Feedback

Your feedback is important to keep improving our website and offer you a more reliable experience.

Linux Kernel Performance

Linux development evolves rapidly. The performance and scalability of the OS kernel has been a key part of its success. However, discussions have appeared on LKML (Linux Kernel Mailing List) regarding large performance regression between kernel versions. These discussions underscore the need for a systematic and disciplined way to characterize, improve, and test Linux kernel performance. Our goal is to work with the Linux community to further enhance the Linux kernel with consistent performance increases (avoiding degradations) across releases. The information available on this site gives community members better information about what 0-Day and LKP (Linux Kernel Performance) are doing to preserve performance integrity of the kernel.

0-Day CI Linux Kernel Performance Report (v5.1)

BY Rong Chen ON May 20, 2019
  1. Introduction

0-Day CI is an automated Linux kernel test service that provides comprehensive test coverage of the Linux kernel. It covers kernel build, static analysis, boot, functional, performance and power tests. This report shows the recent observations of kernel performance status on IA platform based on the test results from 0-Day CI service. It is structured in the following manner:

  • Section 2, merged regressions and improvements in v5.1 release candidates

  • Section 3, test parameter description

  • Section 4, captured regressions and improvements by shift-left testing during developers’ and maintainers’ tree during v5.1 release cycle

  • Section 5, performance comparison among different kernel releases

  • Section 6, test machine list

 

  1. Test parameters

Here are the descriptions for each parameter/field used in the tests.

 

Classification

Name

Description

General

runtime

Run the test case within a certain time period (seconds or minutes)

 

nr_task

If it is an integer, which means the number of processes/threads (to run the workload) of this job. Default is 1.

If it is a percentage, e.g. 200% means the number of processes/threads is double of cpu number

 

nr_threads

Alias of nr_task

 

iterations

Number to repeat this job

 

test_size

Test disk size or memory size

 

set_nic_irq_affinity

Set NIC interrupt affinity

 

disable_latency_stats

Latency_stats may introduce too much noise if there are too many context switches, allow to disable it

 

transparent_hugepage

Set transparent hugepage policy (/sys/kernel/mm/transparent_hugepage)

 

boot_params:bp1_memmap

Boot parameters of memmap

 

disk:nr_pmem

number of pmem partitions used by test

 

swap:priority

Priority means the  priority of the swap device. priority is a value between -1 and 32767, the default is -1 and higher priority with higher value.

Test Machine

model

Name of Intel processor microarchitecture

 

brand

Brand name of cpu

 

cpu_number

Number of cpu

 

memory

Size of memory


 

  1. Linux Kernel v5.1 Release Test

The v5.1 release of the Linux kernel was on May 6, 2019. Some of the significant changes in the release include BPF spinlocks, more year-2038 preparation, the TEO CPU-idle governor, The io_uring fast asynchronous I/O mechanism, initial support for pidfds (file descriptors that refer to a process), the SafeSetID security module, and much more. See the KernelNewbies 5.1 page for lots of details. 0-Day CI monitored the release closely to trace down the performance status on IA platform. 0-Day observed 3 regressions and 4 improvements during feature development phase for v5.1. We will share more detailed information together with correlated patches that led to the results. Note that the assessment is limited by the test coverage 0-Day has now. The list is summarized in the observation summary section.

  1. Observation Summary

0-Day CI observed 3 regressions and 4 improvements during feature development phase for v5.1, which is in the time frame from v5.1-rc1 to v5.1 release.

Test Indicator

Report

Test Scenario

Test Machine

Development Base

Status

vm-scalability.median

[LKP] [block] 6861428921: 1003.2% improvement

runtime: 300

thp_enabled: always

thp_defrag: always

nr_task: 32

nr_ssd: 1

test: swap-w-seq

cpufreq_governor: performance

lkp-skl-2sp7

v5.0-rc4


 

merged at v5.1-rc1

vm-scalability.median

[LKP] [mm] af53d3e9e0: 12.5% improvement

runtime: 300s

test: small-allocs

cpufreq_governor: performance

lkp-bdw-ep2

v5.1-rc5

merged at v5.1-rc6

will-it-scale.per_process_ops

[LKP] [x86/retpolines] a9d57ef15c: -2.9% regression

nr_task: 50%

mode: process

test: poll2

cpufreq_governor: performance

lkp-bdw-ep3d

v5.1-rc2

merged at v5.1-rc3, no response from author yet

will-it-scale.per_thread_ops

[LKP] [x86, retpolines] ce02ef06fc: 3.1% improvement

nr_task: 100%

mode: thread

test: futex3

cpufreq_governor: performance

lkp-bdw-ep3b

v5.0-rc1

merged at v5.1-rc1, has backported to stable


 

