Sorry, you need to enable JavaScript to visit this website.

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 (v5.7)

BY Rong Chen ON Jun 12, 2020
  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 v5.7 release candidates

  • Section 3, test parameter description

  • Section 4, captured regressions and improvements by shift-left testing during developers’ and maintainers’ tree during v5.7 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 v5.7 Release Test

The 5.7 release of the Linux kernel was on May 31, 2020. Headline features in 5.7 include x86 split-lock detection, thermal-pressure management, frequency invariance in the load-tracking code, coexistence between BPF and realtime preemption, support for BPF security hook programs (formerly called the KRSI security module), a new, Microsoft-blessed exFAT filesystem implementation, and more. The final patch to be merged was this one deprecating the long-standing 80-column limit for kernel source. See the KernelNewbies 5.7 page for lots of details.

 

0-Day CI monitored the release closely to trace down the performance status on IA platform. 0-Day observed 5 regressions and 5 improvements during feature development phase for v5.7. 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 5 regressions and 5 improvements during the feature development phase for v5.7, which is in the time frame from v5.7-rc1 to v5.7 release. 

Test Indicator

Report

Test Scenario

Test Machine

Development Base

Status

aim7.jobs-per-min

[LKP] [xfs] 542951592c: 22.9% improvement

disk: 4BRD_12G

md: RAID1

fs: xfs

test: disk_cp

load: 3000

cpufreq_governor: performance

lkp-csl-2ap2

v5.6-rc4

merged at v5.7-rc1

aim7.jobs-per-min

[LKP] [btrfs] c75e839414: aim7.jobs-per-min -9.1% regression

disk: 4BRD_12G

md: RAID0

fs: btrfs

test: disk_wrt

load: 1500

cpufreq_governor: performance

lkp-csl-2ap2

v5.6-rc7

merged at v5.7-rc1, no response from author yet

fxmark.ssd_f2fs_DRBL_1_directio.works/sec

[LKP] [perf/core] 90c91dfb86: 18.2% improvement

disk: 1SSD

media: ssd

test: DRBL

fstype: f2fs

directio: directio

cpufreq_governor: performance

lkp-csl-2ap1

v5.6-rc6

merged at v5.7-rc1

netperf.Throughput_total_tps

[LKP] [net] a6211caa63: 160.7% improvement

ip: ipv4

runtime: 300s

nr_threads: 200%

cluster: cs-localhost

test: UDP_RR

cpufreq_governor: performance

lkp-skl-fpga01

v5.7-rc5

merged at v5.7-rc7

reaim.jobs_per_min

[LKP] [sched/fair] 070f5e860e: -10.5% regression

runtime: 300s

nr_task: 100%

test: five_sec

cpufreq_governor: performance

lkp-ivb-d04

v5.6-rc3

merged at v5.7-rc1, no response from author yet

stress-ng.fiemap.ops_per_sec

[LKP] [ext4] d3b6f23f71: -60.5% regression

nr_threads: 10%

disk: 1HDD

testtime: 1s

class: os

cpufreq_governor: performance

fs: ext4

lkp-csl-2sp5

v5.6-rc4

merged at v5.7-rc1, the author is working on it.

stress-ng.mmapfork.ops_per_sec

[LKP] [sched/fair] 6c8116c914: -38.0% regression

nr_threads: 100%

disk: 1HDD

testtime: 1s

class: scheduler

cpufreq_governor: performance

sc_pid_max: 4194304

lkp-bdw-ep6

v5.6-rc4

merged at v5.7-rc1, the author is working on it.

