Elasticsearch 7.14.1集群压测报告（32核128G*10 本地NVMe SSD，Intel）

2022-08-16 15:58:08 浏览数 (1)

说明

本文描述问题及解决方法同样适用于 腾讯云 Elasticsearch Service（ES）。

另外使用到：腾讯云 云服务器（Cloud Virtual Machine，CVM）

本文另有延续：

Elasticsearch压力测试 - 云 社区 - 腾讯云 (tencent.com)

环境配置

Esrally客户端环境

  • 版本 Linux环境：Centos 7.9 Python：3.8.7 Pip：pip 20.2.3 from pip (python 3.8) Java：openjdk version 1.8.0_302 (build 1.8.0_302-b08) Git：2.7.5 Esrally：2.3.0 测试基准：geonames(http://www.geonames.org/)
  • 配置 节点数量：1 内存：64G 硬盘：SSD云硬盘 100GB CPU个数：1 CPU核心数：32

Elasticsearch服务端环境

  • 版本 Linux环境：Centos 7.2 Java：openjdk version 11.0.9.1-ga (build 11.0.9.1-ga 1, mixed mode) Elasticsearch版本：7.14.1（腾讯云 Elasticsearch Service 基础版）
  • 配置 节点数量：10 内存：128G 硬盘：本地NVMe SSD盘 3.5T*2 CPU个数：1 CPU核心数：32 CPU型号：Intel(R) Xeon(R) Platinum 8255C CPU @ 2.50GHz

压测对比（demo）

该截图为压测对比的片段数据demo，对比对象为同等规格的不同配置项，其中包括分片、副本、索引参数等，我们也可以通过该方式对比不同规格、不同版本之间的压测结果。 本次压测参数：

代码语言：javascript复制
esrally race 
  --track=geonames 
  --target-hosts=10.0.0.1:9200 
  --pipeline=benchmark-only 
  --track-params="number_of_shards:20, number_of_replicas:0" 
  --user-tag="version:Intel_32C128G_7T*10"

配置新增项： 文件位于：.rally/benchmarks/tracks/default/geonames/challenges/default.json

代码语言：javascript复制
      "translog": {
        "sync_interval": "5s",
        "durability": "async"
      }

对比demo压测参数：

代码语言：javascript复制
esrally race 
  --track=geonames 
  --target-hosts=10.0.0.1:9200 
  --pipeline=benchmark-only 
  --track-params="number_of_shards:10, number_of_replicas:1" 
  --user-tag="version:Intel_32C128G_7T*10"

对比命令：

代码语言：javascript复制
esrally compare --baseline=673c88f8-0bad-45f7-887d-b9b0dd8e5221 --contender=825fe220-69c5-4943-ab8f-71a35a989c60

第三列为本次数据，第四列为对比数据，第五列为对比数值差，最后一列为对比百分百差值

● 红色代表优

● 绿色代表劣

压测报告

在 Metric一栏，有非常多的指标数据，详细的解释可以参见 官方文档(http://esrally.readthedocs.io/en/latest/summary_report.html)。 一般要关注的数据有：

  • throughput 每个操作的吞吐量，比如 index、search等
  • latency 每个操作的响应时长数据
  • Heap used for x 记录堆栈的使用情况

