# Preloading docs for cache warming **URL:** \ **Category:** Elasticsearch\ **Created:** [October 30, 2013, 1:56pm UTC](https://discuss.elastic.co/t/preloading-docs-for-cache-warming/14173 "2013-10-30T13:56:05Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![Val\_Vakar](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/val_vakar/32/2017_2.png) [@Val\_Vakar](https://discuss.elastic.co/u/Val_Vakar)\ **Post date:** [October 30, 2013, 1:56pm UTC](https://discuss.elastic.co/t/preloading-docs-for-cache-warming/14173/1 "2013-10-30T13:56:05Z")
Hello experts, My impression so far is that there are two notable bottlenecks in ES: 1. Disk I/O - loading documents 2. CPU pegging while searching/filtering Is there a good way to pre-load documents? In other words, if I don't want to figure out the cache warming queries, can I at least pre-empt the disk I/O problem? Thanks, Val -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Ivan](https://avatars.discourse-cdn.com/v4/letter/i/df788c/32.png) [@Ivan](https://discuss.elastic.co/u/Ivan)\ **Post date:** [October 31, 2013, 5:17pm UTC](https://discuss.elastic.co/t/preloading-docs-for-cache-warming/14173/2 "2013-10-31T17:17:45Z")
There should be a feature coming out in the next version of elasticsearch that will warm up new segments without needing a warming query. > > > Query-based warm-up is convenient as it doesn't require to know how things work …under the hood to warm-up a segment: people just have to put there queries that are similar to their users' queries. > > However, warm-up could be made more efficient through \`IndexWriterConfig.setMergedSegmentWarmer\`/\`IndexWriter.IndexReaderWarmer\` which doesn't require running a query at all and is run during the merging phase, before the new segment goes live. I think we should expose a way to register such warmers and maybe use something like \`SimpleMergedSegmentWarmer\` by default. Never used it myself (I don't work off master builds), but it might be what you are requiring. Cheers, Ivan On Wed, Oct 30, 2013 at 6:56 AM, Val Vakar [val.vakar@gmail.com](mailto:val.vakar@gmail.com) wrote: > Hello experts, > > My impression so far is that there are two notable bottlenecks in ES: > > 1. Disk I/O - loading documents > 2. CPU pegging while searching/filtering > > Is there a good way to pre-load documents? In other words, if I don't want > to figure out the cache warming queries, can I at least pre-empt the disk > I/O problem? > > Thanks, > Val > > -- > You received this message because you are subscribed to the Google Groups > "elasticsearch" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). > For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 6, 2017, 2:09am UTC](https://discuss.elastic.co/t/preloading-docs-for-cache-warming/14173/3 "2017-07-06T02:09:33Z")