[
https://issues.apache.org/jira/browse/PHOENIX-258?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15412436#comment-15412436
]
Lars Hofhansl commented on PHOENIX-258:
---------------------------------------
That does not at all match my experience. It only needs to do 2 (perhaps 3)
seeks per chunk. So at most 234 seeks, which on top happen in parallel. That
should take a few *milli* seconds.
OK... Further it needs to seek each HFile/Memstore. Was the table just seeded?
(Even then I'd expect maybe a few thousand seeks, and still in parallel)
If there are many deleted rows, that might over shadow the savings here, but I
doubt that's the case.
How long does {{SELECT /*+ RANGE_SCAN */ DISTINCT ORGANIZATION_ID FROM T}} take?
> Use skip scan when SELECT DISTINCT on leading row key column(s)
> ---------------------------------------------------------------
>
> Key: PHOENIX-258
> URL: https://issues.apache.org/jira/browse/PHOENIX-258
> Project: Phoenix
> Issue Type: Task
> Reporter: ryang-sfdc
> Assignee: Lars Hofhansl
> Fix For: 4.8.0
>
> Attachments: 258-WIP.txt, 258-v1.txt, 258-v10.txt, 258-v11.txt,
> 258-v12.txt, 258-v13.txt, 258-v14.txt, 258-v15.txt, 258-v16.txt, 258-v17.txt,
> 258-v2.txt, 258-v3.txt, 258-v4.txt, 258-v5.txt, 258-v6.txt, 258-v7.txt,
> 258-v8.txt, 258-v9.txt, 258.txt, DistinctFixedPrefixFilter.java, in-clause.png
>
>
> create table(a varchar(32) not null, date date not null constraint pk primary
> key(a,date))
> [["PLAN"],["CLIENT PARALLEL 94-WAY FULL SCAN OVER foo"],[" SERVER
> AGGREGATE INTO ORDERED DISTINCT ROWS BY [a]"],["CLIENT MERGE SORT"]]
>
> We should skip scan.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
