jjzjj

GreatSQL vs MySQL性能测试来了,速围观~

GreatSQL 2023-03-28 原文
  • GreatSQL社区原创内容未经授权不得随意使用,转载请联系小编并注明来源。
  • GreatSQL是MySQL的国产分支版本,使用上与MySQL一致。

1.结论先行

  1. 无论ibp(innodb_buffer_pool_size)是否充足,MySQL的性能都远不如GreatSQL。

  2. MySQL的性能平均约为GreatSQL的70%(最高84.5%,最低61.7%)。

  3. 在ibp充分的情况下,随着并发数的增加,MySQL并没有表现出该有的性能提升,反倒掉头向下,可见还是不够稳定。

  4. 在ibp不够的情况下,GreatSQL开启thread pool性能有所提升;当ibp充足的情况下,区别就不大了。

MySQL vs GreatSQL性能数据对比

48G 96G 144G 192G avg
MySQL vs GreatSQL 0.6173 0.735 0.721 0.8449 0.7295

各数据库版本分别为

  • MySQL 8.0.30 MySQL Community Server - GPL

  • GreatSQL 8.0.25-16 GreatSQL, Release 16, Revision 8bb0e5af297

2.测试结果数据

2.1 ibp=48G

TPS 8th 16th 32th 64th 128th
GreatSQL-thdpool 969.16 1324.21 1661.57 2007.98 2331.4
GreatSQL 873.06 1146.85 1371.34 1509.8 1699.19
MySQL 686.14 846.5 915.15 1073.95 1439.29

P.S,后缀加上 thdpool 表示启用了thread pool

QPS 8th 16th 32th 64th 128th
GreatSQL-thdpool 19383.2 26484.14 33231.49 40159.56 46627.89
GreatSQL 17461.16 22937.14 27426.87 30196.02 33983.78
MySQL 13722.8 16929.94 18303.03 21479 28785.7

2.2 ibp=96G

TPS 8th 16th 32th 64th 128th
GreatSQL-thdpool 1074.57 1407.54 1706.35 2206.06 2810.39
GreatSQL 1013.2 1198.5 1546.53 2033.04 2419.47
MySQL 751.7 986.11 1218.87 1778.67 2065.69
QPS 8th 16th 32th 64th 128th
GreatSQL-thdpool 21491.46 28150.73 34127 44121.2 56207.88
GreatSQL 20264.04 23969.97 30930.56 40660.83 48389.42
MySQL 15034.11 19722.27 24377.47 35573.39 41313.8

2.3 ibp=144G

TPS 8th 16th 32th 64th 128th
GreatSQL-thdpool 1059.46 1422.72 1853.24 2710.31 3481.66
GreatSQL 857.28 1327.67 1767.78 2660.8 3148.06
MySQL 857.05 1149.79 2038.3 2516.41 2510.15
QPS 8th 16th 32th 64th 128th
GreatSQL-thdpool 21189.17 28454.3 37064.79 54206.13 69633.25
GreatSQL 17145.52 26553.48 35355.47 53215.89 62961.17
MySQL 17140.96 22995.73 40765.95 50328.29 50202.93

2.4 ibp=192G

TPS 8th 16th 32th 64th 128th
GreatSQL 1406.86 1316.02 2144.17 4114.55 3310.67
GreatSQL-thdpool 1391.2 1247.93 2085.81 4053.76 3113.97
MySQL 1367.31 2629.75 2940.51 2687.48 2797.06
QPS 8th 16th 32th 64th 128th
GreatSQL 28137.19 26320.43 42883.45 82291 66213.47
GreatSQL-thdpool 27823.9 24958.68 41716.16 81075.21 62279.48
MySQL 27346.18 52595.01 58810.18 53749.63 55941.29