will-it-scale.per_thread_ops

[LKP] [futex] 5a07168d8d: -2.7% regression

nr_task: 100%

mode: thread

test: futex3

cpufreq_governor: performance

lkp-bdw-ep3b

v5.1-rc1

merged at v5.1-rc2, no response from author yet

 

will-it-scale.per_thread_ops

[LKP] [fsnotify] 5f02a87763: -5.1% regression

nr_task: 16

mode: thread

test: unlink1

cpufreq_governor: performance

lkp-ivb-d02

v5.0-rc4

merged at v5.1-rc1, no response from author yet

  1. vm-scalability.median

vm-scalability exercises functions and regions of the mm subsystem of the Linux kernel. We tested on multiple machines such as HSW EP server, during which we reported improvement on one test scenario.

Scenario: swap-w-seq test

 

Commit 6861428921 was reported to have 1003.2% improvement of vm-scalability.median when comparing to v5.0-rc4." It was merged to mainline at v5.1-rc1.              

 

Correlated commits

6861428921

block: always define BIO_MAX_PAGES as 256

branch

linus/master

report

[LKP] [block] 6861428921: vm-scalability.median 1003.2% improvement

test scenario

runtime: 300

thp_enabled: always

thp_defrag: always

nr_task: 32

nr_ssd: 1

test: swap-w-seq

cpufreq_governor: performance

test machine

lkp-skl-2sp7

status

merged at v5.1-rc1

 

  1. will-it-scale.per_process_ops

Will-it-scale takes a test case and runs it from 1 through to n parallel copies to see if the test case will scale. It builds both process and threads based tests in order to see any differences between the two.

Scenario: process poll2

 

Commit a9d57ef15c was reported to have -2.9% of will-it-scale.per_process_ops when comparing to v5.1-rc2." It was merged to mainline at v5.1-rc3.

 

Correlated commits

a9d57ef15c

x86/retpolines: Disable switch jump tables when retpolines are enabled

branch

linus/master

report

[LKP] [x86/retpolines] a9d57ef15c: will-it-scale.per_process_ops -2.9% regression

test scenario

nr_task: 50%

mode: process

test: poll2

cpufreq_governor: performance

test machine

lkp-bdw-ep3d

status

merged at v5.1-rc3

  1. will-it-scale.per_thread_ops

Will-it-scale takes a test case and runs it from 1 through to n parallel copies to see if the test case will scale. It builds both process and threads based tests in order to see any differences between the two.

Scenario: thread unlink1

 

Commit 5f02a87763 was reported to have -5.1% of will-it-scale.per_thread_ops when comparing to v5.0-rc4." It was merged to mainline at v5.1-rc1.

 

Correlated commits

5f02a87763

fsnotify: annotate directory entry modification events

branch

linus/master

report

[LKP] [fsnotify] 5f02a87763: will-it-scale.per_thread_ops -5.1% regression

test scenario

nr_task: 16

mode: thread

test: unlink1

cpufreq_governor: performance

test machine

lkp-ivb-d02 lkp-bdw-ep3d

status

merged at v5.1-rc1

 

  1. Shift-Left Testing

Beyond testing trees in the upstream kernel, 0-Day CI also tests developers’ and maintainers’ trees, which can catch issues earlier and reduce wider impact. We call it “shift-left” testing. During the v5.0 release cycle, 0-Day CI had reported 14 major performance regressions and 5 major improvements by doing shift-left testing. We will share more detailed information together with possible code changes that led to this result for some of these, though the assessment is limited by the test coverage we have now. The whole list is summarized at report summary section.

  1. Report Summary

0-Day CI had reported 14 performance regressions and 5 improvements by doing shift-left testing on developer and maintainer repos.

