You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Taras Ledkov (JIRA)" <ji...@apache.org> on 2018/09/04 15:34:02 UTC
[jira] [Commented] (IGNITE-9171) Use lazy mode with results
pre-fetch
[ https://issues.apache.org/jira/browse/IGNITE-9171?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16603198#comment-16603198 ]
Taras Ledkov commented on IGNITE-9171:
--------------------------------------
[~vozerov], please take a look at the benchmark results.
Benchmark configuration:
- 4 servers, 8 clients.
- 1 backups;
Benchmark names agenda:
- rN - results set size (one row, 1K and 2k rows, 1M rows)
- lazy - lazy mode on.
Operations per second:
|| || master || ignite-9171 ||
| r1 | 154,550.61 | 153,534.05 |
| r1K | 6,097.21 | 6,027.85 |
| r2K | 3,523.90 | 3,478.26 |
| r1M | 16.51 | 16.45 |
| r1-lazy | 24,474.02 | 146,916.20 |
| r1K-lazy | 2,014.16 | 6,043.35 |
| r2K-lazy | 1,839.11 | 3,503.99 |
| r1M-lazy | 21.09 | 21.00 |
> Use lazy mode with results pre-fetch
> ------------------------------------
>
> Key: IGNITE-9171
> URL: https://issues.apache.org/jira/browse/IGNITE-9171
> Project: Ignite
> Issue Type: Improvement
> Components: sql
> Affects Versions: 2.6
> Reporter: Taras Ledkov
> Assignee: Taras Ledkov
> Priority: Major
> Labels: sql-stability
> Fix For: 2.7
>
>
> Current implementation of the {{lazy}} mode always starts separate thread for {{MapQueryLazyWorker}}. It causes excessive overhead for requests that produces small results set.
> We have to begin execute query at the {{QUERY_POOL}} thread pool and fetch first page of the results. In case results set is bigger than one page {{MapQueryLazyWorker}} is started and link with {{MapNodeResults}} to handle next pages lazy.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)