2.5 GreatSQL不同ibp下的数据

GreatSQL
TPS 8th 16th 32th 64th 128th QPS 8th 16th 32th 64th 128th
GreatSQL-thdpool(48G) 969.16 1324.21 1661.57 2007.98 2331.4 GreatSQL-thdpool(48G) 19383.2 26484.14 33231.49 40159.56 46627.89
GreatSQL(48G) 873.06 1146.85 1371.34 1509.8 1699.19 GreatSQL(48G) 17461.16 22937.14 27426.87 30196.02 33983.78
GreatSQL-thdpool(96G) 1074.57 1407.54 1706.35 2206.06 2810.39 GreatSQL-thdpool(96G) 21491.46 28150.73 34127 44121.2 56207.88
GreatSQL(96G) 1013.2 1198.5 1546.53 2033.04 2419.47 GreatSQL(96G) 20264.04 23969.97 30930.56 40660.83 48389.42
GreatSQL-thdpool(144G) 1059.46 1422.72 1853.24 2710.31 3481.66 GreatSQL-thdpool(144G) 21189.17 28454.3 37064.79 54206.13 69633.25
GreatSQL(144G) 857.28 1327.67 1767.78 2660.8 3148.06 GreatSQL(144G) 17145.52 26553.48 35355.47 53215.89 62961.17
GreatSQL(192G) 1406.86 1316.02 2144.17 4114.55 3310.67 GreatSQL(192G) 28137.19 26320.43 42883.45 82291 66213.47
GreatSQL-thdpool(192G) 1391.2 1247.93 2085.81 4053.76 3113.97 GreatSQL-thdpool(192G) 27823.9 24958.68 41716.16 81075.21 62279.48

2.6 MySQL不同ibp下的数据

MySQL
TPS 8th 16th 32th 64th 128th QPS 8th 16th 32th 64th 128th
MySQL(48G) 686.14 846.5 915.15 1073.95 1439.29 MySQL(48G) 13722.8 16929.94 18303.03 21479 28785.7
MySQL(96G) 751.7 986.11 1218.87 1778.67 2065.69 MySQL(96G) 15034.11 19722.27 24377.47 35573.39 41313.8
MySQL(144G) 857.05 1149.79 2038.3 2516.41 2510.15 MySQL(144G) 17140.96 22995.73 40765.95 50328.29 50202.93
MySQL(192G) 1367.31 2629.75 2940.51 2687.48 2797.06 MySQL(192G) 27346.18 52595.01 58810.18 53749.63 55941.29

测试环境&测试模式

3.1 测试工具

sysbench

/usr/local/bin/sysbench --version
sysbench 1.1.0

P.S,该版本是楼方鑫修改后的,增加了99.9%的RT统计值,例如:

[ 1s ] thds: 128 tps: 10285.06 qps: 208112.71 (r/w/o: 145769.21/41646.36/20697.15) lat (ms,99%,99%,99.9%): 24.83/24.83/28.67 err/s: 0.00 reconn/s: 0.00
[ 2s ] thds: 128 tps: 9968.88 qps: 199013.18 (r/w/o: 139399.13/39676.28/19937.76) lat (ms,99%,99%,99.9%): 20.00/20.00/24.38 err/s: 0.00 reconn/s: 0.00
[ 3s ] thds: 128 tps: 10214.11 qps: 204613.28 (r/w/o: 143162.59/41022.47/20428.22) lat (ms,99%,99%,99.9%): 19.29/19.29/23.10 err/s: 0.00 reconn/s: 0.00
[ 4s ] thds: 128 tps: 10227.68 qps: 204402.77 (r/w/o: 143127.62/40819.79/20455.37) lat (ms,99%,99%,99.9%): 17.95/17.95/20.00 err/s: 0.00 reconn/s: 0.00
[ 5s ] thds: 128 tps: 10466.08 qps: 209233.51 (r/w/o: 146497.06/41804.30/20932.15) lat (ms,99%,99%,99.9%): 19.29/19.29/21.11 err/s: 0.00 reconn/s: 0.00

3.2 测试模式

  • 利用sysbench生成64个表,每个表1250万条记录。
  • 数据库总大小约191G。
  • sysbench采用 oltp_read_write 模式。
  • innodb_flush_method = O_DIRECT_NO_FSYNC。
  • GreatSQL在需要时才开启thread pool,MySQL不支持thread pool。
  • 默认关闭InnoDB PQ。
  • 因为没有额外测试机,所以采用本地socket方式连接,顺便关闭网络监听设置。
  • 测试资源有限,所以只测试单机模式,没有开启MGR。

