性能优化

机构挂靠接口性能优化

问题/需求

【机构挂靠】sfe-affiliate-api/affiliate/open-list接口响应时间慢
- 药店挂靠，接口有点慢，451条数据，3分钟左右。

性能分析

机构挂靠服务

sfe-affiliate-api

服务调用关系

java

@ApiOperation("开放挂靠列表")
@PostMapping(AffiliateRouter.LIST_OPEN_AFFILIATE)
ActionResult<IPage<AffiliateRelationInfo>> getAffiliateRelationOpen(@SpringQueryMap Page page, @RequestBody @Validated AffiliateRelationOpenQO query);

接口监控

近一个月接口监控

抓取对应日志获取入参

json

{
	url: http://10.66.8.197/affiliate/open-list?pages=1&size=1000,
	method: POST,
	headers: [
		host:sfe-affiliate-api.stable:8080,
		tm-header-tenantid:df2c39341afd4ea7bc5c3d7e27c33b67,
		spring-cloud-request-nonce:779104,
		spring-cloud-request-sign:F6C1B9586683C1FFCABDF912A1C57F512004F920B18CAB1BC9872B1BCC1C966D,
		traceparent:00-af99aca185386f2263a7e52bcdf1536c-66dd6081528be55a-00,
		content-type:application/json; charset=utf-8,
		x-request-id:4d013d8c-0de3-41cb-8f4c-6638ddcdaa40,
		x-envoy-attempt-count:1,
		x-envoy-original-path:/gw/open/sfe-affiliate-api/affiliate/open-list?pages=1&size=1000,
		content-length:72,
		forwarded:proto=http;host=10.66.8.197;for="192.168.84.177:37060",
		x-forwarded-client-cert:By=spiffe://cluster.local/ns/stable/sa/default;Hash=0b0446fbc3606fe738b8f314c12f79d36238e143cb8b2c1644b9c00f4e3d976f;Subject="";URI=spiffe://cluster.local/ns/stable/sa/default,
		x-forwarded-proto:http,http,
		x-forwarded-prefix:/open/sfe-affiliate-api,
		x-forwarded-port:80,
		x-forwarded-host:10.66.8.197,
		user-agent:ReactorNetty/1.0.10,
		accept:*/*
	],
	body: {"pages":1,"size":1000,"periodIds":["8ac2ae5e82fe851e0187b2238fa0022e"]},
	protocol: HTTP/1.1
}

接口性能

采用 Arthas排查接口耗时

步骤	执行	备注
1	marketing-sfe-affiliate-api-latest-558fbf674-lkjw4	pod
2	kubectl exec -it marketing-sfe-affiliate-api-latest-558fbf674-lkjw4 -c marketing-sfe-affiliate-api -n stable -- /bin/sh
3	wget https://alibaba.github.io/arthas/arthas-boot.jar	arthas
4	java -jar arthas-boot.jar

json

curl --location 'localhost:8080/affiliate/open-list?pages=1&size=1000' \
--header 'tm-header-tenantid: df2c39341afd4ea7bc5c3d7e27c33b67' \
--header 'spring-cloud-request-nonce: 779104' \
--header 'spring-cloud-request-sign: F6C1B9586683C1FFCABDF912A1C57F512004F920B18CAB1BC9872B1BCC1C966D' \
--header 'traceparent: 00-af99aca185386f2263a7e52bcdf1536c-66dd6081528be55a-00' \
--header 'content-type: application/json; charset=utf-8' \
--header 'x-request-id: 4d013d8c-0de3-41cb-8f4c-6638ddcdaa40' \
--header 'x-envoy-attempt-count: 1' \
--header 'content-length: 72' \
--header 'Cookie: JSESSIONID=4D81FAFACD926C8A2D036F2784388BFC; XSRF-TOKEN=0fd2143d-5210-4fcd-a15f-434d0a1b690d' \
--data '{
    "pages": 1,
    "size": 1000,
    "periodIds": [
        "8ac2ae5e82fe851e0187b2238fa0022e"
    ]
}'

json

trace com.taimei.sfe.affiliate.api.facade.AffiliateRelationFacade getAffiliateRelationByPage  -n 5 --skipJDKMethod false

json

trace com.taimei.sfe.core.affiliate.biz.AffiliateRelationService getAffiliateRelationByPage  -n 5 --skipJDKMethod false

json

trace com.taimei.sfe.core.affiliate.biz.AffiliateRelationService fillDisplayData  -n 5 --skipJDKMethod false

解决方案

批量查询字典数据后赋值，避免在大循环体内批量调用外部接口

性能优化 ​

机构挂靠接口性能优化 ​

问题/需求 ​

性能分析 ​

机构挂靠服务 ​

服务调用关系 ​

接口监控 ​

接口性能 ​

采用 Arthas排查接口耗时 ​

解决方案 ​

性能优化

机构挂靠接口性能优化

问题/需求

性能分析

机构挂靠服务

服务调用关系

接口监控

接口性能

采用 Arthas排查接口耗时

解决方案