vm-scalability.median

[LKP] [sched/fair] c32b430829: 5.8% improvement

runtime: 300s

size: 8T

test: anon-w-seq

cpufreq_governor: performance

lkp-csl-2ap4

v5.6-rc4

merged at v5.7-rc1

will-it-scale.per_process_ops

[LKP] [non] 2aa3847085: -32.0% regression

test: open2

cpufreq_governor: performance

lkp-ivb-d04

v5.6-rc1

merged at v5.7-rc1, but fix commit 5bd73286d5 merged at v5.7-rc1 too

will-it-scale.per_thread_ops

[LKP] [futex] 222993395e: 1507.5% improvement

nr_task: 16

mode: thread

test: futex1

cpufreq_governor: performance

lkp-csl-2ap2

v5.6-rc4

merged at v5.7-rc1

    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: disk_wrt test on btrfs

 

Commit c75e839414 was reported to have -8.9% of aim7.jobs-per-min when comparing to v5.6-rc7. It was merged to mainline at v5.7-rc1.

 

Correlated commits

c75e839414

btrfs: kill the subvol_srcu

branch

kdave-btrfs-devel/misc-5.7

report

[LKP] [btrfs] c75e839414: -8.9% regression

test scenario

disk: 4BRD_12G

md: RAID0

fs: btrfs

test: disk_wrt

load: 1500

cpufreq_governor: performance

test machine

lkp-csl-2ap2

status

merged at v5.7-rc1, no response from author yet

 

    1. reaim.jobs_per_min

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.

 

    1. scenario: five_sec test

 

Commit 070f5e860e was reported to have -10.5% of reaim.jobs_per_min when comparing to v5.6-rc3. It was merged to mainline at v5.7-rc1.

 

Correlated commits

070f5e860e

sched/fair: Take into account runnable_avg to classify group

branch

linux-next/master

report

[LKP] [sched/fair] 070f5e860e: -10.5% regression

test scenario

runtime: 300s

nr_task: 100%

test: five_sec

cpufreq_governor: performance

test machine

lkp-ivb-d04

status

merged at v5.7-rc1, no response from author yet

 

  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.

    1. Scenario: thread futex1

 

 

Commit 222993395e was reported to have 1507.5% of will-it-scale.per_thread_ops when comparing to v5.6-rc4. It was merged to mainline at v5.7-rc1.

 

Correlated commits

222993395e

futex: Remove pointless mmgrap() + mmdrop()

branch

tip/locking/core

report

[LKP] [futex] 222993395e: 1507.5% improvement

test scenario

nr_task: 16

mode: thread

test: futex1

cpufreq_governor: performance

test machine

lkp-csl-2ap2

status

merged at v5.7-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 v5.7 release cycle, 0-Day CI had reported 17 major performance regressions and 12 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 in the report summary section.

    1. Report Summary

0-Day CI had reported 17 performance regressions and 12 improvements by doing shift-left testing on developer and maintainer repos.

 

