KVM and docker LXC Benchmarking with OpenStack - Comment
|
Passive benchmarking with docker LXC and KVM using OpenStack hosted in SoftLayer. These results provide initial incite as to why LXC as a technology choice offers benefits over traditional VMs and seek to provide answers as to the typical initial LXC question -- "why would I consider Linux Containers over VMs" from a performance perspective. Results here provide insight as to: - Cloudy ops times (start, stop, reboot) using OpenStack. - Guest micro benchmark performance (I/O, network, memory, CPU). - Guest micro benchmark performance of MySQL; OLTP read, read / write complex and indexed insertion. - Compute node resource consumption; VM / Container density factors. - Lessons learned during benchmarking. The tests here were performed using OpenStack Rally to drive the OpenStack cloudy tests and various other linux tools to test the guest performance on a "micro level". The nova docker virt driver was used in the Cloud scenario to realize VMs as docker LXC containers and compared to the nova virt driver for libvirt KVM. Please read the disclaimers in the presentation as this is only intended to be the "chip of the ice burg". Topics covered in this presentation slides: 1. Passive Benchmarking with docker LXC, KVM & OpenStack Hosted @ SoftLayer Boden Russell (brussell@us.ibm.com) IBM Global Technology Services Advanced Cloud Solutions & Innovation V2.0 2. FAQ - How is this version (v2.0) different from the initial benchmarks? – See the revision history within this document. - Are there any artifacts associated with the test? – Yes; see my github repo: https://github.com/bodenr/cloudy-docker-kvm-bench - Do these results imply an LXC based technology replaces the need for traditional hypervisors? – In my opinion, traditional VMs will become the “edge case” moving forward for use cases which are currently based on Linux flavored VMs. However I believe there will still be cases for traditional VMs, some of which are detailed in the LXC Realization presentation. - Are these results scientific? – No. Disclaimers have been attached to any documentation related to these tests to indicate such. These tests are meant to be a set of “litmus” tests to gain an initial understanding of how LXC compares to traditional hypervisors specifically in the Cloud space. - Do you welcome comments / feedback on the test? – Yes; the goal of these tests is to educate the community on LXC based technologies vs traditional hypervisors. As such they are fully disclosed in complete and hence open to feedback of any kind. 5/11/2014 2Document v2.0 3. FAQ Continued - Should I act on these results? – I believe the results provide enough information to gain some interest. I expect any organization, group or individual considering actions as a result will perform their own validation to assert the technology choice is beneficial for their consumption prior to adoption. - Is further / deeper testing and investigation warranted? – Absolutely. These tests should be conducted in a more active manner to understand the root causes for any differences. Additional tests and variations are also needed including; various KVM disk cache modes, skinny VM images (i.e. JeOS), impacts of database settings, docker storage drivers, etc. - Is this a direct measurement of the hypervisor (KVM) or LXC engine (docker)? – No, many factors play into results. For example the compute node has the nova virt driver running which is obviously different in implementation between nova libvirt-kvm and nova docker. Thus it’s implementation *may* have an impact on the compute node metrics and performance. 5/11/2014 Document v2.0 3 4. Revision History Revision Overview of changes V1.0 - Initial document release V2.0 - All tests were re-run using a single docker image throughout the tests (see my Dockerfile). - As the result of an astute reader, the 15 VM serial “packing” test reflects VM boot overhead rather than steady- state; this version clarifies such claims. - A new Cloudy test was added to better understand steady-state CPU. - Rather than presenting direct claims of density, raw data and graphs are presented to let the reader draw their own conclusions. - Additional “in the guest” tests were performed including blogbench. 