# Logstash HTTP polling performance **URL:** \ **Category:** Logstash\ **Created:** [June 6, 2017, 7:15am UTC](https://discuss.elastic.co/t/logstash-http-polling-performance/88368 "2017-06-06T07:15:32Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![juri](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/juri/32/18816_2.png) [@juri](https://discuss.elastic.co/u/juri)\ **Post date:** [June 6, 2017, 7:15am UTC](https://discuss.elastic.co/t/logstash-http-polling-performance/88368/1 "2017-06-06T07:15:32Z")
Hello, In my use case I need to periodically query huge population of devices(~1M) via HTTP REST API and store status of each device in elasticsearch. I consider to use http\_poller but I worry about perfomance and scalability of this solution. Does anyone have more experience with this input type ? ---
**Author:** ![magnusbaeck](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/magnusbaeck/32/44943_2.png) [@magnusbaeck](https://discuss.elastic.co/u/magnusbaeck)\ **Post date:** [June 8, 2017, 7:29pm UTC](https://discuss.elastic.co/t/logstash-http-polling-performance/88368/2 "2017-06-08T19:29:33Z")
I suspect Logstash uses one thread per input plugin, so using one http\_poller per device clearly won't scale. I don't think a pure Logstash solution is appropriate here. You need something else to do the HTTP polling, but Logstash could still read the results of that polling and process it further. ---
**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, 7:29pm UTC](https://discuss.elastic.co/t/logstash-http-polling-performance/88368/3 "2017-07-06T19:29:40Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.