Test Indicator

Mail

Test Scenario

Test Machine

Status

aim7.jobs-per-min

[LKP] [pipe] f2af7d90e2: 1827.5% improvement

disk: 4BRD_12G

md: RAID0

fs: xfs

test: disk_src

load: 3000

cpufreq_governor: performance

lkp-skl-2sp7

currently not merged

aim7.jobs-per-min

[LKP] [btrfs] ba60dcc6ad: 17.2% improvement

disk: 4BRD_12G

md: RAID0

fs: btrfs

test: sync_disk_rw

load: 20

cpufreq_governor: performance

lkp-skl-2sp7

currently not merged

aim7.jobs-per-min

[LKP] [xfs] a5949d3fae: -33.6% regression

disk: 1BRD_48G

fs: xfs

test: sync_disk_rw

load: 600

cpufreq_governor: performance

lkp-ivb-2ep1

currently not merged, no response from author yet

filebench.sum_bytes_mb/s

[LKP] [xfs] ef6d7398f5: -47.5% regression

disk: 1HDD

fs: xfs

test: fileserver.f

cpufreq_governor: performance

lkp-hsw-d01

currently not merged, no response from author yet

filebench.sum_bytes_mb/s

[LKP] [f2fs] da55a3ad36: -44.4% regression

disk: 1HDD

fs: f2fs

test: createfiles.f

cpufreq_governor: performance

lkp-hsw-d01

currently not merged, no response from author yet

fio.read_bw_MBps

[LKP] c9d32d8e86: 94.9% improvement

disk: 2pmem

fs: xfs

runtime: 200s

nr_task: 50%

time_based: tb

rw: randrw

bs: 4k

ioengine: sync

test_size: 200G

cpufreq_governor: performance

lkp-csl-2sp6

currently not merged

fio.write_bw_MBps

[LKP] 567fcac239: -7.8% regression

runtime: 300s

disk: 1SSD

fs: ext4

nr_task: 100%

test_size: 128G

rw: write

bs: 4k

ioengine: sync

direct: direct

cpufreq_governor: performance

lkp-bdw-ex2

currently not merged, no response from author yet

fio.write_bw_MBps

[LKP] [btrfs] ca7f34487a: 136.4% improvement

runtime: 300s

disk: 1SSD

fs: btrfs

nr_task: 100%

test_size: 128G

rw: randwrite

bs: 4k

ioengine: sync

cpufreq_governor: performance

lkp-csl-2ap1

currently not merged

fxmark.ssd_btrfs_MWUL_12_bufferedio.works/sec

[LKP] [btrfs] 986b7387a1: -81.0% regression

disk: 1SSD

media: ssd

test: MWUL

fstype: btrfs

directio: bufferedio

cpufreq_governor: performance

lkp-csl-2ap1

currently not merged, no response from author yet

hackbench.throughput

[LKP] 9c921492e9: -33.4% regression

runtime: 20s

nr_threads: 50%

mode: process

ipc: pipe

cpufreq_governor: performance

lkp-csl-2sp7

currently not merged, no response from author yet

netperf.Throughput_total_tps

[LKP] [percpu_counter] 467724688b: 13.0% improvement

ip: ipv4

runtime: 300s

nr_threads: 16

cluster: cs-localhost

test: TCP_CRR

cpufreq_governor: performance

lkp-skl-2sp7

currently not merged

netperf.Throughput_total_tps

[LKP] [hwrng] a583ed310b: 9.7% improvement

ip: ipv4

runtime: 300s

nr_threads: 16

cluster: cs-localhost

test: TCP_CRR

cpufreq_governor: performance

lkp-skl-2sp7

currently not merged

netperf.Throughput_total_tps

[LKP] [net] a6211caa63: 160.7% improvement

ip: ipv4

runtime: 300s

nr_threads: 200%

cluster: cs-localhost

test: UDP_RR

cpufreq_governor: performance

lkp-skl-fpga01

merged at v5.7-rc7

phoronix-test-suite.glmark2.800x600.score

[LKP] [drm/mgag200] e44e907dd8: -64.9% regression

need_x: true

test: glmark2-1.1.0

cpufreq_governor: performance

lkp-nhm-2ep1

currently not merged, no response from author yet

pmbench.latency.ns.average

[LKP] [swap] 9c9c831c31: -33.9% improvement

runtime: 1800s

mode: raw

nr_dev: 2

nr_part: 2

part_type: S

all: 1

thp_enabled: never

thp_defrag: never

nr_processes: 16

nr_threads: 1

pattern: uniform

ratio: 80

cold: 1

initialize: 1

setsize: 4800M

cpu_node_bind: even

cpufreq_governor: performance

debug-setup: yhss9

sc_numa_balancing: 0

sc_swappiness: 100

lkp-csl-2sp4

