mikewalch commented on a change in pull request #92: Document Scan Executors URL: https://github.com/apache/accumulo-website/pull/92#discussion_r201510738
########## File path: _docs-2-0/administration/scan-exec.md ########## @@ -0,0 +1,113 @@ +--- +title: Scan Executors +category: administration +order: 11 +--- + +Accumulo scans operate by repeatedly fetching batches of data from a [tablet +server][tserver]. On the tablet server side, a thread pool fetches batches. +In Java threads pools are called executors. By default, a single executor per +tablet server handles all scans in FIFO order. For some workloads, the single +FIFO executor is suboptimal. For example, consider many unimportant scans +reading lots of data mixed with a few important scans reading small amounts of +data. The long scans noticeably increase the latency of the short scans. +Accumulo offers two mechanisms to help improve situations like this: multiple +scan executors and per executor prioritizers. Additional scan executors can +give tables dedicated resources. For each scan executor, an optional +prioritizer can reorder queued work. + +### Configuring and using Scan Executors + +To configure additional scan executors, chose a unique name and set the +following property. This causes each tablet server to create a scan executor +with the specified threads. By default, Accumulo sets Review comment: `By default, Accumulo sets...` should be first sentence in paragraph. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