5/11/2014 Document v2.0 4 5. Why Linux Containers (LXC) - Fast – Runtime performance near bare metal speeds – Management operations (run, stop , start, etc.) in seconds / milliseconds - Agile – VM-like agility – it’s still “virtualization” – Seamlessly “migrate” between virtual and bare metal environments - Flexible – Containerize a “system” – Containerize “application(s)” - Lightweight – Just enough Operating System (JeOS) – Minimal per container penalty - Inexpensive – Open source – free – lower TCO – Supported with out-of-the-box modern Linux kernel - Ecosystem – Growing in popularity – Vibrant community & numerous 3rd party apps 5/11/2014 5Document v2.0 6. Hypervisors vs. Linux Containers Hardware Operating System Hypervisor Virtual Machine Operating System Bins / libs App App Virtual Machine Operating System Bins / libs App App Hardware Hypervisor Virtual Machine Operating System Bins / libs App App Virtual Machine Operating System Bins / libs App App Hardware Operating System Container Bins / libs App App Container Bins / libs App App Type 1 Hypervisor Type 2 Hypervisor Linux Containers 5/11/2014 6 Containers share the OS kernel of the host and thus are lightweight. However, each container must have the same OS kernel. Containers are isolated, but share OS and, where appropriate, libs / bins. Document v2.0 7. Hypervisor VM vs. LXC vs. Docker LXC 5/11/2014 7Document v2.0 8. Docker in OpenStack - Havana – Nova virt driver which integrates with docker REST API on backend – Glance translator to integrate docker images with Glance - Icehouse – Heat plugin for docker - Both options are still under development 5/11/2014 8 nova-docker virt driver docker heat plugin DockerInc::Docke r::Container (plugin) Document v2.0 9. About This Benchmark - Use case perspective – As an OpenStack Cloud user I want a Ubuntu based VM with MySQL… Why would I choose docker LXC vs a traditional hypervisor? - OpenStack “Cloudy” perspective – LXC vs. traditional VM from a Cloudy (OpenStack) perspective – VM operational times (boot, start, stop, snapshot) – Compute node resource usage (per VM penalty); density factor - Guest runtime perspective – CPU, memory, file I/O, MySQL OLTP, etc. - Why KVM? – Exceptional performance DISCLAIMERS The tests herein are semi-active litmus tests – no in depth tuning, analysis, etc. More active testing is warranted. These results do not necessary reflect your workload or exact performance nor are they guaranteed to be statistically sound. 5/11/2014 9Document v2.0 10. Benchmark Environment Topology @ SoftLayer glance api / reg nova api / cond / etc keystone … rally nova api / cond / etc cinder api / sch / vol docker lxc dstat controller compute node glance api / reg nova api / cond / etc keystone … rally nova api / cond / etc cinder api / sch / vol KVM dstat controller compute node 5/11/2014 10 + Awesome! + Awesome! Document v2.0 11. Benchmark Specs 5/11/2014 11 Spec Controller Node (4CPU x 8G RAM) Compute Node (16CPU x 96G RAM) Environment Bare Metal @ SoftLayer Bare Metal @ SoftLayer Mother Board SuperMicro X8SIE-F Intel Xeon QuadCore SingleProc SATA [1Proc] SuperMicro X8DTU-F_R2 Intel Xeon HexCore DualProc [2Proc] CPU Intel Xeon-Lynnfield 3470-Quadcore [2.93GHz] (Intel Xeon-Westmere 5620-Quadcore [2.4GHz]) x 2 Memory (Kingston 4GB DDR3 2Rx8 4GB DDR3 2Rx8 [4GB]) x2 (Kingston 16GB DDR3 2Rx4 16GB DDR3 2Rx4 [16GB]) x 6 HDD (LOCAL) Digital WD Caviar RE3 WD5002ABYS [500GB]; SATAII Western Digital WD Caviar RE4 WD5003ABYX [500GB]; SATAII NIC eth0/eth1 @ 100 Mbps eth0/eth1 @100 Mbps Operating System Ubuntu 12.04 LTS 64bit Ubuntu 12.04 LTS 64bit Kernel 3.5.0-48-generic 3.8.0-38-generic IO Scheduler deadline deadline Hypervisor tested NA - KVM 1.0 + virtio + KSM (memory deduplication) - docker 0.10.0 + go1.2.1 + commit dc9c28f + AUFS OpenStack Trunk master via devstack Trunk master via devstack. Libvirt KVM nova driver / nova-docker virt driver OpenStack Benchmark Client OpenStack project rally NA Metrics Collection NA dstat Guest Benchmark Driver NA - Sysbench 0.4.12 - mbw 1.1.1.