3.3 测试机硬件配置

  • 最大物理内存:376G,但数据库分配IBP时分别为48G、96G、144G、192G,没有将物理内存全部耗尽。
  • 磁盘:Dell NVMe SSD
$ nvme list | grep nvme1
/dev/nvme1n1          90L0A019TAHR        
Dell Express Flash CD5 3.84T SFF         1     
2.86  TB /   3.84  TB    512   B +  0 B   1.1.1
  • 文件系统、ioscheduler
$ df -hT | grep nvme1
/dev/nvme1n1p1 xfs       3.5T  2.9T  706G  81% /data_nvme1n1p1

$ cat /sys/block/nvme1n1/queue/scheduler
[none] mq-deadline kyber bfq
  • CPU
Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              176
On-line CPU(s) list: 0-175
Thread(s) per core:  2
Core(s) per socket:  22
Socket(s):           4
NUMA node(s):        4
Vendor ID:           GenuineIntel
BIOS Vendor ID:      Intel
CPU family:          6
Model:               85
Model name:          Intel(R) Xeon(R) Gold 6238 CPU @ 2.10GHz
BIOS Model name:     Intel(R) Xeon(R) Gold 6238 CPU @ 2.10GHz
Stepping:            7
CPU MHz:             2800.924
CPU max MHz:         3700.0000
CPU min MHz:         1000.0000
BogoMIPS:            4200.00
L1d cache:           32K
L1i cache:           32K
L2 cache:            1024K
L3 cache:            30976K
NUMA node0 CPU(s):   0,4,8,12,16,20,24,28,32,36,40,44,48,52,56,60,64,68,72,76,80,84,88,92,96,100,104,108,112,116,120,124,128,132,136,140,144,148,152,156,160,164,168,172
NUMA node1 CPU(s):   1,5,9,13,17,21,25,29,33,37,41,45,49,53,57,61,65,69,73,77,81,85,89,93,97,101,105,109,113,117,121,125,129,133,137,141,145,149,153,157,161,165,169,173
NUMA node2 CPU(s):   2,6,10,14,18,22,26,30,34,38,42,46,50,54,58,62,66,70,74,78,82,86,90,94,98,102,106,110,114,118,122,126,130,134,138,142,146,150,154,158,162,166,170,174
NUMA node3 CPU(s):   3,7,11,15,19,23,27,31,35,39,43,47,51,55,59,63,67,71,75,79,83,87,91,95,99,103,107,111,115,119,123,127,131,135,139,143,147,151,155,159,163,167,171,175
Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single intel_ppin ssbd mba ibrs ibpb stibp ibrs_enhanced fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts pku ospke avx512_vnni md_clear flush_l1d arch_capabilities

3.4 数据库配置选项参数

[mysqld]
skip-networking
lower_case_table_names = 1
character-set-server = UTF8MB4
skip_name_resolve = 1
default_time_zone = "+8:00"

#performance setttings
lock_wait_timeout = 3600
open_files_limit    = 65535
back_log = 1024
max_connections = 1024
max_connect_errors = 1000000
table_open_cache = 1024
table_definition_cache = 1024
thread_stack = 512K
sort_buffer_size = 4M
join_buffer_size = 4M
read_buffer_size = 8M
read_rnd_buffer_size = 4M
bulk_insert_buffer_size = 64M
thread_cache_size = 768
interactive_timeout = 600
wait_timeout = 600
tmp_table_size = 32M
max_heap_table_size = 32M

#log settings
log_timestamps = SYSTEM
log_error = error.log
log_error_verbosity = 3
slow_query_log = 1
log_slow_extra = 1
#log_slow_verbosity = FULL
slow_query_log_file = slow.log
long_query_time = 0.01
log_queries_not_using_indexes = 1
log_throttle_queries_not_using_indexes = 60
min_examined_row_limit = 0
log_slow_admin_statements = 1
log_slow_slave_statements = 1
log_bin = binlog

binlog_format = ROW
sync_binlog = 1
binlog_cache_size = 4M
max_binlog_cache_size = 2G
max_binlog_size = 1G
binlog_rows_query_log_events = 1
binlog_expire_logs_seconds = 604800
#binlog_expire_logs_auto_purge = 1
binlog_checksum = CRC32
gtid_mode = ON
enforce_gtid_consistency = TRUE

#myisam settings
key_buffer_size = 32M
myisam_sort_buffer_size = 128M

#replication settings
#master_info_repository = TABLE
#relay_log_info_repository = TABLE
relay_log_recovery = 1
slave_parallel_type = LOGICAL_CLOCK
slave_parallel_workers = 2
binlog_transaction_dependency_tracking = WRITESET
slave_preserve_commit_order = 1
slave_checkpoint_period = 2