currently not merged

reaim.jobs_per_min

[LKP] [mm/lru] 94a3fe59f5: -37.0% regression

runtime: 300s

nr_task: 1000t

test: mem_rtns_1

cpufreq_governor: performance

lkp-ivb-2ep1

currently not merged, no response from author yet

stress-ng.af-alg.ops_per_sec

[LKP] [crypto] beeb460cd1: 17454.7% improvement

nr_threads: 100%

disk: 1HDD

testtime: 1s

class: cpu

cpufreq_governor: performance

lkp-bdw-ep6

currently not merged

stress-ng.futex.ops_per_sec

[LKP] [uswsusp] 08e6583083: -11.4% regression

nr_threads: 100%

disk: 1HDD

testtime: 1s

class: scheduler

cpufreq_governor: performance

sc_pid_max: 4194304

lkp-csl-2sp5

currently not merged, no response from author yet

stress-ng.icache.ops_per_sec

[LKP] [ima] 8eb613c0b8: -84.2% regression

nr_threads: 100%

disk: 1HDD

testtime: 30s

class: cpu-cache

cpufreq_governor: performance

lkp-csl-2sp5

currently not merged, no response from author yet

stress-ng.stream.ops_per_sec

[LKP] [kernfs] ea7c5fc39a: 11827.2% improvement

nr_threads: 100%

disk: 1HDD

testtime: 1s

class: cpu-cache

cpufreq_governor: performance

lkp-csl-2sp5

currently not merged, the improvement shouldn’t be such huge. 

vm-scalability.median

[LKP] 5fba1f38eb: -10.6% regression

runtime: 300

thp_enabled: never

thp_defrag: always

nr_task: 8

nr_pmem: 1

test: swap-w-rand-mt

bp_memmap: 96G!18G

cpufreq_governor: performance

lkp-csl-2sp6

currently not merged, no response from author yet

vm-scalability.median

[LKP] [mm] 134c8b410f: -7.9% regression

runtime: 300s

size: 8T

test: anon-cow-seq

cpufreq_governor: performance

lkp-cfl-e1

currently not merged, no response from author yet

vm-scalability.median

[LKP] [thp] fd80f22b29: 10.3% improvement

runtime: 300s

size: 8T

test: anon-cow-seq

cpufreq_governor: performance

lkp-cfl-e1

currently not merged

vm-scalability.median

[LKP] [mm] 9266baf8ee: -17.0% regression

runtime: 300s

size: 8T

test: anon-w-seq-hugetlb

cpufreq_governor: performance

lkp-cfl-e1

currently not merged, no response from author yet

will-it-scale.per_process_ops

[LKP] [x86/mce] 1de08dccd3: -14.1% regression

nr_task: 100%

mode: process

test: malloc1

cpufreq_governor: performance

lkp-knm01

currently not merged, the author doubt it.

will-it-scale.per_process_ops

[LKP] a530431919: 2097.3% improvement

nr_task: 100%

mode: process

test: mmap2

cpufreq_governor: performance

lkp-hsw-4ex1

currently not merged, no response from author yet

will-it-scale.per_process_ops

[LKP] [eventfd] a4ef93e263: -2.2% regression

nr_task: 100%

mode: process

test: eventfd1

cpufreq_governor: performance

lkp-skl-fpga01

currently not merged, no response from author yet

will-it-scale.per_thread_ops

[LKP] [mm] 5ae8a9d7c8: -2.1% regression

nr_task: 100%

mode: thread

test: page_fault1

cpufreq_governor: performance

lkp-ivb-d01

currently not merged, no response from author yet

will-it-scale.per_thread_ops

[LKP] [firmware] 88a5883981: -66.8% regression

nr_task: 100%

mode: thread

test: sched_yield

cpufreq_governor: performance

lkp-skl-fpga01

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 btrfs

 

Commit ba60dcc6ad was reported to have 17.2% of aim7.jobs-per-min when comparing to v5.7-rc5.

 

Correlated commits

ba60dcc6ad

btrfs: speed up btrfs_get_##bits helpers

branch

linux-next/master

report

[LKP] [btrfs] ba60dcc6ad: 17.2% improvement

test scenario

disk: 4BRD_12G

md: RAID0

fs: btrfs

test: sync_disk_rw

load: 20

cpufreq_governor: performance

test machine

lkp-skl-2sp7

status

currently not merged

 

    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.

