Container-based Virtualization University of Amsterdam MSc. System - - PowerPoint PPT Presentation

▶
container based virtualization
SMART_READER_LITE
LIVE PREVIEW

Container-based Virtualization University of Amsterdam MSc. System - - PowerPoint PPT Presentation

Mar 04, 2024 482 likes •762 views

Power Efficiency of Hypervisor and Container-based Virtualization University of Amsterdam MSc. System & Network Engineering Research Project II Jeroen van Kessel 02-02-2016 Supervised by: dr. ir. Arie Taal dr. Paola Grosso Significance

slide-1
SLIDE 1

Power Efficiency of Hypervisor and Container-based Virtualization

University of Amsterdam

  • MSc. System & Network Engineering

Research Project II

Jeroen van Kessel 02-02-2016 Supervised by:

  • dr. ir. Arie Taal
  • dr. Paola Grosso
slide-2
SLIDE 2

Significance

  • Growing consumption of cloud services

Cisco claims: "by 2019, more than 86 percent of all workload will be processed by cloud data centers"

  • More awareness on power consumption
  • Container-based virtualization is an emerging technology

Docker became very popular in a relative short time

2/26

slide-3
SLIDE 3

Related Work (1)

  • Van der Poll [2015]
  • Power consumption of two open-source hypervisors, KVM and Xen
  • KVM as a more green solution than Xen.
  • IBM research division [2014]
  • Performance comparison of virtual nodes ran by a hypervisor versus Linux containers
  • Docker had equal or faster performance compared to KVM
  • Ericsson research division [2015] [1]

Empirical Investigation of power consumption of virtualization platforms

  • Compared number of VMs and Containers
  • Power impact of CPU, memory and HDD
  • Used a Power Measurement Device
  • Results: virtualization platforms behave similarly

in idle state and in CPU/Memory stress test

Figure 1: Power Measurement Device

Image source: http://media.bestofmicro.com/green-power-cpu-performance,E-0-228600-13.jpg

3/26

slide-4
SLIDE 4

Related Work (2)

  • Ericsson research division [2015] [2]
  • Performance comparison between traditional hypervisors and Linux containers
  • Containers achieve generally better performance compared to traditional VMs

Figure 2: LINPACK CPU performance

Source: Ericsson, Hypervisors vs. Lightweight Virtualization: a Performance Comparison [2015]

4/26

slide-5
SLIDE 5

Research Question (1)

Is there a difference in power efficiency under a traditional hypervisor-based virtualization versus Linux containers?

5 /26

slide-6
SLIDE 6

VMs and Containers

  • Container virtualization is a lightweight alternative to hypervisor-based

virtualization

  • Container runs on top of the same shared host OS kernel
  • While VMs install a full Guest OS
  • Containers do not isolate resources as well as hypervisors

Figure 3: Image source: http://gordonsun-blog.s3.amazonaws.com/wp-content/uploads/2015/05/docker-containers-vs-vms.png

Xen 4.5.1 Docker 1.9.1

6 /26

slide-7
SLIDE 7

Approach

Figure 4: Power Measurement Setup

7 /26

slide-8
SLIDE 8

Measurement Setup

Running Ubuntu 15.10 x64

Table 1: IBM 1U Server Specifications

8 /26

slide-9
SLIDE 9

9 /26

slide-10
SLIDE 10
slide-11
SLIDE 11

(1) Power Usage Sensor (2) Data Acquisition Interface Board

Measurement Sensors

Image source: eurocircuits.com blueprints

11 /26

slide-12
SLIDE 12

Measurement results

Synthetics Applications

Table 2: Dominant components and their synthetics benchmark applications

But first, IDLE is measured as a baseline

12 /26

slide-13
SLIDE 13

IDLE Power consumption (1)

13 /26

7.73 7.39 7.27 6.91 6.63 6.56 4.93 4.87 4.86 1 2 3 4 5 6 7 8 9 Xen running 1 Virtual Node Docker running 1 Container Native OS

Average Power Usage in Watt (W)

CPU Memory HDD

slide-14
SLIDE 14

Total IDLE Power consumption (2)

7.73 7.38 7.27 6.91 6.63 6.55 4.93 4.87 4.86

2 4 6 8 10 12 14 16 18 20 Xen running 1 Virtual Node Docker running 1 Container Native OS

Total Power Usage in Watt (W)

HDD Memory CPU

19.57 18.88 18.68

14 /26

slide-15
SLIDE 15

CPUs and Cores

Experiment 1: 4 cores on CPU1 Experiment 2: 2 cores both CPU1 and CPU2

Image source: http://img.tomshardware.com/us/2007/10/29/hitting_4ghz_with_air_cooling/intel_penryn_45_nm_octo_core.jpg

CPU1 CPU2

15 /26

CPU1 CPU2

slide-16
SLIDE 16

Power Consumption during LINPACK on Docker (1) and Xen (2)

(1) Docker: All 4 cores used on physical CPU1

1 2 3

(2) Xen: All 4 cores used on physical CPU1

16 /26

slide-17
SLIDE 17

Power Consumption during LINPACK on Docker (1) and Xen (2)

(1) Docker: 2 cores used of each CPU1 and CPU2 (2) Xen: 2 cores used of each CPU1 and CPU2

17 /26

slide-18
SLIDE 18

Total Power consumption of LINPACK CPU Docker and Xen

18 /26

slide-19
SLIDE 19

Energy efficiency during LINPACK (CPU intensive)

19 /26

1397 1341 1581 1502 200 400 600 800 1000 1200 1400 1600 1800 Xen 1 CPU Xen 2 CPUs Docker 1 CPU Docker 2 CPUs

Power Efficiency in MFLOPS / W

slide-20
SLIDE 20

Power Consumption during sysbench 350GiB memory transfer

Power Usage of all memory banks (1) Power Usage of all memory banks (2)

20 /26

slide-21
SLIDE 21

Total Power efficiency during a sysbench 350GiB memory transfer

21 /26

100.661 101.560 104.411 0.000 20.000 40.000 60.000 80.000 100.000 120.000 Xen Memory Docker Memory Native OS Memory

Power Efficiency in (Ops/sec) / W

slide-22
SLIDE 22

Power Consumption during Bonnie++ (25GiB)

22 /26

slide-23
SLIDE 23

Total Power efficiency during Bonnie++ 25GiB (Sequential Writes in Kb/sec)

23 /26

11.555 12.252 12.140 0.000 2.000 4.000 6.000 8.000 10.000 12.000 14.000 Xen HDD Docker HDD Native OS HDD

Power Efficiency in (Kb/sec) / W (Sequential Write)

slide-24
SLIDE 24

Conclusion

Research question: Is there a difference in power efficiency under a traditional hypervisor-based virtualization versus Linux containers? Performance results match with the Ericsson research Power Efficiency results: CPU: Docker is more efficient in terms of power Memory, HDD (Writes) and IDLE: Docker is more efficient but almost negligible

24 /26

slide-25
SLIDE 25

Future Work

  • Benchmark multiple virtual nodes and containers
  • Investigate the energy footprint of the network component
  • Test other applications

Performance evaluation of real life applications

  • Futher investigation of other platforms such as LXC, KVM and VMware
  • Investigate energy impact of (Xen) Paravirtualization

25 /26

slide-26
SLIDE 26

Thanks for your attention