#innodb settings
transaction_isolation = REPEATABLE-READ
innodb_buffer_pool_size = 2G
innodb_buffer_pool_instances = 8
innodb_data_file_path = ibdata1:12M:autoextend
innodb_flush_log_at_trx_commit = 1
innodb_log_buffer_size = 32M
innodb_log_file_size = 2G
innodb_log_files_in_group = 3
innodb_doublewrite_files = 2
innodb_max_undo_log_size = 4G
innodb_io_capacity = 400000
innodb_io_capacity_max = 800000
innodb_open_files = 65534
#本次测试采用O_DIRECT_NO_FSYNC模式
innodb_flush_method = O_DIRECT_NO_FSYNC
innodb_lru_scan_depth = 4000
innodb_lock_wait_timeout = 10
innodb_rollback_on_timeout = 1
innodb_print_all_deadlocks = 1
innodb_online_alter_log_max_size = 4G
innodb_print_ddl_logs = 0
innodb_status_file = 1
innodb_status_output = 0
innodb_status_output_locks = 1
innodb_sort_buffer_size = 67108864
#innodb_thread_concurrency = 176
#innodb_spin_wait_delay = 3
#innodb_sync_spin_loops = 10

#innodb monitor settings
innodb_monitor_enable = "module_innodb"
innodb_monitor_enable = "module_server"
innodb_monitor_enable = "module_dml"
innodb_monitor_enable = "module_ddl"
innodb_monitor_enable = "module_trx"
innodb_monitor_enable = "module_os"
innodb_monitor_enable = "module_purge"
innodb_monitor_enable = "module_log"
innodb_monitor_enable = "module_lock"
innodb_monitor_enable = "module_buffer"
innodb_monitor_enable = "module_index"
innodb_monitor_enable = "module_ibuf_system"
innodb_monitor_enable = "module_buffer_page"
innodb_monitor_enable = "module_adaptive_hash"

#pfs settings
performance_schema = 1
#performance_schema_instrument = '%memory%=on'
performance_schema_instrument = '%lock%=on'

#thread pool,需要开启thread pool时才取消下面两行注释
#thread_handling = 'pool-of-thread'
#thread_pool_stall_limit = 50

有关GreatSQL vs MySQL性能测试来了,速围观~的更多相关文章

  1. ruby-on-rails - 使用 Ruby on Rails 进行自动化测试 - 最佳实践 - 2

    很好奇,就使用ruby​​onrails自动化单元测试而言,你们正在做什么?您是否创建了一个脚本来在cron中运行rake作业并将结果邮寄给您?git中的预提交Hook?只是手动调用?我完全理解测试,但想知道在错误发生之前捕获错误的最佳实践是什么。让我们理所当然地认为测试本身是完美无缺的,并且可以正常工作。下一步是什么以确保他们在正确的时间将可能有害的结果传达给您? 最佳答案 不确定您到底想听什么,但是有几个级别的自动代码库控制:在处理某项功能时,您可以使用类似autotest的内容获得关于哪些有效,哪些无效的即时反馈。要确保您的提

  2. ruby - 使用 C 扩展开发 ruby​​gem 时,如何使用 Rspec 在本地进行测试? - 2

    我正在编写一个包含C扩展的gem。通常当我写一个gem时,我会遵循TDD的过程,我会写一个失败的规范,然后处理代码直到它通过,等等......在“ext/mygem/mygem.c”中我的C扩展和在gemspec的“扩展”中配置的有效extconf.rb,如何运行我的规范并仍然加载我的C扩展?当我更改C代码时,我需要采取哪些步骤来重新编译代码?这可能是个愚蠢的问题,但是从我的gem的开发源代码树中输入“bundleinstall”不会构建任何native扩展。当我手动运行rubyext/mygem/extconf.rb时,我确实得到了一个Makefile(在整个项目的根目录中),然后当

  3. ruby - Ruby 的 Hash 在比较键时使用哪种相等性测试? - 2

    我有一个围绕一些对象的包装类,我想将这些对象用作散列中的键。包装对象和解包装对象应映射到相同的键。一个简单的例子是这样的:classAattr_reader:xdefinitialize(inner)@inner=innerenddefx;@inner.x;enddef==(other)@inner.x==other.xendenda=A.new(o)#oisjustanyobjectthatallowso.xb=A.new(o)h={a=>5}ph[a]#5ph[b]#nil,shouldbe5ph[o]#nil,shouldbe5我试过==、===、eq?并散列所有无济于事。

