phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lars Hofhansl (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (PHOENIX-5349) Speed up AlterTableWithViewsIT
Date Sat, 15 Jun 2019 22:01:00 GMT

     [ https://issues.apache.org/jira/browse/PHOENIX-5349?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Lars Hofhansl updated PHOENIX-5349:
-----------------------------------
    Description: 
This is - after ViewIT - the slowest test we have.

Right now we're testing all combinations of columnEncoded, multiTenant, and salted, i.e. 2^3
combinations.

I think that is not necessary. We can instead do the following four combinations instead:
(false, false, false), (true, false, true), (true, true, false), (true, true, true)

We'd cover no-columnencoding only once, and make sure we have the other three remaining combinations
for a total of 4. That will cut the runtime to 1/2.

  was:
This is - after ViewIT - the slowest test we have.

Right now we're testing all combinations of columnEncoded, multiTenant, and salted, i.e. 2^3
combinations.

I think that is not necessary. We can instead do the following four combinations instead:
{ false, false, false }, { true, false, true }, { true, true, false }, { true, true, true
}

We'd cover no-columnencoding only once, and make sure we have the other three remaining combinations
for a total of 4. That will cut the runtime to 1/2.


> Speed up AlterTableWithViewsIT
> ------------------------------
>
>                 Key: PHOENIX-5349
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-5349
>             Project: Phoenix
>          Issue Type: Sub-task
>            Reporter: Lars Hofhansl
>            Priority: Major
>
> This is - after ViewIT - the slowest test we have.
> Right now we're testing all combinations of columnEncoded, multiTenant, and salted, i.e.
2^3 combinations.
> I think that is not necessary. We can instead do the following four combinations instead:
> (false, false, false), (true, false, true), (true, true, false), (true, true, true)
> We'd cover no-columnencoding only once, and make sure we have the other three remaining
combinations for a total of 4. That will cut the runtime to 1/2.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message