 

Test Indicator

Mail

Test Scenario

Test Machine

Status

aim7.jobs-per-min

[LKP] [MD] 4bc034d353: -86.0% regression

disk: 4BRD_12G

md: RAID0

fs: xfs

test: sync_disk_rw

load: 300

cpufreq_governor: performance

lkp-ivb-ep01

Currently not merged, "That is expected.  The following commit should restore the performance."

aim7.jobs-per-min

[LKP] [xfs] fa3fe73bed: -10.9% regression

disk: 4BRD_12G

md: RAID0

fs: xfs

test: disk_rr

load: 3000

cpufreq_governor: performance

lkp-ivb-ep01

Currently not merged, no response from author yet

aim7.jobs-per-min

[LKP] [btrfs] c99d2765f3: 4.5% improvement

disk: 1BRD_48G

fs: btrfs

test: disk_src

load: 500

cpufreq_governor: performance

lkp-skl-2sp7

Currently not merged

aim7.jobs-per-min

[LKP] [btrfs] c8eaeac7b7: -11.7% regression

disk: 4BRD_12G

md: RAID0

fs: btrfs

test: disk_rr

load: 1500

cpufreq_governor: performance

lkp-ivb-ep01

Currently not merged, no response from author yet

fio.write_bw_MBps

[LKP] [btrfs] 8d47a0d8f7: -28.6% regression

disk: 2pmem

fs: btrfs

runtime: 200s

nr_task: 50%

time_based: tb

rw: randwrite

bs: 4k

ioengine: libaio

test_size: 100G

cpufreq_governor: performance

lkp-hsw-ep2

Currently not merged, no response from author yet

fxmark.hdd_btrfs_MWUM_1_bufferedio.works/sec

[LKP] [loop] 56a85fd837: 224.0% improvement

disk: 1HDD

media: hdd

test: MWUM

fstype: btrfs

directio: bufferedio

cpufreq_governor: performance

ivb44

Currently not merged

lmbench3.TCP.socket.bandwidth.10MB.MB/sec

[LKP] [tcp] 01b4c2aab8: -20.2% regression

test_memory_size: 50%

nr_threads: 100%

mode: development

test: TCP

cpufreq_governor: performance

lkp-bdw-ep4

Currently not merged, no response from author yet

netperf.Throughput_Mbps

[LKP] [tcp] 8b27dae5a2: -25.7% regression

ip: ipv4

runtime: 900s

nr_threads: 200%

cluster: cs-localhost

test: TCP_STREAM

cpufreq_governor: performance

lkp-skl-fpga01

Currently not merged, no response from author yet

stress-ng.bad-altstack.ops_per_sec

[LKP] [locking/rwsem] 1b94536f2d: -32.7% regression

nr_threads: 100%

disk: 1HDD

testtime: 5s

class: memory

cpufreq_governor: performance

lkp-knm02

Currently not merged, "later patches that enable reader optimistic spinning should

bring back some of the lost performance.”

stress-ng.schedpolicy.ops_per_sec

[LKP] [sched/core] 900e532bf6: -91.5% regression

nr_threads: 100%

disk: 1HDD

testtime: 1s

class: interrupt

cpufreq_governor: performance

lkp-bdw-ep3

Currently not merged, no response from author yet

vm-scalability.median

[LKP] [mm/page_alloc.c] 733ad8b6f8: 4.6% improvement

runtime: 300s

size: 1T

test: lru-shm

cpufreq_governor: performance

lkp-bdw-ep2

Currently not merged

vm-scalability.median

[LKP] [tmpfs, devtmpfs, ramfs, rootfs] 9d65ac8a35: -19.4% regression

runtime: 300s

size: 16G

test: shm-pread-rand

cpufreq_governor: performance

lkp-bdw-ep2

Currently not merged, no response from author yet

vm-scalability.median

[LKP] [vfs] 27eb9d500d: -19.4% regression

runtime: 300s

size: 16G

test: shm-pread-rand

cpufreq_governor: performance

lkp-bdw-ep2

Currently not merged, no response from author yet

vm-scalability.median

[LKP] [vfs] 9af1f3b769: 18.5% improvement

runtime: 300s

size: 256G

test: lru-shm-rand

cpufreq_governor: performance

lkp-bdw-ep2

Currently not merged

vm-scalability.median

[LKP] [locking/rwsem] e59710760a: -32.1% regression

runtime: 300s

test: small-allocs

cpufreq_governor: performance

lkp-bdw-ex2

Currently not merged, no response from author yet

will-it-scale.per_process_ops

[LKP] [mm] 71ee870ccb: -2.9% regression

nr_task: 50%

mode: process

test: poll2

cpufreq_governor: performance

lkp-bdw-ep3d

Currently not merged, no response from author yet

will-it-scale.per_process_ops

[LKP] [x86/fpu] 657f807aa7: -1.5% regression

nr_task: 50%

mode: process

test: futex2

cpufreq_governor: performance

lkp-bdw-ep3d

Currently not merged, no response from author yet

will-it-scale.per_thread_ops

[LKP] [crypto] 71052dcf4b: 59.5% improvement

nr_task: 100%

mode: thread

test: signal1

cpufreq_governor: performance

lkp-knm01

Currently not merged

will-it-scale.per_thread_ops

[LKP] [locking/rwsem] adc32e8877: -21.0% regression

nr_task: 50%

mode: thread

test: page_fault1

cpufreq_governor: performance

lkp-bdw-ep3d

Currently not merged, no response from author yet

  1. aim7.jobs-per-min

aim7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of a multiuser system.