具体的指标含义，可以参阅另一篇文档：压测标准

Metric

Task

Value

Unit

Cumulative indexing time of primary shards

11.6213

min

Min cumulative indexing time across primary shards

0

min

Median cumulative indexing time across primary shards

0.565267

min

Max cumulative indexing time across primary shards

0.591267

min

Cumulative indexing throttle time of primary shards

0

min

Min cumulative indexing throttle time across primary shards

0

min

Median cumulative indexing throttle time across primary shards

0

min

Max cumulative indexing throttle time across primary shards

0

min

Cumulative merge time of primary shards

2.4651

min

Cumulative merge count of primary shards

1380

Min cumulative merge time across primary shards

0

min

Median cumulative merge time across primary shards

0.0945333

min

Max cumulative merge time across primary shards

0.740583

min

Cumulative merge throttle time of primary shards

0.528317

min

Min cumulative merge throttle time across primary shards

0

min

Median cumulative merge throttle time across primary shards

0

min

Max cumulative merge throttle time across primary shards

0.0492667

min

Cumulative refresh time of primary shards

2.23803

min

Cumulative refresh count of primary shards

13298

Min cumulative refresh time across primary shards

0

min

Median cumulative refresh time across primary shards

0.0526

min

Max cumulative refresh time across primary shards

0.999583

min

Cumulative flush time of primary shards

0.00281667

min

Cumulative flush count of primary shards

31

Min cumulative flush time across primary shards

0

min

Median cumulative flush time across primary shards

0

min

Max cumulative flush time across primary shards

0.000416667

min

Total Young Gen GC time

7.046

s

Total Young Gen GC count

1333

Total Old Gen GC time

0

s

Total Old Gen GC count

0

Store size

3.34291

GB

Translog size

0.0862316

GB

Heap used for segments

2.21271

MB

Heap used for doc values

0.155579

MB

Heap used for terms

1.74126

MB

Heap used for norms

0.186829

MB

Heap used for points

0

MB

Heap used for stored fields

0.129044

MB

Segment count

262

Total Ingest Pipeline count

0

Total Ingest Pipeline time

0

s

Total Ingest Pipeline failed

0

error rate

index-append

0

%

Min Throughput

index-stats

90

ops/s

Mean Throughput

index-stats

90.01

ops/s

Median Throughput

index-stats

90.02

ops/s

Max Throughput

index-stats

90.03

ops/s

50th percentile latency

index-stats

2.59685

ms

90th percentile latency

index-stats

3.96404

ms

99th percentile latency

index-stats

4.29628

ms

99.9th percentile latency

index-stats

4.9505

ms

100th percentile latency

index-stats

5.76097

ms

50th percentile service time

index-stats

1.83113

ms

90th percentile service time

index-stats

1.96145

ms

99th percentile service time

index-stats

2.49059

ms

99.9th percentile service time

index-stats

2.99369

ms

100th percentile service time

index-stats

3.38487

ms

error rate

index-stats

0

%

Min Throughput

node-stats

42.88

ops/s

Mean Throughput

node-stats

44.05

ops/s

Median Throughput

node-stats

44.21

ops/s

Max Throughput

node-stats

44.45

ops/s

50th percentile latency

node-stats

6908.99

ms

90th percentile latency

node-stats

11407.1

ms

99th percentile latency

node-stats

12427.2

ms

99.9th percentile latency

node-stats

12526.8

ms

100th percentile latency

node-stats

12537.6

ms

50th percentile service time

node-stats

12.3356

ms

90th percentile service time

node-stats

13.456

ms

99th percentile service time

node-stats

15.9214

ms

99.9th percentile service time

node-stats

17.1677

ms

100th percentile service time

node-stats

18.1915

ms

error rate

node-stats

0

%

Min Throughput

default

50.02

ops/s

Mean Throughput

default

50.04

ops/s

Median Throughput

default

50.04

ops/s

Max Throughput

default

50.07

ops/s

50th percentile latency

default

2.96665

ms

90th percentile latency

default

3.80139

ms

99th percentile latency

default

4.19294

ms

99.9th percentile latency

default

8.99626

ms

100th percentile latency

default

9.71564

ms

50th percentile service time

default

1.76794

ms

