[
https://issues.apache.org/jira/browse/EAGLE-542?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hao Chen updated EAGLE-542:
---------------------------
Description:
Eagle hbase scan takes 3 seconds fro 100k records which is because it scans
sequential TS rows.
We should resolve this problem to use eagle hbase as metric store.
As to eagle query engine, we maybe could simply add an annotation
@Bucket(Number) so that we could scan with buckets number of scanner
concurrently.
The storage guarantee is:
* Different bucket should be put into different region
* Better to have is these regions are separated into different region servers.
* Scan bucketed services with multiple concurrent scanner
* Merge & sort result.
Reference:
* https://github.com/sematext/HBaseWD
* http://opentsdb.net/docs/build/html/user_guide/writing.html
was:
Eagle hbase scan takes 3 seconds fro 100k records which is because it scans
sequential TS rows.
We should resolve this problem to use eagle hbase as metric store.
As to eagle query engine, we maybe could simply add an annotation
@Bucket(Number) so that we could scan with buckets number of scanner
concurrently.
The storage guarantee is different bucket should be put into different region,
better to have is these regions are separated into different region servers.
> Support @Bucket annotation in query engine
> -------------------------------------------
>
> Key: EAGLE-542
> URL: https://issues.apache.org/jira/browse/EAGLE-542
> Project: Eagle
> Issue Type: New Feature
> Affects Versions: v0.5.0
> Reporter: Hao Chen
> Assignee: wujinhu
> Fix For: v0.5.0
>
>
> Eagle hbase scan takes 3 seconds fro 100k records which is because it scans
> sequential TS rows.
> We should resolve this problem to use eagle hbase as metric store.
> As to eagle query engine, we maybe could simply add an annotation
> @Bucket(Number) so that we could scan with buckets number of scanner
> concurrently.
> The storage guarantee is:
> * Different bucket should be put into different region
> * Better to have is these regions are separated into different region servers.
> * Scan bucketed services with multiple concurrent scanner
> * Merge & sort result.
> Reference:
> * https://github.com/sematext/HBaseWD
> * http://opentsdb.net/docs/build/html/user_guide/writing.html
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
