1.docker安装 minio/warp
docker pull minio/warp
这个命令是获取最新版本的镜像，如果需要获取指定版本的镜像在后面加上 :版本号 就可以了。
[~]$ docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
minio/warp latest 8ebca3ef7060 8 days ago 17.9MB
需要看看具体的版本号可以通过 inspect查看
docker image inspect minio/warp:latest |grep -i version
常用参数
| 参数 | 释义 |
|---|---|
| --benchdata | 指定压测结果文件的名称（包括文件路径） |
| --obj.size | 指定对象的大小 |
| --host | 指定服务器 |
| --access-key | 账号 |
| --secret-key | 密码 |
| --concurrent | 并发数 |
| --obj.randsize | 随机大小 |
| --duration | 持续时间 |
| --autoterm | 当基准被认为稳定时自动终止 |
2.利用minio/warp 进行压测
docker run --net host --name warp\
-v /home/test/result:/result \
minio/warp put --host=192.168.1.{7...9}:9000 --access-key=minioadmin --secret-key=minioadmin \
--benchdata /result/put-100 \
--concurrent 20 \
--obj.randsize \
--duration 10m0s \
--autoterm
由于warp是运行完之后容器就停止运行了，无法进入容器内部（运行任务的时候也进不去），所以要获取生成的结果文件， 要么是通过 docker cp把文件复制出来（生成的文件默认是直接在根目录下的）
docker cp 0f411107bb2f:warp-mixed-2022-09-27[090405]-rk20.csv.zst /home/test/minio/warp
| 参数 | 释义 |
|---|---|
| 0f411107bb2f | docker run生成的容器的id，如果执行的时候用--name指定了名字，可换成指定的名字 |
| warp-mixed-2022-09-27[090405]-rk20.csv.zst | warp生成的结果文件，可以根据--benchdata指定名字及路径 |
| /home/test/minio/warp | 你的linux的路径 |
要么就像上面一样，通过docker的文件映射将文件直接映射到linux上：
| 参数 | 释义 |
|---|---|
| --benchdata /result/put-100 | 指定将压测结果生成到 容器内result文件夹下，文件名为 put-100 （最终生成的文件的名称为 put-100.csv.zst） |
| --host=192.168.1.{7...9}:9000 | 压测的环境 7...9 是minio支持的写法，表示 7 8 9 |
| --access-key | minio的登录名 |
| --secret-key | minio的密码 |
| -v /home/test/result:/result | 设置映射 |
3.结果分析
warp执行完会把结果打印的控制台，可以在命令后面加上 >result.log 输出到文件，压缩的详细数据存储在.csv.zst后缀的文件中，可以通过 minio/warp analyze 解析文件
docker run --name warp-analyze --net host \
-v /home/test/result/put-100.csv.zst:/result/put-100.csv.zst \
minio/warp analyze --analyze.v /result/put-100.csv.zst >put-100-result-detail.log
因为是创建容器去分析，所以需要把文件映射到容器内部
| 参数 | 释义 |
|---|---|
| -v /home/test/result/put-100.csv.zst:/result/put-100.csv.zst | 把文件映射到容器 |
| >put-100-result-detail.log | 把解析结果输出到log文件 |
结果
warp: Preparing server.
Clearing Prefix "warp-benchmark-bucket/"...warp: Starting benchmark in 3s...
warp: Benchmark starting...
Throughput 110.6MiB/s within 7.500000% for 14.623s. Assuming stability. Terminating benchmark.
warp: Saving benchmark data...
warp: Benchmark data written to "/result/put-100.csv.zst"
----------------------------------------
Operation: PUT
* Average: 111.06 MiB/s, 11.11 obj/s
Throughput by host:
* http://192.168.1.7:9000: Avg: 17.79 MiB/s, 1.78 obj/s
* http://192.168.1.8:9000: Avg: 17.27 MiB/s, 1.73 obj/s
* http://192.168.1.9:9000: Avg: 76.05 MiB/s, 7.61 obj/s
Throughput, split into 57 x 1s:
* Fastest: 121.3MiB/s, 12.13 obj/s
* 50% Median: 111.5MiB/s, 11.15 obj/s
* Slowest: 99.0MiB/s, 9.90 obj/s
warp: Starting cleanup...
Clearing Prefix "warp-benchmark-bucket/0uvYTk18/"...warp: Cleanup Done.
analyze的详情数据
----------------------------------------
Operation: PUT (766). Ran 1m9s. Concurrency: 20. Hosts: 3.
