[jira] [Comment Edited] (JENA-1550) Bulk loader for TDB2.

["Andy Seaborne (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/JENA-1550?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16481944#comment-16481944 ] 

Andy Seaborne edited comment on JENA-1550 at 5/20/18 1:34 PM:
--------------------------------------------------------------

The speed jumps in parse to node table don't free anything until the end of the parser. The count output is coming from the parser.

There is a chance it is due to the data - reaching a point where there is a lower density of less new nodes .

But also it could be that it is another level of JIT compilation cutting in. Up to 25M, file caching is comfortable within memory so the task is not I/O bound.  The memory bandwidth goes up with increased speed so that leaves CPU.



was (Author: andy.seaborne):
The speed jumps in parse to node table don't free anything until the end of the parser. The count output is coming from the parser.

My guess is that it is another level of JIT compilation cutting in. Up to 25M, file caching is comfortable within memory so the task is not I/O bound.  The memory bandwidth goes up with increased speed so that leaves CPU.

> Bulk loader for TDB2.
> ---------------------
>
>                 Key: JENA-1550
>                 URL: https://issues.apache.org/jira/browse/JENA-1550
>             Project: Apache Jena
>          Issue Type: Improvement
>          Components: TDB2
>    Affects Versions: Jena 3.7.0
>            Reporter: Andy Seaborne
>            Assignee: Andy Seaborne
>            Priority: Major
>
> Provide a bulk loader for TDB2.



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