AvinoamBN opened a new issue #9017: help understanding some basics to overcome 
query lag
URL: https://github.com/apache/incubator-druid/issues/9017
 
 
   Hi Gian, thank for reaching out.
   So I'm kinda new to apache druid, and I've been assigned with finding the 
source of a lag when querying.
   so far, I've learnt how to read the cluster conf files and how to query for 
metrics.
   I was trying to follow the docs to check how I can fine tune the 
configuration and optimize the historical and broker, but some options values 
are a mystery to me.
   like "heap size" and `druid.cache.sizeInBytes` and `druid.server.maxSize`.
   
   1. to reach max heap size of 24G RAM, do i need a 48 cpu cores server?
   2. what is the free system memory / druid.server.maxSize ratio? the docs 
mention "should be set such that a Historical is not allocated an excessive 
amount of segment data"... what is considered excessive, what the formula?
   
   regarding querying the metrics datasource:
   I'm doing a select query with mysql, and sending it via POST to the broker.
   example query:
   ```
   {
   "query":
   "SELECT * FROM \"metrics\" 
       WHERE \"__time\" BETWEEN 
           TIMESTAMP '2019-12-11 00:00:00' AND 
           TIMESTAMP '2019-12-11 23:59:59' 
         AND 
           \"dataSource\" = 'orgrimmar'
         AND
           \"metric\" = 'query/time'
         AND 
           \"type\" = 'select'
         AND
            \"value\" > 10000 
     ORDER BY 1 limit 10"
   }
   ```
   
   this query give results that look like this:
   
   ```
   {
     "__time": "2019-12-11T15:00:49.878Z",
     "count": 1,
     "dataSource": "orgrimmar",
     "hasFilters": "true",
     "host": "druid-broker1:8083",
     "interval": "2019-11-26T10:18:51.681Z/2019-12-11T15:00:34.664Z",
     "metric": "query/time",
     "numComplexMetrics": 0,
     "numDimensions": 0,
     "numMetrics": 0,
     "server": "",
     "type": "select",
     "value": 13288
   }
   ```
   
   How can i find out why did this query take 13 seconds to complete?
   How can i find out what and where is the bottle-neck?

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


With regards,
Apache Git Services

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

Reply via email to