Requests considered: 616:
* Avg: 1.73s, 50%: 917ms, 90%: 4.098s, 99%: 10.721s, Fastest: 176ms, Slowest: 13.593s
Requests by host:
* http://192.168.1.7:9000 - 125 requests:
- Avg: 3.759s Fastest: 1.019s Slowest: 13.593s 50%: 2.801s 90%: 7.984s
* http://192.168.1.8:9000 - 121 requests:
- Avg: 3.907s Fastest: 1.131s Slowest: 12.458s 50%: 3.108s 90%: 7.635s
* http://192.168.1.9:9000 - 520 requests:
- Avg: 833ms Fastest: 176ms Slowest: 2.428s 50%: 771ms 90%: 1.347s
Throughput:
* Average: 111.06 MiB/s, 11.11 obj/s
Throughput by host:
* http://192.168.1.7:9000:
- Average: 17.79 MiB/s, 1.78 obj/s
- Fastest: 31.6MiB/s
- 50% Median: 17.7MiB/s
- Slowest: 9.9MiB/s
* http://192.168.1.8:9000:
- Average: 17.27 MiB/s, 1.73 obj/s
- Fastest: 26.0MiB/s
- 50% Median: 17.3MiB/s
- Slowest: 11.4MiB/s
* http://192.168.1.9:9000:
- Average: 76.05 MiB/s, 7.61 obj/s
- Fastest: 98.9MiB/s
- 50% Median: 76.4MiB/s
- Slowest: 49.7MiB/s
Throughput, split into 57 x 1s:
* Fastest: 121.3MiB/s, 12.13 obj/s (1s, starting 06:26:42 UTC)
* 50% Median: 111.5MiB/s, 11.15 obj/s (1s, starting 06:26:59 UTC)
* Slowest: 99.0MiB/s, 9.90 obj/s (1s, starting 06:26:41 UTC)
4.文档
参数释义
NAME:
warp - Benchmark tool for S3 compatible object storage systems.
For usage details see https://github.com/minio/warp
USAGE:
warp [global options] command [command options] [arguments...]
VERSION:
0.6.3 - d691c32
AUTHOR:
MinIO, Inc.
COMMANDS:
mixed benchmark(基准) mixed objects
get benchmark get objects
put benchmark put objects
delete benchmark delete objects
list benchmark list objects
stat benchmark stat objects (get file info)
select benchmark select objects
versioned benchmark mixed versioned objects
retention benchmark PutObjectRetention
multipart benchmark multipart object
analyze analyze existing benchmark data
cmp compare existing benchmark data
merge merge existing benchmark data
client run warp in client mode, accepting connections to run benchmarks
GLOBAL FLAGS:
--no-color disable color theme
--debug enable debug output
--insecure disable TLS certificate verification
--autocompletion install auto-completion for your shell
--help, -h show help
--version, -v print the version
NAME:
warp put - benchmark put objects
USAGE:
warp put [FLAGS]
-> see https://github.com/minio/warp#put
FLAGS:
--no-color disable color theme
--debug enable debug output
--insecure disable TLS certificate verification
--autocompletion install auto-completion for your shell
--host value host. Multiple hosts can be specified as a comma separated list. (default: "127.0.0.1:9000") [$WARP_HOST]
--access-key value Specify access key [$WARP_ACCESS_KEY]
--secret-key value Specify secret key [$WARP_SECRET_KEY]
--tls Use TLS (HTTPS) for transport [$WARP_TLS]
--region value Specify a custom region [$WARP_REGION]
--encrypt encrypt/decrypt objects (using server-side encryption with random keys)
--bucket value Bucket to use for benchmark data. ALL DATA WILL BE DELETED IN BUCKET! (default: "warp-benchmark-bucket")
--host-select value Host selection algorithm. Can be "weighed" or "roundrobin" (default: "weighed")
--concurrent value Run this many concurrent operations (default: 20)
--noprefix Do not use separate prefix for each thread
--prefix value Use a custom prefix for each thread
--disable-multipart disable multipart uploads
--md5 Add MD5 sum to uploads
--storage-class value Specify custom storage class, for instance 'STANDARD' or 'REDUCED_REDUNDANCY'.
--obj.size value Size of each generated object. Can be a number or 10KiB/MiB/GiB. All sizes are base 2 binary. (default: "10MiB")
--obj.generator value Use specific data generator (default: "random")
--obj.randsize Randomize size of objects so they will be up to the specified size
--benchdata value Output benchmark+profile data to this file. By default unique filename is generated.
--serverprof value Run MinIO server profiling during benchmark; possible values are 'cpu', 'mem', 'block', 'mutex' and 'trace'.
--duration value Duration to run the benchmark. Use 's' and 'm' to specify seconds and minutes. (default: 5m0s)
--autoterm Auto terminate when benchmark is considered stable.
--autoterm.dur value Minimum duration where output must have been stable to allow automatic termination. (default: 10s)
--autoterm.pct value The percentage the last 6/25 time blocks must be within current speed to auto terminate. (default: 7.5)
--noclear Do not clear bucket before or after running benchmarks. Use when running multiple clients.
--syncstart value Specify a benchmark start time. Time format is 'hh:mm' where hours are specified in 24h format, server TZ.
--warp-client value Connect to warp clients and run benchmarks there.
--analyze.dur value Split analysis into durations of this length. Can be '1s', '5s', '1m', etc.
--analyze.out value Output aggregated data as to file
--analyze.op value Only output for this op. Can be GET/PUT/DELETE, etc.
--analyze.host value Only output for this host.
--analyze.skip value Additional duration to skip when analyzing data. (default: 0s)
--analyze.v Display additional analysis data.
--serve value When running benchmarks open a webserver to fetch results remotely, eg: localhost:7762
--help, -h show help