 

      1. scenario: anon-w-seq-hugetlb test

 

 

Commit 9266baf8ee was reported to have -17.0% of vm-scalability.median when comparing to v5.7-rc1.

 

Correlated commits

9266baf8ee

mm: Optimized hugepage zeroing & copying from user

branch

linux-review/UPDATE-20200501-184514/Prathu-Baronia/mm-Optimized-hugepage-zeroing-copying-from-user/20200415-041712

report

[LKP] [mm] 9266baf8ee: -17.0% regression

test scenario

runtime: 300s

size: 8T

test: anon-w-seq-hugetlb

cpufreq_governor: performance

test machine

lkp-cfl-e1

status

currently not merged, no response from author yet

 

    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: randwrite on btrfs

 

Commit ca7f34487a was reported to have 136.4% of fio.write_bw_MBps when comparing to v5.7-rc5.

 

Correlated commits

ca7f34487a

btrfs: Speed up buffered writes exceeding memory size

branch

linux-review/Nikolay-Borisov/btrfs-Speed-up-buffered-writes-exceeding-memory-size/20200519-202239

report

[LKP] [btrfs] ca7f34487a: 136.4% improvement

Test Scenario

runtime: 300s

disk: 1SSD

fs: btrfs

nr_task: 100%

test_size: 128G

rw: randwrite

bs: 4k

ioengine: sync

cpufreq_governor: performance

Test Machine

lkp-csl-2ap1

status

currently not merged

 

  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.

    1. Scenario: thread sched_yield

 

 

Commit 88a5883981 was reported to have -66.8% of will-it-scale.per_thread_ops when comparing to v5.7-rc4.

 

Correlated commits

88a5883981

firmware: arm_scmi/mailbox: ignore notification for tx done using irq

branch

linux-review/joe_zhuchg-126-com/firmware-arm_scmi-mailbox-ignore-notification-for-tx-done-using-irq/20200509-220000

report

[LKP] [firmware] 88a5883981: -66.8% regression

test scenario

nr_task: 100%

mode: thread

test: sched_yield

cpufreq_governor: performance

test machine

lkp-skl-fpga01

status

currently not merged

 

  1. Latest Release Performance Comparing

 

This session gives some information about the performance difference among different kernel releases, especially between v5.7 and v5.6. 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 comparison, 0-Day did not successfully bisect it thus no related report sent out during the release development period, but it is still worth checking. The root cause to cause the regressions won’t be covered in this session. 

 

In the following figures, the value on the Y-axis is the relative performance number. We used the v5.6 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 2 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: Skylake

cpu_number: 104

memory: 192G

model: Broadwell-EX

brand: Intel(R) Xeon(R) CPU E7-8890 v4 @ 2.20GHz

cpu_number: 160

memory: 256G

runtime

300s

300s

size

No requirement

1T

vm-scalability test parameter

test case: small-allocs

test case: lru-shm

performance summary

vm-scalability.throughput on kernel v5.7 has -16.43% regression when comparing to v5.6

vm-scalability.throughput on kernel v5.6 has 9.83% improvement when comparing to v5.5

 

    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: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

model: Broadwell-EP

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

cpu_number: 88

memory: 128G

nr_task

100%

100%

50%

50%

will-it-scale test parameter

mode: process

test case: unlink1

mode: thread

test: pwrite1

mode: process

test: mmap2

mode: thread

test: futex2

summary

will-it-scale.per_process_ops on kernel v5.7 has -8.99% regression when comparing to v5.6

will-it-scale.per_thread_ops on kernel v5.7 has 4.97% improvement when comparing to v5.6

will-it-scale.per_process_ops on kernel v5.7 has 5.38% improvement when comparing to v5.6

will-it-scale.per_thread_ops on kernel v5.7 has 917.97% improvement when comparing to v5.6

 

    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

Unixbench Test 2

 

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

 