90th percentile service time

default

1.93181

ms

99th percentile service time

default

2.25012

ms

99.9th percentile service time

default

8.38265

ms

100th percentile service time

default

8.84024

ms

error rate

default

0

%

Min Throughput

term

100.03

ops/s

Mean Throughput

term

100.05

ops/s

Median Throughput

term

100.04

ops/s

Max Throughput

term

100.09

ops/s

50th percentile latency

term

2.885

ms

90th percentile latency

term

3.28685

ms

99th percentile latency

term

3.62102

ms

99.9th percentile latency

term

8.40079

ms

100th percentile latency

term

14.5035

ms

50th percentile service time

term

2.12238

ms

90th percentile service time

term

2.27681

ms

99th percentile service time

term

2.47782

ms

99.9th percentile service time

term

7.35007

ms

100th percentile service time

term

13.5583

ms

error rate

term

0

%

Min Throughput

phrase

110

ops/s

Mean Throughput

phrase

110.02

ops/s

Median Throughput

phrase

110.02

ops/s

Max Throughput

phrase

110.03

ops/s

50th percentile latency

phrase

2.82409

ms

90th percentile latency

phrase

3.23931

ms

99th percentile latency

phrase

7.67388

ms

99.9th percentile latency

phrase

21.0077

ms

100th percentile latency

phrase

27.748

ms

50th percentile service time

phrase

2.06775

ms

90th percentile service time

phrase

2.21518

ms

99th percentile service time

phrase

4.07794

ms

99.9th percentile service time

phrase

13.957

ms

100th percentile service time

phrase

27.1629

ms

error rate

phrase

0

%

Min Throughput

country_agg_uncached

3.01

ops/s

Mean Throughput

country_agg_uncached

3.01

ops/s

Median Throughput

country_agg_uncached

3.01

ops/s

Max Throughput

country_agg_uncached

3.01

ops/s

50th percentile latency

country_agg_uncached

41.801

ms

90th percentile latency

country_agg_uncached

61.1074

ms

99th percentile latency

country_agg_uncached

64.1245

ms

100th percentile latency

country_agg_uncached

64.1869

ms

50th percentile service time

country_agg_uncached

40.5648

ms

90th percentile service time

country_agg_uncached

60.413

ms

99th percentile service time

country_agg_uncached

63.2005

ms

100th percentile service time

country_agg_uncached

63.2089

ms

error rate

country_agg_uncached

0

%

Min Throughput

country_agg_cached

99.44

ops/s

Mean Throughput

country_agg_cached

99.61

ops/s

Median Throughput

country_agg_cached

99.62

ops/s

Max Throughput

country_agg_cached

99.72

ops/s

50th percentile latency

country_agg_cached

2.63589

ms

90th percentile latency

country_agg_cached

3.02715

ms

99th percentile latency

country_agg_cached

3.35078

ms

99.9th percentile latency

country_agg_cached

9.61668

ms

100th percentile latency

country_agg_cached

17.3212

ms

50th percentile service time

country_agg_cached

1.86138

ms

90th percentile service time

country_agg_cached

2.00815

ms

99th percentile service time

country_agg_cached

2.15493

ms

99.9th percentile service time

country_agg_cached

2.37638

ms

100th percentile service time

country_agg_cached

16.6456

ms

error rate

country_agg_cached

0

%

Min Throughput

scroll

20.03

pages/s

Mean Throughput

scroll

20.04

pages/s

Median Throughput

scroll

20.04

pages/s

Max Throughput

scroll

20.05

pages/s

50th percentile latency

scroll

614.353

ms

90th percentile latency

scroll

625.645

ms

99th percentile latency

scroll

632.598

ms

100th percentile latency

scroll

633.479

ms

50th percentile service time

scroll

612.703

ms

90th percentile service time

scroll

624.571

ms

99th percentile service time

scroll

630.818

ms

100th percentile service time

scroll

632.295

ms

error rate

scroll

0

%

Min Throughput

expression

1.5

ops/s

Mean Throughput

expression

1.5

ops/s

Median Throughput

expression

1.5

ops/s

Max Throughput

expression

1.51

ops/s

50th percentile latency

expression

83.2414

ms

90th percentile latency

expression

140.869

ms

99th percentile latency

expression

146.941

ms

100th percentile latency

expression

149.266

ms

50th percentile service time

expression

81.8575

ms

90th percentile service time

expression

139.747