 

  1. scenario: sync_disk_rw test on xfs

 

Commit 4bc034d353 was reported to have -86.0% of aim7.jobs-per-min when comparing to v5.1-rc3.

 

Correlated commits

4bc034d353

Revert "MD: fix lock contention for flush bios"

branch

block/for-5.2/block

report

[LKP] [MD] 4bc034d353: aim7.jobs-per-min -86.0% regression

test scenario

disk: 4BRD_12G

md: RAID0

fs: xfs

test: sync_disk_rw

load: 300

cpufreq_governor: performance

test machine

lkp-ivb-ep01

status

Not merged at v5.1, “That is expected. The following commit should restore the performance.”

 

  1. fio.write_bw_MBps

Fio is a tool that will spawn a number of threads or processes doing a particular type of I/O action as specified by the user.

  1. scenario: rand write

 

Commit 8d47a0d8f7 was reported to have -28.6% of fio.write_bw_MBps when comparing to v5.1-rc7.

 

Correlated commits

8d47a0d8f7

btrfs: Do mandatory tree block check before submitting bio

branch

linus/master

report

[LKP] [btrfs] 8d47a0d8f7: fio.write_bw_MBps -28.6% regression

test scenario

disk: 2pmem

fs: btrfs

runtime: 200s

nr_task: 50%

time_based: tb

rw: randwrite

bs: 4k

ioengine: libaio

test_size: 100G

cpufreq_governor: performance

test machine

lkp-hsw-ep2

status

Not merged at v5.1,  no response from author yet

 

  1. netperf.Throughput_Mbps

Netperf is a benchmark that can be used to measure the performance of many different types of networking. It provides tests for both unidirectional throughput, and end-to-end latency.

  1. scenario: TCP_STREAM test on localhost

 

Commit 8b27dae5a2 was reported to have -25.7% of netperf.Throughput_Mbps when comparing to v5.0.

 

Correlated commits

8b27dae5a2

tcp: add one skb cache for rx

branch

linux-next/master

report

[LKP] [tcp] 8b27dae5a2: netperf.Throughput_Mbps -25.7% regression

test scenario

ip: ipv4

runtime: 900s

nr_threads: 200%

cluster: cs-localhost

test: TCP_STREAM

cpufreq_governor: performance

test machine

lkp-skl-fpga01

status

Not merged at v5.1, no response from author yet

 

  1. vm-scalability.median

vm-scalability exercises functions and regions of the mm subsystem of the Linux kernel. We tested on multiple machines such as HSW EP server, during which we reported improvement on one test scenario.

Scenario: shm-pread-rand test

 

Commit 27eb9d500d was reported to have -19.4% of vm-scalability.median when comparing to v5.1-rc2.

 

Correlated commits

27eb9d500d

vfs: Convert ramfs, shmem, tmpfs, devtmpfs, rootfs to use the new mount API

branch

dhowells-fs/mount-api-viro

report

[LKP] [vfs] 27eb9d500d: vm-scalability.median -19.4% regression

test scenario

runtime: 300s

size: 16G

test: shm-pread-rand

cpufreq_governor: performance

test machine

lkp-bdw-ep2

status

Not merged at v5.1, no response from author yet

 

  1. Latest Release Performance Comparing

 

This session gives some information about the performance difference among different kernel releases, especially between v5.1 and v5.0. There are 50+ performance benchmarks running in 0-Day CI, and we selected 9 benchmarks which historically showed the most regressions/improvements reported by 0-Day CI. Some typical configuration/parameters are used to run the test. For some of the regressions from the comparing, 0-Day did not successfully bisect it thus no related report sent out during the release development period, but it is still worth to check.  

 

In the following figures, the value on the Y-axis is the relative performance number. We used the v5.0 data as the base (performance number is 100).