-2 - iibench (py) - netperf 2.5.0-1 - Blogbench 1.1 - cpu_bench.py VM Image NA - Scenario 1 (KVM): official ubuntu 12.04 image + mysql snapshotted and exported to qcow2 – 1080 MB - Scenario 2 (docker): guillermo/mysql -- 381.5 MB Hosted @Document v2.0 12. Test Descriptions: Cloudy Benchmarks 5/11/2014 12 Benchmark Benchmark Driver Description OpenStack Cloudy Benchmarks Serial VM boot (15 VMs) OpenStack Rally - Boot VM from image - Wait for ACTIVE state - Repeat the above a total of 15 times - Delete VMs Compute node steady-state VM packing cpu_bench.py - Boot 15 VMs in async fashion - Sleep for 5 minutes (wait for steady-state) - Delete all 15 VMs in async fashion VM reboot (5 VMs rebooted 5 times each) OpenStack Rally - Boot VM from image - Wait for ACTIVE state - Soft reboot VM 5 times - Delete VM - Repeat the above a total of 5 times VM snapshot (1 VM, 1 snapshot) OpenStack Rally - Boot VM from image - Wait for ACTIVE state - Snapshot VM to glance image - Delete VM Document v2.0 13. Test Descriptions: Guest Benchmarks 5/11/2014 13 Benchmark Benchmark Driver Description Guest Runtime Benchmarks CPU performance Sysbench from within the guest - Clear memory cache - Run sysbench cpu test - Repeat a total of 5 times - Average results over the 5 times OLTP (MySQL) performance Sysbench from within the guest - Clear memory cache - Run sysbench OLTP test - Repeat a total of 5 times - Average results over the 5 times MySQL Indexed insertion benchmark - Clear memory cache - Run iibench for a total of 1M inserts printing stats at 100K intervals - Collect data over 5 runs & average File I/O performance Sysbench from within the guest - Synchronous IO - Clear memory cache - Run sysbench OLTP test - Repeat a total of 5 times - Average results over the 5 times Memory performance Mbw from within the guest - Clear memory cache - Run mbw with array size of 1000 MiB and each test 10 times - Collect average over 10 runs per test Network performance Netperf - Run netperf server on controller - From guest run netperf client in IPv4 mode - Repeat text 5x - Average results Application type performance Blogbench - Clear memory cache - Run blogbench for 5 minutes - Repeat 5 times - Average read / write scores Document v2.0 14. STEADY STATE VM PACKING OpenStack Cloudy Benchmark 5/11/2014 14Document v2.0 15. Cloudy Performance: Steady State Packing - Benchmark scenario overview – Pre-cache VM image on compute node prior to test – Boot 15 VM asynchronously in succession – Wait for 5 minutes (to achieve steady-state on the compute node) – Delete all 15 VMs asynchronously in succession - Benchmark driver – cpu_bench.py - High level goals – Understand compute node characteristics under steady-state conditions with 15 packed / active VMs 5/11/2014 15 0 2 4 6 8 10 12 14 16 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 ActiveVMs Time Benchmark Visualization VMs Document v2.0 16. Cloudy Performance: Steady State Packing 5/11/2014 16 0 10 20 30 40 50 60 70 80 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281 289 297 305 313 321 CPUUsageInPercent Time Docker: Compute Node CPU (full test duration) usr sys Averages – 0.54 – 0.17 0 10 20 30 40 50 60 70 80 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281 289 297 305 313 321 329 337 345 CPUUsageInPercent Time KVM: Compute Node CPU (full test duration) usr sys Averages – 7.64 – 1.4 Document v2.0 17. Cloudy Performance: Steady State Packing 5/11/2014 17 0 2 4 6 8 10 12 14 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126 131 136 141 146 151 156 161 166 171 176 181 186 191 196 201 206 211 CPUUsageInPercent Time (31s – 243s) Docker: Compute Node Steady-State CPU (segment: 31s – 243s) usr sys 0 2 4 6 8 10 12 14 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126 131 136 141 146 151 156 161 166 171 176 181 186 191 196 201 206 211 CPUUsageInPercent Time (95s - 307s) KVM: Compute Node Steady-State CPU (segment: 95s – 307s) usr sys Averages – 0.2 – 0.03 Averages – 1.91 – 0.36 31 seconds 243 seconds 95 seconds 307 seconds Document v2.0 18. Cloudy Performance: Steady State Packing 5/11/2014 18 0 2 4 6 8 10 12 14 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 121 127 133 139 145 151 157 163 169 175 181 187 193 199 205 211 CPUUsageInPercent Time: KVM(95s - 307s) Docker(31s – 243s) Docker / KVM: Compute Node Steady-State CPU (Segment Overlay) docker-usr docker-sys kvm-usr kvm-sys docker: 31s KVM: 95s docker: 243s KVM: 307s Docker Averages – 0.2 – 0.03 KVM Averages – 1.91 – 0.36 Document v2.0 19. Cloudy Performance: Steady State Packing 5/11/2014 19 0.00E+00 1.00E+09 2.00E+09 3.00E+09 4.00E+09 5.00E+09 6.00E+09 7.00E+09 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281 289 297 305 313 321 MemoryUsed Time Docker: Compute Node Used Memory (full test duration) Memory Delta 734 MB Per VM 49 MB 0.00E+00 1.00E+09 2.00E+09 3.00E+09 4.00E+09 5.00E+09 6.00E+09 7.00E+09 1 10 19 28 37 46 55 64 73 82 91 100 109 118 127 136 145 154 163 172 181 190 199 208 217 226 235 244 253 262 271 280 289 298 307 316 325 334 MemoryUsed Time KVM: Compute Node