ms

99th percentile service time

expression

145.243

ms

100th percentile service time

expression

148.174

ms

error rate

expression

0

%

Min Throughput

painless_static

1.4

ops/s

Mean Throughput

painless_static

1.4

ops/s

Median Throughput

painless_static

1.4

ops/s

Max Throughput

painless_static

1.41

ops/s

50th percentile latency

painless_static

99.7975

ms

90th percentile latency

painless_static

158.673

ms

99th percentile latency

painless_static

161.488

ms

100th percentile latency

painless_static

161.666

ms

50th percentile service time

painless_static

98.5624

ms

90th percentile service time

painless_static

157.462

ms

99th percentile service time

painless_static

159.464

ms

100th percentile service time

painless_static

160.565

ms

error rate

painless_static

0

%

Min Throughput

painless_dynamic

1.4

ops/s

Mean Throughput

painless_dynamic

1.4

ops/s

Median Throughput

painless_dynamic

1.4

ops/s

Max Throughput

painless_dynamic

1.41

ops/s

50th percentile latency

painless_dynamic

101.198

ms

90th percentile latency

painless_dynamic

159.528

ms

99th percentile latency

painless_dynamic

164.991

ms

100th percentile latency

painless_dynamic

165.723

ms

50th percentile service time

painless_dynamic

99.9259

ms

90th percentile service time

painless_dynamic

158.47

ms

99th percentile service time

painless_dynamic

164.141

ms

100th percentile service time

painless_dynamic

164.411

ms

error rate

painless_dynamic

0

%

Min Throughput

decay_geo_gauss_function_score

1

ops/s

Mean Throughput

decay_geo_gauss_function_score

1

ops/s

Median Throughput

decay_geo_gauss_function_score

1

ops/s

Max Throughput

decay_geo_gauss_function_score

1

ops/s

50th percentile latency

decay_geo_gauss_function_score

106.114

ms

90th percentile latency

decay_geo_gauss_function_score

131.256

ms

99th percentile latency

decay_geo_gauss_function_score

134.307

ms

100th percentile latency

decay_geo_gauss_function_score

135.118

ms

50th percentile service time

decay_geo_gauss_function_score

104.521

ms

90th percentile service time

decay_geo_gauss_function_score

129.729

ms

99th percentile service time

decay_geo_gauss_function_score

132.644

ms

100th percentile service time

decay_geo_gauss_function_score

133.117

ms

error rate

decay_geo_gauss_function_score

0

%

Min Throughput

decay_geo_gauss_script_score

1

ops/s

Mean Throughput

decay_geo_gauss_script_score

1

ops/s

Median Throughput

decay_geo_gauss_script_score

1

ops/s

Max Throughput

decay_geo_gauss_script_score

1

ops/s

50th percentile latency

decay_geo_gauss_script_score

101.618

ms

90th percentile latency

decay_geo_gauss_script_score

129.315

ms

99th percentile latency

decay_geo_gauss_script_score

132.952

ms

100th percentile latency

decay_geo_gauss_script_score

133.003

ms

50th percentile service time

decay_geo_gauss_script_score

99.8716

ms

90th percentile service time

decay_geo_gauss_script_score

127.959

ms

99th percentile service time

decay_geo_gauss_script_score

131.571

ms

100th percentile service time

decay_geo_gauss_script_score

131.693

ms

error rate

decay_geo_gauss_script_score

0

%

Min Throughput

field_value_function_score

1.5

ops/s

Mean Throughput

field_value_function_score

1.51

ops/s

Median Throughput

field_value_function_score

1.51

ops/s

Max Throughput

field_value_function_score

1.51

ops/s

50th percentile latency

field_value_function_score

43.0776

ms

90th percentile latency

field_value_function_score

57.9185

ms

99th percentile latency

field_value_function_score

59.9253

ms

100th percentile latency

field_value_function_score

59.9498

ms

50th percentile service time

field_value_function_score

41.8078

ms

90th percentile service time

field_value_function_score

56.8933

ms

99th percentile service time

field_value_function_score

58.5603

ms

100th percentile service time

field_value_function_score

58.8797

ms

error rate

field_value_function_score

0

%

Min Throughput

field_value_script_score

1.5

ops/s

Mean Throughput

field_value_script_score

1.51

ops/s

Median Throughput

field_value_script_score

