[GitHub] [skywalking] aderm commented on issue #4364: Support Downsampling Data Packing feature in ES storage implementation

[GitBox <gi...@apache.org>]

aderm commented on issue #4364: Support Downsampling Data Packing feature in ES storage implementation
URL: https://github.com/apache/skywalking/pull/4364#issuecomment-586660956
 
 
   > About `writing to/querying from the merged index`, we always query/write from metrics in minute precision, if there is no performance issue there, there is no issue for merged index. Right?
   
   Regarding the query of the index, if 3 precisions of metrics are merged into the minute index, have two small problems:
   1. When the ES server loads index memory data, because lucene index storage is similar to columnar storage, the merge will take up more memory. Will timebucket or other field queries consume more resources? evaluate the memory resources at the minute level? Although the result set of the search is the same size;
   2. When the SW daily data volume is large, reaching a single index to the level of 10 million or 100 million, will the performance not change if the index conditions are not changed? Maybe more scenarios are minute-level queries.

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services