[ 
https://issues.apache.org/jira/browse/IMPALA-8542?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16866820#comment-16866820
 ] 

ASF subversion and git services commented on IMPALA-8542:
---------------------------------------------------------

Commit f4ef9ca87352406cc7a484e90ed1d1d702745810 in impala's branch 
refs/heads/master from Todd Lipcon
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=f4ef9ca ]

IMPALA-8542. Add an access trace for the data cache

This adds a relatively simple JSON-formatted access trace for the data
cache feature. Each partition stores a trace file named 'trace.txt',
with each line representing a hit, miss, or store into the cache.

The trace is collected using the kudu::AsyncLogger class which handles
buffering and deferring the actual IO to a background thread.

By default, the full cache key info is written to the trace (including
the file paths), but a flag can enable anonymization (128-bit
city-hashing) of the file names in case any user would like to capture a
trace to be shared publically without divulging their table names.

Change-Id: I2302c19abb5db19f1d3d1cd727a82977a9e2ba9c
Reviewed-on: http://gerrit.cloudera.org:8080/13425
Tested-by: Impala Public Jenkins <impala-public-jenk...@cloudera.com>
Reviewed-by: Michael Ho <k...@cloudera.com>


> Access trace collection for data cache
> --------------------------------------
>
>                 Key: IMPALA-8542
>                 URL: https://issues.apache.org/jira/browse/IMPALA-8542
>             Project: IMPALA
>          Issue Type: Improvement
>          Components: Backend
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>            Priority: Major
>             Fix For: Impala 3.3.0
>
>
> Now that we have a remote-read data cache, it would be useful to log an 
> access trace. The trace can be then fed back into various cache policy 
> simulators to compare the relative performance, and do "what if" analysis 
> (how would hit rate react with larger/smaller capacities)



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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-all-unsubscr...@impala.apache.org
For additional commands, e-mail: issues-all-h...@impala.apache.org

Reply via email to