874580011 opened a new issue, #12930:
URL: https://github.com/apache/druid/issues/12930

   This is the case. At present, the company always uses HTTP to request 
Druids, but the problem is also very obvious. HTTP makes a query cumbersome. 
Can we make the Druid request as simple as JDBC. Therefore, I learned that 
Druid was requested by JDBC, and everything was very smooth by integrating 
mybatis plus. Just when I wanted to finish work, the problem occurred。By means 
of jemter pressure measurement,I found that the speed of HTTP query for the 
same SQL is 10 times faster than that of JDBC. The pressure test configuration 
is 10 threads,After checking the relevant information, I still can't get the 
answer, so please help me to see what the situation is
   
   HTTP request 1 is jdbc test
   HTTP request2 is http test
   
   <img width="1001" alt="企业微信截图_7e585721-3363-4b2e-9cb5-45afd398de82" 
src="https://user-images.githubusercontent.com/53367253/185832668-6481b32c-abfa-4a95-9b9c-74eb9a328aba.png";>
   
   `@GetMapping("testList1")
       public String list1() {
           StopWatch stopWatch = new StopWatch();
           stopWatch.start();
           List<ApmServiceMetric> appExceptionDataMinList = 
druidMapper.getApmService();
           stopWatch.stop();
           System.out.println("stopWatch.getTotalTimeSeconds() = " + 
stopWatch.getTotalTimeSeconds());
           for (ApmServiceMetric apmServiceMetric : appExceptionDataMinList) {
               System.out.println(apmServiceMetric);
           }
           return "success";
       }`
   
   `@Options(statementType = StatementType.STATEMENT)
       List<ApmServiceMetric> getApmService();`
   
   `<select id="getApmService" resultType="com.tingyun.entity.ApmServiceMetric">
           select
               (CURRENT_TIMESTAMP - INTERVAL '5' MINUTE) as "datetime",
               TIMESTAMP_TO_MILLIS(CURRENT_TIMESTAMP - INTERVAL '5' MINUTE) as 
"timestamp",
               biz_system_id as "bizSystemId",
               application_id as "applicationId",
               sum("count") / 5.0 as "APMServiceCPM",
               sum("count") as "APMServiceCallCount",
               sum("count") - sum("success_count") as "APMServiceFailCount",
               (sum("resp_time") * 1.0) / sum("success_count") as 
"APMServiceAvgResponseTime"
           from SVR_ACTION_DATA_MIN
           where
               "data_type" = 'APP'
               and "__time" > CURRENT_TIMESTAMP - INTERVAL '10' MINUTE and 
"__time" <![CDATA[ <= ]]> CURRENT_TIMESTAMP - INTERVAL '5' MINUTE
           group by
               biz_system_id, application_id
       </select>`
   
   
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to