  1. test suite: vm-scalability

vm-scalability exercises functions and regions of the mm subsystem of the Linux kernel. Below 4 tests show the typical test results.

 

vm-scalability Test 1

vm-scalability Test 2

 

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                           

 

vm-scalability Test 1

vm-scalability Test 2

test machine

model: Haswell-EX

brand: Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz

cpu_number: 144

memory: 512G

model: Broadwell-EP

brand: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz

cpu_number: 88

memory: 128G

runtime

No requirement

300s

nr_task

32

No requirement

boot_params

bp1_memmap: 120G!4G

bp2_memmap: 120G!130G

bp3_memmap: 120G!258G

bp4_memmap: 120G!386G

No requirement

disk

nr_pmem: 4

No requirement

swap

priority: 1

No requirement

transparent_hugepage

thp_enabled: never

thp_defrag: never

No requirement

vm-scalability test parameter

test case: swap-w-seq

test case: small-allocs

performance summary

vm-scalability.throughput on kernel v5.1 has -9.1% regression when comparing to v5.0

vm-scalability.throughput on kernel v5.1 has 12.48% improvement when comparing to v5.0

 

  1. test suite: will-it-scale

Will-it-scale takes a test case and runs it from 1 through to n parallel copies to see if the test case will scale. It builds both process and threads based tests in order to see any differences between the two.

 

will-it-scale  Test 1

Will-it-scale Test 2

will-it-scale  Test 3

will-it-scale  Test 4

 

Here are the parameters and performance test summary for above tests:                                                                                                                                                                                                                          

 

will-it-scale Test 1

will-it-scale Test 2

will-it-scale Test 3

will-it-scale Test 4

test machine

model: Knights Mill

brand: Intel(R) Xeon Phi(TM) CPU 7295 @ 1.50GHz

cpu_number: 288

memory: 80G

model: Knights Mill

brand: Intel(R) Xeon Phi(TM) CPU 7295 @ 1.50GHz

cpu_number: 288

memory: 80G

model: Broadwell-EP

brand: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz

cpu_number: 88

memory: 128G

model: Ivy Bridge

brand: Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz

cpu_number: 4

memory: 4G

nr_task

100%

100%

100%

1

will-it-scale test parameter

mode: thread

test case: dup1

mode: process

test case: pread1

mode: process
test: futex4

mode: process
test: poll2

summary

will-it-scale.per_thread_ops on kernel v5.1 has -8.85% regression when comparing to v5.0

will-it-scale.per_thread_ops on kernel v5.1 has -8.56% regression when comparing to v5.0

will-it-scale.per_process_ops on kernel v5.1 has 3.54% improvement when comparing to v5.0

will-it-scale.per_process_ops on kernel v5.1 has 11.13% improvement when comparing to v5.0

 

  1. test suite: unixbench

UnixBench is a system benchmark to provide a basic indicator of the performance of a Unix-like system.

 

Unixbench Test 1

 

 

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                           

 

Unixbench Test 1

test machine

model: Skylake-SP

cpu_number: 192

memory: 256G

runtime

300

nr_task

30%

unixbench test parameter

test case: shell8

performance summary

unixbench.score on kernel v5.1 has 18.6% improvement when comparing to v5.0


 

 

  1. test suite: reaim

reaim updates and improves the existing Open Source AIM 7 benchmark. aim7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of a multiuser system.

 

reaim Test 1                             

reaim Test 2


 

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                                          

 

reaim Test 1

reaim Test 2

test machine

model: Haswell-EP

brand: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz

cpu_number: 72

memory: 256G

model: Haswell-EX

brand: Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz

cpu_number: 144

memory: 512G

runtime

300s

No requirement

nr_task

100%

100%

reaim test parameter

nr_job: 10000

test case: short

iterations: 20

nr_job: 3000

test: dbase

performance  summary

reaim.jobs_per_min on kernel v5.1 almost the same with v5.0

reaim.jobs_per_min on kernel v5.1 has -3.09% regression when comparing to v5.0

 

  1. test suite: pigz

pigz, which stands for Parallel Implementation of GZip, is a fully functional replacement for gzip that exploits multiple processors and multiple cores to the hilt when compressing data.

pigz Test 1


 

 

Here are the test configuration and performance test summary for above tests:          

 

 

pigz Test 1

test machine

model: Ivy Bridge

nr_node: 1

nr_cpu: 4

memory: 4G

nr_threads

100%

pigz Test parameter

blocksize: 512K

performance  summary

pigz.throughput on kernel v5.1 almost the same with v5.0

           

  1. test suite: netperf

Netperf is a benchmark that can be used to measure the performance of many different types of networking. It provides tests for both unidirectional throughput, and end-to-end latency.