1.51

ops/s

Max Throughput

field_value_script_score

1.51

ops/s

50th percentile latency

field_value_script_score

51.2358

ms

90th percentile latency

field_value_script_score

78.1297

ms

99th percentile latency

field_value_script_score

81.5962

ms

100th percentile latency

field_value_script_score

85.1375

ms

50th percentile service time

field_value_script_score

50.0607

ms

90th percentile service time

field_value_script_score

77.0218

ms

99th percentile service time

field_value_script_score

79.9932

ms

100th percentile service time

field_value_script_score

83.5557

ms

error rate

field_value_script_score

0

%

Min Throughput

large_terms

1.1

ops/s

Mean Throughput

large_terms

1.1

ops/s

Median Throughput

large_terms

1.1

ops/s

Max Throughput

large_terms

1.1

ops/s

50th percentile latency

large_terms

759.051

ms

90th percentile latency

large_terms

803.56

ms

99th percentile latency

large_terms

1171.68

ms

100th percentile latency

large_terms

1281.92

ms

50th percentile service time

large_terms

752.205

ms

90th percentile service time

large_terms

790.845

ms

99th percentile service time

large_terms

1045.43

ms

100th percentile service time

large_terms

1275.7

ms

error rate

large_terms

0

%

Min Throughput

large_filtered_terms

1.1

ops/s

Mean Throughput

large_filtered_terms

1.1

ops/s

Median Throughput

large_filtered_terms

1.1

ops/s

Max Throughput

large_filtered_terms

1.1

ops/s

50th percentile latency

large_filtered_terms

761.057

ms

90th percentile latency

large_filtered_terms

802.481

ms

99th percentile latency

large_filtered_terms

867.485

ms

100th percentile latency

large_filtered_terms

883.59

ms

50th percentile service time

large_filtered_terms

754.452

ms

90th percentile service time

large_filtered_terms

795.488

ms

99th percentile service time

large_filtered_terms

861.441

ms

100th percentile service time

large_filtered_terms

876.858

ms

error rate

large_filtered_terms

0

%

Min Throughput

large_prohibited_terms

1.1

ops/s

Mean Throughput

large_prohibited_terms

1.1

ops/s

Median Throughput

large_prohibited_terms

1.1

ops/s

Max Throughput

large_prohibited_terms

1.1

ops/s

50th percentile latency

large_prohibited_terms

758.152

ms

90th percentile latency

large_prohibited_terms

801.5

ms

99th percentile latency

large_prohibited_terms

868.599

ms

100th percentile latency

large_prohibited_terms

895.196

ms

50th percentile service time

large_prohibited_terms

751.518

ms

90th percentile service time

large_prohibited_terms

795.278

ms

99th percentile service time

large_prohibited_terms

862.411

ms

100th percentile service time

large_prohibited_terms

887.207

ms

error rate

large_prohibited_terms

0

%

Min Throughput

desc_sort_population

1.5

ops/s

Mean Throughput

desc_sort_population

1.51

ops/s

Median Throughput

desc_sort_population

1.51

ops/s

Max Throughput

desc_sort_population

1.51

ops/s

50th percentile latency

desc_sort_population

23.6787

ms

90th percentile latency

desc_sort_population

36.879

ms

99th percentile latency

desc_sort_population

38.4983

ms

100th percentile latency

desc_sort_population

38.7709

ms

50th percentile service time

desc_sort_population

22.2716

ms

90th percentile service time

desc_sort_population

35.7709

ms

99th percentile service time

desc_sort_population

37.5019

ms

100th percentile service time

desc_sort_population

37.8242

ms

error rate

desc_sort_population

0

%

Min Throughput

asc_sort_population

1.5

ops/s

Mean Throughput

asc_sort_population

1.51

ops/s

Median Throughput

asc_sort_population

1.51

ops/s

Max Throughput

asc_sort_population

1.51

ops/s

50th percentile latency

asc_sort_population

24.1694

ms

90th percentile latency

asc_sort_population

37.0402

ms

99th percentile latency

asc_sort_population

38.6004

ms

100th percentile latency

asc_sort_population

39.1573

ms

50th percentile service time

asc_sort_population

22.7906

ms

90th percentile service time

asc_sort_population

35.8008

ms

99th percentile service time

asc_sort_population

37.3242

ms

100th percentile service time

