ajayky-os opened a new pull request, #14873: URL: https://github.com/apache/iceberg/pull/14873
[Changelog](https://github.com/GoogleCloudPlatform/gcs-analytics-core/blob/main/CHANGELOG.md) of gcs-analytics-core: - Fix issue with vectored IO where read requests were not bounded, resulting in poor performance with larger data files. - Disable small object prefetch optimization by default, does not contribute to performance gain for larger data set. Consecutive Benchmark Results with updated version for 1 TB schema size: **benchmark**|**scan\_time (analytics core disabled)**|**scan\_time (analytics core enabled)**|**% Scan time improvement** :-----:|:-----:|:-----:|:-----: tpcds\_sf1000|29,946,605|19,370,576|35.32 tpcds\_sf1000|29,626,952|18,743,899|36.73 tpcds\_sf1000|30,714,037|19,084,079|37.87 tpcds\_sf1000|29,281,478|18,700,487|36.14 **benchmark**|**scan\_time (analytics core disabled)**|**scan\_time (analytics core enabled)**|**% Scan time improvement** :-----:|:-----:|:-----:|:-----: tpch\_sf1000|16,753,626|13,833,931|17.43 tpch\_sf1000|17,615,900|14,144,223|19.71 tpch\_sf1000|17,780,989|13,958,524|21.50 tpch\_sf1000|17,025,561|13,746,540|19.26 *scan_time is total of BatchScan nodes for all queries. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