  4. ruby - RSpec - 使用测试替身作为 block 参数 - 2

    我有一些Ruby代码,如下所示:Something.createdo|x|x.foo=barend我想编写一个测试,它使用double代替block参数x,这样我就可以调用:x_double.should_receive(:foo).with("whatever").这可能吗? 最佳答案 specify'something'dox=doublex.should_receive(:foo=).with("whatever")Something.should_receive(:create).and_yield(x)#callthere

  5. ruby - Sinatra:运行 rspec 测试时记录噪音 - 2

    Sinatra新手;我正在运行一些rspec测试,但在日志中收到了一堆不需要的噪音。如何消除日志中过多的噪音?我仔细检查了环境是否设置为:test,这意味着记录器级别应设置为WARN而不是DEBUG。spec_helper:require"./app"require"sinatra"require"rspec"require"rack/test"require"database_cleaner"require"factory_girl"set:environment,:testFactoryGirl.definition_file_paths=%w{./factories./test/

  6. ruby-on-rails - 迷你测试错误 : "NameError: uninitialized constant" - 2

    我遵循MichaelHartl的“RubyonRails教程:学习Web开发”,并创建了检查用户名和电子邮件长度有效性的测试(名称最多50个字符,电子邮件最多255个字符)。test/helpers/application_helper_test.rb的内容是:require'test_helper'classApplicationHelperTest在运行bundleexecraketest时,所有测试都通过了,但我看到以下消息在最后被标记为错误:ERROR["test_full_title_helper",ApplicationHelperTest,1.820016791]test

  7. ruby - 即使失败也继续进行多主机测试 - 2

    我已经构建了一些serverspec代码来在多个主机上运行一组测试。问题是当任何测试失败时,测试会在当前主机停止。即使测试失败,我也希望它继续在所有主机上运行。Rakefile:namespace:specdotask:all=>hosts.map{|h|'spec:'+h.split('.')[0]}hosts.eachdo|host|begindesc"Runserverspecto#{host}"RSpec::Core::RakeTask.new(host)do|t|ENV['TARGET_HOST']=hostt.pattern="spec/cfengine3/*_spec.r

  8. ruby-on-rails - 如何使辅助方法在 Rails 集成测试中可用? - 2

    我在app/helpers/sessions_helper.rb中有一个帮助程序文件,其中包含一个方法my_preference,它返回当前登录用户的首选项。我想在集成测试中访问该方法。例如,这样我就可以在测试中使用getuser_path(my_preference)。在其他帖子中,我读到这可以通过在测试文件中包含requiresessions_helper来实现,但我仍然收到错误NameError:undefinedlocalvariableormethod'my_preference'.我做错了什么?require'test_helper'require'sessions_hel

  9. ruby-on-rails - Cucumber 是否只是 rspec 的包装器以帮助将测试组织成功能? - 2

    只是想确保我理解了事情。据我目前收集到的信息,Cucumber只是一个“包装器”,或者是一种通过将事物分类为功能和步骤来组织测试的好方法,其中实际的单元测试处于步骤阶段。它允许您根据事物的工作方式组织您的测试。对吗? 最佳答案 有点。它是一种组织测试的方式,但不仅如此。它的行为就像最初的Rails集成测试一样,但更易于使用。这里最大的好处是您的session在整个Scenario中保持透明。关于Cucumber的另一件事是您(应该)从使用您的代码的浏览器或客户端的角度进行测试。如果您愿意,您可以使用步骤来构建对象和设置状态,但通常您

  10. ruby-on-rails - 如何调试 cucumber 测试? - 2

    我有:When/^(?:|I)follow"([^"]*)"(?:within"([^"]*)")?$/do|link,selector|with_scope(selector)doclick_link(link)endend我打电话的地方:Background:GivenIamanexistingadminuserWhenIfollow"CLIENTS"我的HTML是这样的:CLIENTS我一直收到这个错误:.F-.F--U-----U(::)failedsteps(::)nolinkwithtitle,idortext'CLIENTS'found(Capybara::Element

随机推荐