Unixbench Test 1 

Unixbench Test 2

test machine

model: Coffee Lake

brand: Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz

cpu_number: 16

memory: 32G

model: Coffee Lake

brand: Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz

cpu_number: 16

memory: 32G

runtime

300s

300s

nr_task

1

30%

unixbench test parameter

test case: fsdisk

test case: syscall

performance summary

unixbench.score on kernel v5.7 has -4.3% regression when comparing to v5.6

unixbench.score on kernel v5.7 has 7.64% improvement when comparing to v5.6

 

    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 

eaim Test 2

test machine

model: Haswell-EX

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

cpu_number: 144

memory: 512G

model: Haswell-EP

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

cpu_number: 72

memory: 256G

runtime

No requirement

300s

nr_task

1600%

1600%

reaim test parameter

test case: high_systime

iterations: 4

test: aim9

performance  summary

reaim.jobs_per_min on kernel v5.7 has -12.4% regression when comparing to v5.6

reaim.jobs_per_min on kernel v5.7 has 9.35% improvement when comparing to v5.6

                                                                                                                                                                                                     

    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

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

cpu_number: 4

memory: 8G

nr_threads

100%

pigz Test parameter

blocksize: 512K

performance  summary

pigz.throughput on kernel v5.7 is almost the same as that in v5.6

            

  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: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

disable_latency_stats

1

1

set_nic_irq_affinity

1

1

runtime

300s

300s

nr_threads

200%

200%

ip

ipv4

ipv4

netperf test parameter

test case: TCP_RR

test case: UDP_RR

performance  summary

netperf.Throughput_tps on kernel v5.7 is almost the same as that in v5.6

netperf.Throughput_tps on kernel v5.7 has 309.75% improvement when comparing to v5.6

 

    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

hackbench Test 3

hackbench Test 4

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

 

hackbench Test 1 

hackbench Test 2 

hackbench Test 3 

hackbench Test 4 

test machine

model: Haswell-EP

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

cpu_number: 72

memory: 256G

model: Haswell-EP

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

cpu_number: 72

memory: 256G

model: Broadwell-EP

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

cpu_number: 88

memory: 128G

model: Haswell-EP

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

cpu_number: 72

memory: 256G

disable_latency_stats

1

1

1

1

nr_task

50%

50%

1600%

1600%

unixbench test parameter

iterations: 12

mode: threads

ipc: socket

iterations: 12

mode: process

ipc: pipe

mode: threads

ipc: pipe

iterations: 18

mode: process

ipc: pipe

performance summary

hackbench.throughput on kernel v5.7 is almost the same as that in v5.6

hackbench.throughput on kernel v5.7 is almost the same as that in v5.6

hackbench.throughput on kernel v5.7 has 7.31% improvement when comparing to v5.6

hackbench.throughput on kernel v5.7 has 13.58% improvement when comparing to v5.6

            

    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: Cascade Lake

brand: Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz

cpu_number: 96

memory: 256G

model: Cascade Lake

brand: Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz

cpu_number: 96

memory: 192G

runtime

200s

300s

file system

xfs

xfs

mount_option

dax

No requirement

disk

2pmem

1HDD

boot_params

bp1_memmap: 104G!8G

bp2_memmap: 104G!132G

No requirement

nr_task

50%

1

time_based

tb

No requirement

fio test parameter

fio-setup-basic:

  rw: randwrite

  bs: 4k

  ioengine: libaio

  test_size: 200G

fio-setup-basic:

  rw: write

  bs: 4k

  ioengine: sync

  test_size: 128G

performance  summary

fio.write_bw_MBps on kernel v5.7 has -98.04% regression when comparing to v5.6

fio.write_bw_MBps on kernel v5.7 has 51.59% improvement when comparing to v5.6

 

    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: Haswell

brand: Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz

cpu_number: 8

memory: 8G

nr_threads

200%

iterations

100x

ebizzy test parameter

duration: 10s

performance  summary

ebizzy.throughput on kernel v5.7 is almost the same as that in v5.6


 

  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 SP

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