an energy-efficient hypervisor scheduler for asymmetric multi- core 1 ching-chi lin institute of...

22
An Energy-Efficient Hypervisor Scheduler for Asymmetric Multi-core 1 Ching-Chi Lin Institute of Information Science, Academia Sinica Department of Computer Science and Information Engineering, National Taiwan University You-Cheng Syu, Pangfeng Liu Department of Computer Science and Information Engineering, National Taiwan University Graduate Institute of Networking and Multimedia, Nation Taiwan University Chao-Jui Chang, Jan-Jan Wu Institute of Information Science, Academia Sinica Research Center for Information Technology Innovation, Academia Sinica Po-Wen Cheng, Wei-Te Hsu Information and Communications Research Laboratories, Industrial Technology Research Institute

Upload: esther-white

Post on 02-Jan-2016

217 views

Category:

Documents


3 download

TRANSCRIPT

An Energy-Efficient Hypervisor Scheduler for Asymmetric Multi-core

1

Ching-Chi LinInstitute of Information Science, Academia Sinica

Department of Computer Science and Information Engineering, National Taiwan University

You-Cheng Syu, Pangfeng LiuDepartment of Computer Science and Information Engineering, National Taiwan University

Graduate Institute of Networking and Multimedia, Nation Taiwan University

Chao-Jui Chang, Jan-Jan WuInstitute of Information Science, Academia Sinica

Research Center for Information Technology Innovation, Academia Sinica

Po-Wen Cheng, Wei-Te HsuInformation and Communications Research Laboratories, Industrial Technology Research Institute

AgendaIntroductionVirtual Core Scheduling ProblemSimulationConclusion

BackgroundAsymmetric multi-core architecture.

◦Consists of cores with different capabilities. ARM: big.LITTLE architecture. Qualcomm: asynchronous Symmetrical Multi-

Processing (aSMP) Nvidia: variable Symmetric Multiprocessing

(vSMP) …etc.

◦Aim to achieve both performance and energy-efficient. CPU-intensive v.s. non-CPU-intensive

3

MotivationDifferent scheduling goals

between homogenous and asymmetric multi-core platforms.◦Homogeneous multi-core: load-

balancing. Distributes workloads evenly to obtain

maximum performance.

◦Asymmetric multi-core: maximize power efficiency with modest performance sacrifices.

4

Hypervisor SchedulerAssigns the virtual cores to

physical cores for execution.◦Determines the execution order and

amount of time assigned to each virtual core according to a scheduling policy.

◦Current solutions Xen - credit-based scheduler KVM - completely fair scheduler

5

OS Kernel

GUEST2

Scheduler

VCPU VCPU

OS Kernel

GUEST2

Scheduler

VCPU VCPU

6

ARM Cortex-A15

ARM Cortex-A7

OS Kernel

GUEST1

Scheduler

VCPU VCPU

Hypervisor

vCPU Scheduler

Performance Power-saving

Low computing resource requirement

High computing resource requirement

If Guest OS scheduler is not asymmetric-aware, it will assign tasks to vCPUs evenly in order to achieve load balancing.

Task 1

Task 2

Task 3

Task 4

Hypervisor vCPU scheduler will assign vCPUs evenly to physical ARM cores since it is not asymmetric-aware.

Cannot take advantage on

big.LITTLE core architeture

Current Hypervisor Architecture and Problem

OS Kernel

GUEST2

Scheduler

VCPU VCPU

7

ARM Cortex-A15

ARM Cortex-A7

OS Kernel

GUEST1

Scheduler

VCPU VCPU

Hypervisor

vCPU Scheduler

Performance Power-saving

Assume that the scheduler in the Guest OS is asymmetric-aware.

Hypervisor vCPU scheduler will assign vCPUs evenly to physical ARM cores in order to achieve load-balancing.

Cannot take advantage on

big.LITTLE core architeture

Current Hypervisor Architecture and Problem(Cont.)

VCPUVCPU VCPUVCPUWaste energyPerformance Degradation

GoalDesign a new hypervisor

scheduling mechanism for asymmetric multi-core platform.◦Periodically generates an energy-

efficient scheduling plan. the amount of time each virtual core

should run on each physical core.

8

AssumptionsThe scheduling mechanism in the

guest OS is already asymmetry-aware.

The hypervisor is aware of the frequency of each virtual core.

9

Virtual Core Scheduling Problem

For every time period, given the operating frequency of each virtual core, the scheduler has to generate a scheduling plan such that◦The power consumption is minimized.◦Guarantee performance.

Scheduling plan◦the amount of time each virtual core

should run on each physical core.10

ModelTwo types of cores – virtual cores

and physical cores.

Power consumption of a physical core◦a function of core type, core

frequency, and the load of the core. Load: the percentage of time a core is

executing virtual cores.

11

PerformanceA ratio between the computing

resource assigned, to the computing resource requested.◦Ex: request 800 CPU cycles, assign

to 720 CPU cycles, the performance is 720/800 = 0.9.

12

Optimization ProblemObjective function:

n: number of physical core

Generate a scheduling plan◦ai,j:the amount of time executing

virtual core j on physical core i.◦Some constraints.

13

)min(1

n

iiPower

ConstraintsEqual performance of each

virtual core.◦Resource sufficient: all virtual core

with performance = 1.◦Resource insufficient: all virtual core

with equal performance less than 1.

Time assign to a virtual core should be less than a time interval.

14

Constraints(Cont.)A physical core has a fixed

amount of computing resources in a time interval.◦Load of a physical core ≦ 100%.

15

Our SolutionGiven the objective function and

constraints.Apply linear programming to

generate a feasible scheduling plan.◦Can find solution in a short time

since the number of virtual and physical cores are small constants.

16

SimulationCompare the power consumption

of our asymmetry-aware scheduler with that of a credit-based scheduler.

17

Simulation EnvironmentTwo types of physical cores

◦power-hunger “big” cores frequency: 1600MHz

◦power-efficient “little” cores frequency: 600MHz

◦The DVFS mechanism is disabled.

18

ScenarioTwo “big” and two “little’’ cores.Dual-core VM.Two sets of input:

◦Case 1: Both VMs with light workloads. 250MHz for each virtual core.

◦Case 2: One VM with heavy workloads, the other with modest workloads. Heavy:1200MHz for each virtual core Modest:600MHz for each virtual core.

19

Simulation Results

◦Case 1: asymmetry-aware method is about 43.2% of that of credit-based method.

◦Case 2:asymmetry-aware method uses 95.6% of energy used by the credit-base method.

20

Power(Watt)

Case 1Light-load VMs

Asymmetry-aware 0.295

Credit-based 0.683

Case 2Heavy-load VM + Modest-load

VM

Asymmetry-aware 2.382

Credit-based 2.491

SummaryWe develop an energy-efficient

asymmetry-aware scheduling mechanism for asymmetric multi-core platforms.◦Generate energy-efficient scheduling

plans with performance guarantee.Simulation results show that the

asymmetry-aware strategy saves up to 57.2% energy against credit-based method.

21

Thank you!