lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ferenczi Jim (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (LUCENE-7579) Sorting on flushed segment
Date Thu, 01 Dec 2016 17:42:58 GMT

    [ https://issues.apache.org/jira/browse/LUCENE-7579?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15712572#comment-15712572
] 

Ferenczi Jim commented on LUCENE-7579:
--------------------------------------

I ran the test from a clean state and I can see a nice improvement with the sparsetaxis use
case. 

I use https://github.com/mikemccand/luceneutil/blob/master/src/python/sparsetaxis/runBenchmark.py
and compare two checkouts of Lucene, one with my branch and the other with master.
For the master branch I have:
{noformat}
838.0 sec:  20.0 M docs;  23.9 K docs/sec
{noformat}

... vs the branch with the flush sort:
{noformat}
 612.2 sec:  20.0 M docs;  32.7 K docs/sec
{noformat}

I reproduce the same diff on each run :)




> Sorting on flushed segment
> --------------------------
>
>                 Key: LUCENE-7579
>                 URL: https://issues.apache.org/jira/browse/LUCENE-7579
>             Project: Lucene - Core
>          Issue Type: Bug
>            Reporter: Ferenczi Jim
>
> Today flushed segments built by an index writer with an index sort specified are not
sorted. The merge is responsible of sorting these segments potentially with others that are
already sorted (resulted from another merge). 
> I'd like to investigate the cost of sorting the segment directly during the flush. This
could make the merge faster since they are some cheap optimizations that can be done only
if all segments to be merged are sorted.
>  For instance the merge of the points could use the bulk merge instead of rebuilding
the points from scratch.
> I made a small prototype which sort the segment on flush here:
> https://github.com/apache/lucene-solr/compare/master...jimczi:flush_sort
> The idea is simple, for points, norms, docvalues and terms I use the SortingLeafReader
implementation to translate the values that we have in RAM in a sorted enumeration for the
writers.
> For stored fields I use a two pass scheme where the documents are first written to disk
unsorted and then copied to another file with the correct sorting. I use the same stored field
format for the two steps and just remove the file produced by the first pass at the end of
the process.
> This prototype has no implementation for index sorting that use term vectors yet. I'll
add this later if the tests are good enough.
> Speaking of testing, I tried this branch on [~mikemccand] benchmark scripts and compared
master with index sorting against my branch with index sorting on flush. I tried with sparsetaxis
and wikipedia and the first results are weird. When I use the SerialScheduler and only one
thread to write the docs,  index sorting on flush is slower. But when I use two threads the
sorting on flush is much faster even with the SerialScheduler. I'll continue to run the tests
in order to be able to share something more meaningful.
> The tests are passing except one about concurrent DV updates. I don't know this part
at all so I did not fix the test yet. I don't even know if we can make it work with index
sorting ;).
>  [~mikemccand] I would love to have your feedback about the prototype. Could you please
take a look ? I am sure there are plenty of bugs, ... but I think it's a good start to evaluate
the feasibility of this feature.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org


Mime
View raw message