 

netperf Test 1

netperf Test 2



 

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                                          

 

netperf Test 1

netperf Test 2

test machine

model: Ivy Bridge

brand: Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz

cpu_number: 4

memory: 8G

model: Broadwell-EP

brand: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz

cpu_number: 88

memory: 128G

disable_latency_stats

1

1

set_nic_irq_affinity

1

1

runtime

900s

900s

nr_threads

200%

25%

ip

ipv4

ipv4

netperf test parameter

test case: TCP_MAERTS

test case: TCP_MAERTS

performance  summary

netperf.Throughput_Mbps on kernel v5.1 has -3.76% regression when comparing to v5.0

netperf.Throughput_Mbps on kernel v5.1 almost the same with v5.0

 

  1. test suite: hackbench

Hackbench is both a benchmark and a stress test for the Linux kernel scheduler. It's  main job is to create a specified number of pairs of schedulable entities (either threads or traditional processes) which communicate via either sockets or pipes and time how long  it takes for each pair to send data back and forth.

hackbench Test 1


 

hackbench Test 2


 

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                                          

 

hackbench Test 1

hackbench Test 2

test machine

model: Haswell-EP

brand: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz

cpu_number: 72

memory: 128G

model: Broadwell-EP

brand: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz

cpu_number: 88

memory: 64G

disable_latency_stats

1

1

nr_threads

1600%

50%

hackbench test parameter

iterations: 18

mode: threads

ipc: socket

mode: process

ipc: pipe

performance  summary

hackbench.throughput on kernel v5.1 has -3.75% regression when comparing to v5.0

hackbench.throughput on kernel v5.1 has 4.75% improvement when comparing to v5.0

 

  1. test suite: fio

Fio was originally written to save me the hassle of writing special test case programs when I wanted to test a specific workload, either for performance reasons or to find/reproduce a bug.

fio Test 1

 

fio Test 2

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                                          

 

fio Test 1

fio Test 2

test machine

model: Ivy Bridge-EP

brand: Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz

cpu_number: 40

memory: 384G

model: Haswell-EP

brand: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz

cpu_number: 72

memory: 256G

disk

1HDD

2pmem

boot_params

No requirement

bp1_memmap: 104G!4G

bp2_memmap: 104G!132G

runtime

300s

200s

file system

btrfs

btrfs

test_size

128G

No requirement

nr_task

1

50%

time_based

No requirement

tb

fio test parameter

fio-setup-basic:

 rw: write

 bs: 4k

 ioengine: sync

fio-setup-basic:

 rw: randread

 bs: 2M

 ioengine: libaio

 test_size: 100G

performance  summary

Fio.write_iops on kernel v5.1 almost the same with v5.0

fio.read_bw_MBps on kernel v5.1 has 5.7% improvement when comparing to v5.0

 

  1. test suite: ebizzy

ebizzy is designed to generate a workload resembling common web application server workloads. It is highly threaded, has a large in-memory working set, and allocates and deallocates memory frequently.

ebizzy Test 1

 

 

Here are the test configuration and performance test summary for above test:                                                                                                                                                                                                                          

 

ebizzy Test 1

test machine

model: Broadwell-EP

nr_node: 2

nr_cpu: 88

memory: 64G

nr_threads

200%

iterations

100x

ebizzy test parameter

duration: 10s

performance  summary

ebizzy.throughput on kernel v5.1 almost the same with v5.0

 

  1. Test Machines

  1. IVB Desktop

model

Ivy Bridge

brand

Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz

cpu number

8

memory

16G

 

model

Ivy Bridge

brand

Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz

cpu number

4

memory

8G

 

  1. SKL Desktop

model

Skylake

brand

Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz

cpu number

80

memory

64G

 

  1. BDW EP

model

Broadwell-EP

brand

Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz

cpu number

88

memory

128G

 

  1. HSW EP

model

Haswell-EP

brand

Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz

cpu number

72

memory

128G

 

  1. IVB EP

model

Ivy Bridge-EP

brand

Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz

cpu number

40

memory

384G

 

model

Ivytown Ivy Bridge-EP

brand

Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz

cpu number

48

memory

64G

 

  1. HSX EX

model

Brickland Haswell-EX

brand

Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz

cpu number

144

memory

512G