asc_sort_population

37.448

ms

error rate

asc_sort_population

0

%

Min Throughput

asc_sort_with_after_population

1.5

ops/s

Mean Throughput

asc_sort_with_after_population

1.51

ops/s

Median Throughput

asc_sort_with_after_population

1.51

ops/s

Max Throughput

asc_sort_with_after_population

1.51

ops/s

50th percentile latency

asc_sort_with_after_population

28.2968

ms

90th percentile latency

asc_sort_with_after_population

44.5625

ms

99th percentile latency

asc_sort_with_after_population

49.3172

ms

100th percentile latency

asc_sort_with_after_population

49.7945

ms

50th percentile service time

asc_sort_with_after_population

26.7906

ms

90th percentile service time

asc_sort_with_after_population

43.682

ms

99th percentile service time

asc_sort_with_after_population

48.0795

ms

100th percentile service time

asc_sort_with_after_population

48.2475

ms

error rate

asc_sort_with_after_population

0

%

Min Throughput

desc_sort_geonameid

6.02

ops/s

Mean Throughput

desc_sort_geonameid

6.02

ops/s

Median Throughput

desc_sort_geonameid

6.02

ops/s

Max Throughput

desc_sort_geonameid

6.03

ops/s

50th percentile latency

desc_sort_geonameid

5.72709

ms

90th percentile latency

desc_sort_geonameid

6.3101

ms

99th percentile latency

desc_sort_geonameid

6.73087

ms

100th percentile latency

desc_sort_geonameid

6.74572

ms

50th percentile service time

desc_sort_geonameid

4.87845

ms

90th percentile service time

desc_sort_geonameid

5.18251

ms

99th percentile service time

desc_sort_geonameid

5.64189

ms

100th percentile service time

desc_sort_geonameid

5.78124

ms

error rate

desc_sort_geonameid

0

%

Min Throughput

desc_sort_with_after_geonameid

6.02

ops/s

Mean Throughput

desc_sort_with_after_geonameid

6.02

ops/s

Median Throughput

desc_sort_with_after_geonameid

6.02

ops/s

Max Throughput

desc_sort_with_after_geonameid

6.02

ops/s

50th percentile latency

desc_sort_with_after_geonameid

27.6875

ms

90th percentile latency

desc_sort_with_after_geonameid

40.815

ms

99th percentile latency

desc_sort_with_after_geonameid

46.1911

ms

100th percentile latency

desc_sort_with_after_geonameid

49.8309

ms

50th percentile service time

desc_sort_with_after_geonameid

26.249

ms

90th percentile service time

desc_sort_with_after_geonameid

39.8379

ms

99th percentile service time

desc_sort_with_after_geonameid

45.2571

ms

100th percentile service time

desc_sort_with_after_geonameid

48.6642

ms

error rate

desc_sort_with_after_geonameid

0

%

Min Throughput

asc_sort_geonameid

6.02

ops/s

Mean Throughput

asc_sort_geonameid

6.02

ops/s

Median Throughput

asc_sort_geonameid

6.02

ops/s

Max Throughput

asc_sort_geonameid

6.03

ops/s

50th percentile latency

asc_sort_geonameid

5.41121

ms

90th percentile latency

asc_sort_geonameid

5.97722

ms

99th percentile latency

asc_sort_geonameid

6.54733

ms

100th percentile latency

asc_sort_geonameid

6.5514

ms

50th percentile service time

asc_sort_geonameid

4.49445

ms

90th percentile service time

asc_sort_geonameid

5.06776

ms

99th percentile service time

asc_sort_geonameid

5.25747

ms

100th percentile service time

asc_sort_geonameid

5.25751

ms

error rate

asc_sort_geonameid

0

%

Min Throughput

asc_sort_with_after_geonameid

6.02

ops/s

Mean Throughput

asc_sort_with_after_geonameid

6.02

ops/s

Median Throughput

asc_sort_with_after_geonameid

6.02

ops/s

Max Throughput

asc_sort_with_after_geonameid

6.02

ops/s

50th percentile latency

asc_sort_with_after_geonameid

25.004

ms

90th percentile latency

asc_sort_with_after_geonameid

37.8372

ms

99th percentile latency

asc_sort_with_after_geonameid

41.8776

ms

100th percentile latency

asc_sort_with_after_geonameid

42.2647

ms

50th percentile service time

asc_sort_with_after_geonameid

23.2009

ms

90th percentile service time

asc_sort_with_after_geonameid

35.8662

ms

99th percentile service time

asc_sort_with_after_geonameid

39.9241

ms

100th percentile service time

asc_sort_with_after_geonameid

40.0065

ms

error rate

asc_sort_with_after_geonameid

0

%

ElasticsearchService 压力测试 大数据解决方案 大数据

0 人点赞