Used Memory (full test duration) Memory Delta 4387 MB Per VM 292 MB Document v2.0 20. Cloudy Performance: Steady State Packing 5/11/2014 20 0.00E+00 1.00E+09 2.00E+09 3.00E+09 4.00E+09 5.00E+09 6.00E+09 7.00E+09 1 10 19 28 37 46 55 64 73 82 91 100 109 118 127 136 145 154 163 172 181 190 199 208 217 226 235 244 253 262 271 280 289 298 307 316 325 334 MemoryUsed Axis Title Docker / KVM: Compute Node Used Memory (Overlay) kvm docker Document v2.0 21. Cloudy Performance: Steady State Packing 5/11/2014 21 0 10 20 30 40 50 60 70 80 90 100 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281 289 297 305 313 321 1MinuteLoadAverage Time Docker: Compute Node 1m Load Average (full test duration) 1m Average 0.15 % 0 10 20 30 40 50 60 70 80 90 100 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281 289 297 305 313 321 329 337 1MinuteLoadAverage Time KVM: Compute Node 1m Load Average (full test duration) 1m Average 35.9 % Document v2.0 22. SERIALLY BOOT 15 VMS OpenStack Cloudy Benchmark 5/11/2014 22Document v2.0 23. Cloudy Performance: Serial VM Boot - Benchmark scenario overview – Pre-cache VM image on compute node prior to test – Boot VM – Wait for VM to become ACTIVE – Repeat the above steps for a total of 15 VMs – Delete all VMs - Benchmark driver – OpenStack Rally - High level goals – Understand compute node characteristics under sustained VM boots 5/11/2014 23 0 2 4 6 8 10 12 14 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ActiveVMs Time Benchmark Visualization VMs Document v2.0 24. Cloudy Performance: Serial VM Boot 5/11/2014 24 3.529113102 5.781662448 0 1 2 3 4 5 6 7 docker KVM TimeInSeconds Average Server Boot Time docker KVM Document v2.0 25. Cloudy Performance: Serial VM Boot 5/11/2014 25 0 5 10 15 20 25 30 35 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 CPUUsageInPercent Time Docker: Compute Node CPU usr sys Averages – 1.39 – 0.57 0 5 10 15 20 25 30 35 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 118 121 124 127 CPUUsageInPercent Time KVM: Compute Node CPU Usage usr sys Averages – 13.45 – 2.23 Document v2.0 26. Cloudy Performance: Serial VM Boot 5/11/2014 26 0 5 10 15 20 25 30 35 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101105109113117121125 CPUUsageInPercent Time Docker / KVM: Compute Node CPU (Unnormalized Overlay) kvm-usr kvm-sys docker-usr docker-sys Document v2.0 27. Cloudy Performance: Serial VM Boot 5/11/2014 27 y = 0.0095x + 1.008 y = 0.3582x + 1.0633 0 5 10 15 20 25 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 UsrCPUInPercent Time (8s - 58s) Docker / KVM: Serial VM Boot Usr CPU (segment: 8s - 58s) docker(8-58) kvm(8-58) Linear (docker(8-58)) Linear (kvm(8-58)) 8 seconds 58 seconds Document v2.0 28. Cloudy Performance: Serial VM Boot 5/11/2014 28 0.00E+00 5.00E+08 1.00E+09 1.50E+09 2.00E+09 2.50E+09 3.00E+09 3.50E+09 4.00E+09 4.50E+09 5.00E+09 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 MemoryUsed Time Docker: Compute Node Memory Used Memory Delta 677 MB Per VM 45 MB 0.00E+00 1.00E+09 2.00E+09 3.00E+09 4.00E+09 5.00E+09 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101105109113117121125 MemoryUsed Time KVM: Compute Node Memory Used Memory Delta 2737 MB Per VM 182 MB Document v2.0 29. Cloudy Performance: Serial VM Boot 5/11/2014 29 0.00E+00 5.00E+08 1.00E+09 1.50E+09 2.00E+09 2.50E+09 3.00E+09 3.50E+09 4.00E+09 4.50E+09 5.00E+09 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101105109113117121125 MemoryUsed Time Docker / KVM: Compute Node Memory Used (Unnormalized Overlay) kvm docker Document v2.0 30. Cloudy Performance: Serial VM Boot 5/11/2014 30 y = 1E+07x + 1E+09 y = 3E+07x + 1E+09 0.00E+00 5.00E+08 1.00E+09 1.50E+09 2.00E+09 2.50E+09 3.00E+09 3.50E+09 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 MemoryUsage Time (1s - 67s) Docker / KVM: Serial VM Boot Memory Usage (segment: 1s - 67s) docker kvm Linear (docker) Linear (kvm) 1 second 67 seconds Document v2.0 31. Cloudy Performance: Serial VM Boot 5/11/2014 31 0 5 10 15 20 25 30 35 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 1MinuteLoadAverage Time Docker: Compute Node 1m Load Average 1m Average 0.25 % 0 5 10 15 20 25 30 35 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 118 121 124 127 1MinuteLoadAverage Time KVM: Compute Node 1m Load Average 1m Average 11.18 % Document v2.0 32. SERIAL VM SOFT REBOOT OpenStack Cloudy Benchmark 5/11/2014 32Document v2.0 33. Cloudy Performance: Serial VM Reboot - Benchmark scenario overview – Pre-cache VM image on compute node prior to test – Boot a VM & wait for it to become ACTIVE – Soft reboot the VM and wait for it to become ACTIVE • Repeat reboot a total of 5 times – Delete VM – Repeat the above for a total of 5 VMs - Benchmark driver – OpenStack Rally - High level goals – Understand compute node characteristics under sustained VM reboots 5/11/2014 33 0 1 2 3 4 5 6 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 ActiveVMs Time Benchmark Visualization Active VMs Document v2.0 34. Cloudy Performance: Serial VM Reboot 5/11/2014 34 2.577879581 124.433239 0 20 40 60 80 100 120 140 docker KVM TimeInSeconds Average Server Reboot Time docker KVM Document v2.0 35. Cloudy Performance: Serial VM Reboot 5/11/2014 35 3.567586041 3.479760051 0 0.5 1 1.5 2 2.5 3 3.5 4 docker KVM TimeInSeconds Average Server Delete Time docker KVM Document v2.0 36. Cloudy Performance: Serial VM Reboot 5/11/2014 36 0 1 2 3 4 5 6 7 8 9 10 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 CPUUsageInPercent Time Docker: Compute Node CPU usr sys 0 1 2 3 4 5 6 7 8 9 10 1 72 143 214 285 356 427 498 569 640 711 782 853 924 995 1066 1137 1208 1279 1350 1421 1492 1563 1634 1705 1776 1847 1918 1989 2060 2131 2202 2273 2344 2415 2486 2557 2628 2699 2770 2841 2912 2983 3054 3125 CPUUsageInPercent Time KVM: Compute Node CPU usr sys Averages – 0.69 – 0.26 Averages – 0.84 – 0.18 Document v2.0 37. Cloudy Performance: Serial VM Reboot 5/11/2014 37 0.00E+00 5.00E+08 1.00E+09 1.50E+09 2.00E+09 2.50E+09 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 MemoryUsed Time Docker: Compute Node Used Memory Memory Delta 48 MB 0.00E+00 5.00E+08 1.00E+09 1.50E+09 2.00E+09 2.50E+09 1 81 161 241 321 401 481 561 641 721 801 881 961 1041 1121 1201 1281 1361 1441 1521 1601 1681 1761 1841 1921 2001 2081 2161 2241 2321 2401 2481 2561 2641 2721 2801 2881 2961 3041 3121 MemoryUsed Time KVM: Compute Node Used Memory Memory Delta 486 MB Document v2.0 38. Cloudy Performance: Serial VM Reboot 5/11/2014 38 0 0.5 1 1.5 2 2.5 3 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 1MinuteLoadAverage Time Docker: Compute Node 1m Load Average 1m Average 0.4 % 0 0.5 1 1.5 2 2.5 3 1 71 141 211 281 351 421 491 561 631 701 771 841 911 981 1051 1121 1191 1261 1331 1401 1471 1541 1611 1681 1751 1821 1891 1961 2031 2101 2171 2241 2311 2381 2451 2521 2591 2661 2731 2801 2871 2941 3011 3081 3151 1MinuteLoadAverage Time KVM: Compute Node 1m Load Average 1m Average 0.33 % Document v2.0 39. SNAPSHOT VM TO IMAGE OpenStack Cloudy Benchmark 5/11/2014 39Document v2.0 40. Cloudy Performance: Snapshot VM To Image - Benchmark scenario overview – Pre-cache VM image on compute node prior to test – Boot a VM – Wait for it to become ACTIVE – Snapshot the VM – Wait for image to become ACTIVE – Delete VM - Benchmark driver – OpenStack Rally - High level goals – Understand cloudy ops times from a user perspective 5/11/2014 40Document v2.0 41. Cloudy Performance: Snapshot VM To Image 5/11/2014 41 36.88756394 48.02313805 0 10 20 30 40 50 60 docker KVM TimeInSeconds Average Snapshot Server Time docker KVM Document v2.0 42. Cloudy Performance: Snapshot VM To Image 5/11/2014 42 0 1 2 3 4 5 6 7 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 CPUUsageInPercent Time Docker: Compute Node CPU usr sys Averages – 0.42 – 0.15 0 1 2 3 4 5 6 7 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 CPUUsageInPercent Time KVM: Compute Node CPU usr sys Averages – 1.46 – 1.0 Document v2.0 43. Cloudy Performance: Snapshot VM To Image 5/11/2014 43 1.48E+09 1.5E+09 1.52E+09 1.54E+09 1.56E+09 1.58E+09 1.6E+09 1.62E+09 1.64E+09 1.66E+09 1.68E+09 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 MemoryUsed Time KVM: Compute Node Used Memory Memory Delta 114 MB 1.6E+09 1.61E+09 1.62E+09 1.63E+09 1.64E+09 1.65E+09 1.66E+09 1.67E+09 1.68E+09 1.69E+09 1.7E+09 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 MemoryUsed Time Docker: Compute Node Memory Used Memory Delta 57 MB Document v2.0 44. Cloudy Performance: Snapshot VM To Image 5/11/2014 44 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 1MinuteLoadAverage Time Docker: Compute Node 1m Load Average 1m Average 0.06 % 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 1MinuteLoadAverage Time KVM: Compute node 1m Load Average 1m Average 0.47 % Document v2.0 45. GUEST PERFORMANCE BENCHMARKS Guest VM Benchmark 5/11/2014 45Document v2.0 46. Configuring Docker Container for 2CPU x 4G RAM - Configuring docker LXC for 2CPU x 