.

8 . At least one non-transitory computer readable medium, comprising computer readable instructions for optimizing workloads on virtualized information handling systems, the instructions, when executed, cause a processor to:
receive a workload request for a first virtual machine executing on a hypervisor executing on a virtualized information handling system, wherein the workload request specifies a computing task for execution at the first virtual machine, and wherein the hypervisor executes a plurality of virtual machines, including the first virtual machine; based on the workload request, determine workload attributes of the computing task, the workload attributes including dependency of the computing task on hardware resources accessible to the hypervisor; identify the hardware resources available for allocation to the first virtual machine; based on the workload attributes determined and the hardware resources identified, determine a workload policy for the computing task, wherein the workload policy specifies the hardware resources used to execute the computing task at the first virtual machine; implement the workload policy at the first virtual machine, including instructions to allocate the hardware resources specified in the workload policy to the first virtual machine; and execute the computing task at the first virtual machine in accordance with the workload policy.

9 . The computer readable medium of claim 8 , wherein the instructions to allocate the hardware resources include instructions to perform at least one of:
access a basic input/output system of the virtualized information handling system; and access respective hardware drivers for the hardware resources, the hardware drivers installed in the hypervisor.

10 . The computer readable medium of claim 8 , wherein the workload request further specifies an application executing on the first virtual machine associated with the computing task, and wherein the workload attributes are selected from at least one of:
a degree of multithreading used by the computing task; a thread priority used by the computing task; an instruction set architecture used by the computing task; a memory capacity used by the computing task; a memory data throughput used by the computing task; a memory cache capacity used by the computing task; a memory cache architecture used by the computing task; a degree of non-uniform memory access permitted by the computing task; a degree of background execution permitted by the computing task; a network latency permitted by the computing task; a storage capacity used by the computing task; a storage data throughput used by the computing task; a storage latency permitted by the computing task; a degree of

Assignees

• Dell Products Lp

Inventors

• Kelly John
• Morton Charles G
• Farhan Munif M

Classifications

• G06F2009/45579

  I/O management, e.g. providing access to device drivers or storage · CPC title

• G06F9/5077

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F9/5044

  considering hardware capabilities · CPC title

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title

• G06F9/5094

  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 title