4G RAM – Pin container to 2 CPUs / Mems • Create cpuset cgroup • Pin group to cpuset.mems to 0,1 • Pin group to cpuset.cpus to 0,1 • Add container root proc to tasks – Limit container memory to 4G • Create memory cgroup • Set memory.limit_in_bytes to 4G • Add container root proc to tasks – Limit blkio • Create blkio cgroup • Add container root process of LXC to tasks • Default blkio.weight of 500 5/11/2014 46Document v2.0 47. Guest Performance: CPU - Linux sysbench 0.4.12 cpu test - Calculate prime numbers up to 20000 - 2 threads - Instance size – 4G RAM – 2 CPU cores – 20G disk 5/11/2014 47Document v2.0 48. Guest Performance: CPU 5/11/2014 48 15.26 15.22 15.13 0 2 4 6 8 10 12 14 16 18 Bare Metal docker KVM Seconds Calculate Primes Up To 20000 Bare Metal docker KVM Document v2.0 49. Guest Performance: Memory - Linux mbw 1.1.1-2 - Instance size – 2 CPU – 4G memory - Execution options – 10 runs; average – 1000 MiB 5/11/2014 49Document v2.0 50. Guest Performance: Memory 5/11/2014 50 3823.3 4393.3 12881.61 3813.38 4395.92 12905.68 3428.95 3461.59 7223.23 0 2000 4000 6000 8000 10000 12000 14000 MEMCPY DUMB MCBLOCK MiB/s Memory Test Memory Benchmark Performance Bare Metal (MiB/s) docker (MiB/s) KVM (MiB/s) Document v2.0 51. Guest Performance: Network - Netperf 2.5.0-1 – Netserver running on controller – Netperf on guest – Run netperf 5 times & average results - Instance size – 2 CPU – 4G memory - Execution options – IPv4 / TCP 5/11/2014 51Document v2.0 52. Guest Performance: Network 5/11/2014 52 940.26 940.56 0 100 200 300 400 500 600 700 800 900 1000 docker KVM ThroughputIn10^6bits/second Network Throughput docker KVM Document v2.0 53. Guest Performance: File I/O Random Read - Linux sysbench 0.4.12 fileio test – Synchronous IO – Random read – Total file size of 150G – 16K block size – Test duration of 100s - Thread variations: 1, 8, 16, 32, 64 - Instance size – 4G RAM – 2 CPU cores – 200G disk - KVM specs – Disk cache mode set to none – Virtio – Deadline scheduler (host & guest) - Docker specs – AUFS storage driver – Deadline scheduler 5/11/2014 53Document v2.0 54. Guest Performance: File I/O Random Read 5/11/2014 54 0 500 1000 1500 2000 2500 1 2 4 8 16 32 64 TotalTransferredInKb/sec Threads Sysbench Synchronous File I/O Random Read docker KVM Document v2.0 55. Guest Performance: File I/O Random Read / Write - Linux sysbench 0.4.12 fileio test – Synchronous IO – Random read – Total file size of 150G – 16K block size – Read/Write ratio for combined random IO test: 1.50 – Test duration of 100s - Thread variations: 1, 8, 16, 32, 64 - Instance size – 4G RAM – 2 CPU cores – 200G disk - KVM specs – Disk cache mode set to none – Virtio – Deadline scheduler (host & guest) - Docker specs – AUFS storage driver – Deadline scheduler 5/11/2014 55Document v2.0 56. Guest Performance: File I/O Random Read / Write 5/11/2014 56 0 200 400 600 800 1000 1200 1400 1600 1 2 4 8 16 32 64 TotalTransferredInKb/sec Threads Sysbench Synchronous File I/O Random Read/Write @ R/W Ratio of 1.50 docker KVM Document v2.0 57. Guest Performance: MySQL OLTP - Linux sysbench 0.4.12 oltp test – Table size of 2,000,000 – MySQL 5.5 (installed on Ubuntu 12.04 LTS with apt-get) – 60 second iterations – Default MySQL cnf settings - Variations – Number of threads – Transactional random read & transactional random read / write - Instance size – 4G RAM – 2 CPU cores – 20G disk 5/11/2014 57Document v2.0 58. Guest Performance: MySQL OLTP 5/11/2014 58 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 1 2 4 8 16 32 64 TotalTransactions Threads MySQL OLTP Random Transactional Reads (60s) docker KVM Document v2.0 59. Guest Performance: MySQL OLTP 5/11/2014 59 0 2000 4000 6000 8000 10000 12000 14000 1 2 4 8 16 32 64 TotalTransactions Threads MySQL OLTP Random Transactional R/W (60s) docker KVM Document v2.0 60. Guest Performance: MySQL Indexed Insertion - Indexed insertion benchmark (iibench python script) – A total of 1,000,000 insertions – Print stats at 100K intervals - Instance size – 4G RAM – 2 CPU cores – 20G disk 5/11/2014 60Document v2.0 61. Guest Performance: MySQL Indexed Insertion 5/11/2014 61 0 20 40 60 80 100 120 140 100000 200000 300000 400000 500000 600000 700000 800000 900000 1000000 SecondsPer100KInsertionBatch Table Size In Rows MySQL Indexed Insertion @ 100K Intervals docker kvm Document v2.0 62. Guest Performance: BlogBench - Blogbench 1.1 – Test duration of 5m – Average results over 5 iterations of test - Instance size – 4G RAM – 2 CPU cores – 200G disk - KVM specs – Disk cache mode set to none – Virtio – Deadline scheduler (host & guest) - Docker specs – AUFS storage driver – Deadline scheduler 5/11/2014 62Document v2.0 63. Guest Performance: BlogBench 5/11/2014 63 398772.6 384769 0 50000 100000 150000 200000 250000 300000 350000 400000 450000 docker KVM Score Blogbench Read Scores docker KVM 1526.6 1285 0 200 400 600 800 1000 1200 1400 1600 1800 docker KVM Score Blogbench Write Scores docker KVM Document v2.0 64. OTHER CONSIDERATIONS 5/11/2014 64Document v2.0 65. Cloud Management Impacts on LXC 5/11/2014 65 0.17 3.529113102 0 0.5 1 1.5 2 2.5 3 3.5 4 docker cli nova-docker Seconds Docker: Boot Container - CLI vs Nova Virt docker cli nova-docker Cloud management often caps true ops performance of LXC Document v2.0 66. Ubuntu MySQL Image Size 5/11/2014 Document v2.0 66 381.5 1080 0 200 400 600 800 1000 1200 docker kvm SizeInMB Docker / KVM: Ubuntu MySQL docker kvm Out of the box JeOS images for docker are lightweight 67. Other Observations - Micro “synthetic” benchmarks do not reflect macro “application” performance – Always benchmark your “real” workload - Nova-docker virt driver still under development – Great start, but additional features needed for parity (python anyone?) – Additions to the nova-docker driver could change Cloudy performance - Docker LXC is still under development – Docker has not yet released v1.0 for production readiness - KVM images can be made skinnier, but requires additional effort - Increased density / oversubscription imposes additional complexity – Techniques to handle resource consumption surges which exceed capacity 5/11/2014 Document v2.0 67 68. REFERENCE 5/11/2014 68Document v2.0 69. References & Related Links - http://www.slideshare.net/BodenRussell/realizing-linux-containerslxc - http://www.slideshare.net/BodenRussell/kvm-and-docker-lxc-benchmarking-with- openstack - https://github.com/bodenr/cloudy-docker-kvm-bench - https://www.docker.io/ - http://sysbench.sourceforge.net/ - http://dag.wiee.rs/home-made/dstat/ - http://www.openstack.org/ - https://wiki.openstack.org/wiki/Rally - https://wiki.openstack.org/wiki/Docker - http://devstack.org/ - http://www.linux-kvm.org/page/Main_Page - https://github.com/stackforge/nova-docker - https://github.com/dotcloud/docker-registry - http://www.netperf.org/netperf/ - http://www.tokutek.com/products/iibench/ - http://www.brendangregg.com/activebenchmarking.html 5/11/2014 69Document v2.0 70. Cloudy Benchmark: Serially Boot 15 VMs - KVM +------------------+-------+---------------+---------------+---------------+---------------+---------------+ | action | count | max (sec) | avg (sec) | min (sec) | 90 percentile | 95 percentile | +------------------+-------+---------------+---------------+---------------+---------------+---------------+ | nova.boot_server | 15 | 7.37148094177 | 5.78166244825 | 4.77369403839 | 6.67956886292 | 7.07061390877 | +------------------+-------+---------------+---------------+---------------+---------------+---------------+ +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ | max (sec) | avg (sec) | min (sec) | 90 pecentile | 95 percentile | success/total | total times | +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ | 7.58968496323 | 6.00853565534 | 4.99443006516 | 6.91288709641 | 7.28662061691 | 1.0 | 15 | +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ - Docker +------------------+-------+---------------+---------------+---------------+---------------+---------------+ | action | count | max (sec) | avg (sec) | min (sec) | 90 percentile | 95 percentile | +------------------+-------+---------------+---------------+---------------+---------------+---------------+ | nova.boot_server | 15 | 5.18499684334 | 3.52911310196 | 2.93864893913 | 4.74490590096 | 4.95752367973 | +------------------+-------+---------------+---------------+---------------+---------------+---------------+ +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ | max (sec) | avg (sec) | min (sec) | 90 pecentile | 95 percentile | success/total | total times | +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ | 5.43275094032 | 3.77053097089 | 3.12985610962 | 4.95886874199 | 5.18047580719 | 1.0 | 15 | +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ 5/11/2014 70Document v2.0 71. Cloudy Performance: Serial VM Reboot - KVM +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ | action | count | max (sec) | avg (sec) | min (sec) | 90 percentile | 95 percentile | +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ | nova.reboot_server | 10 | 124.900292158 | 124.433238959 | 123.947879076 | 124.881286669 | 124.890789413 | | nova.boot_server | 2 | 7.05096197128 | 6.82815694809 | 6.6053519249 | 7.00640096664 | 7.02868146896 | | nova.delete_server | 2 | 4.46658396721 | 3.47976005077 | 2.49293613434 | 4.26921918392 | 4.36790157557 | +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ | max (sec) | avg (sec) | min (sec) | 90 pecentile | 95 percentile | success/total | total times | +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ | 633.087348938 | 632.493344903 | 631.899340868 | 632.968548131 | 633.027948534 | 0.4 | 5 | +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ - Docker +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ | action | count | max (sec) | avg (sec) | min (sec) | 90 percentile | 95 percentile | +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ | nova.reboot_server | 25 | 4.48567795753 | 2.57787958145 | 2.35410904884 | 3.0847319603 | 3.48342533112 | | nova.boot_server | 5 | 4.16244912148 | 3.5675860405 | 3.05103397369 | 4.03664107323 | 4.09954509735 | | nova.delete_server | 5 | 3.54331803322 | 3.52483625412 | 3.50456190109 | 3.53761086464 | 3.54046444893 | +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ +---------------+---------------+---------------+--------------+---------------+---------------+-------------+ | max (sec) | avg (sec) | min (sec) | 90 pecentile | 95 percentile | success/total | total times | +---------------+---------------+---------------+--------------+---------------+---------------+-------------+ | 21.5702910423 | 19.9976443768 | 18.7037060261 | 20.997631073 | 21.2839610577 | 1.0 | 5 | +---------------+---------------+---------------+--------------+---------------+---------------+-------------+ 5/11/2014 71Document v2.0 72. Cloud Performance: Snapshot VM To Image - KVM +--------------------+-------+----------------+----------------+----------------+----------------+----------------+ | action | count | max (sec) | avg (sec) | min (sec) | 90 percentile | 95 percentile | +--------------------+-------+----------------+----------------+----------------+----------------+----------------+ | nova.delete_image | 1 | 0.726859092712 | 0.726859092712 | 0.726859092712 | 0.726859092712 | 0.726859092712 | | nova.create_image | 1 | 48.0231380463 | 48.0231380463 | 48.0231380463 | 48.0231380463 | 48.0231380463 | | nova.boot_server | 2 | 32.7824101448 | 19.4164011478 | 6.05039215088 | 30.1092083454 | 31.4458092451 | | nova.delete_server | 2 | 12.3564949036 | 8.40917897224 | 4.46186304092 | 11.5670317173 | 11.9617633104 | +--------------------+-------+----------------+----------------+----------------+----------------+----------------+ +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ | max (sec) | avg (sec) | min (sec) | 90 pecentile | 95 percentile | success/total | total times | +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ | 104.401446104 | 104.401446104 | 104.401446104 | 104.401446104 | 104.401446104 | 1.0 | 1 | +---------------+---------------+---------------+---------------+---------------+---------------+-------------+ - Docker (defect deleting image) +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ | action | count | max (sec) | avg (sec) | min (sec) | 90 percentile | 95 percentile | +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ | nova.create_image | 1 | 36.8875639439 | 36.8875639439 | 36.8875639439 | 36.8875639439 | 36.8875639439 | | nova.boot_server | 2 | 3.96964478493 | 3.84809792042 | 3.72655105591 | 3.94533541203 | 3.95749009848 | | nova.delete_server | 2 | 4.48610281944 | 4.46519696712 | 4.44429111481 | 4.48192164898 | 4.48401223421 | +--------------------+-------+---------------+---------------+---------------+---------------+---------------+ +-----------+-----------+-----------+--------------+---------------+---------------+-------------+ | max (sec) | avg (sec) | min (sec) | 90 pecentile | 95 percentile | success/total | total times | +-----------+-----------+-----------+--------------+---------------+---------------+-------------+ | n/a | n/a | n/a | n/a | n/a | 0 | 1 | +-----------+-----------+-----------+--------------+---------------+---------------+-------------+ 5/11/2014 72Document v2.0 |
|||
| Posted by : peter88 | Post date : 2020-01-06 14:29 | ||
| Category : Technology | Views : 387 | ||
New Comment
