kubernetes - Resource metrics (memory, CPU etc) should I be looking at for auto scaling purpose -


what cloud resource metrics (memory, cpu, disk io etc) should looking @ auto scaling purpose? fyi, metrics strictly used auto scaling purpose. have kubernetes architecture , prometheus (for monitoring , scraping metrics)

i have kubernetes cluster set in local cloud. using prometheus tool(https://prometheus.io/) set scraping system level metrics. now, want have auto-scaling feature in system. have been using prometheus saving metrics this. "memory , cpu used, allocated, total last 24 hours." want save more metrics. list of metrics getting prometheus: http://demo.robustperception.io:9100/metrics can't decide more metrics going need auto scaling purpose. can suggest metrics purpose? tia.

normally, common bottleneck memory hierarchy rather cpu usage. more requests application receives, more have out-of-memory error. more, if application not hpc, not needs cpu-intensive.

in memory hierarchy, disk i/o can dramatically affect performance. need check how disk i/o intensive application is. in sense, changing disk hardware better solution rather spinning more instances. however, depends on application.

in case, interesting if measure average response time, , take decisions accordingly.


Comments

Popular posts from this blog

python Tkinter Capturing keyboard events save as one single string -

android - InAppBilling registering BroadcastReceiver in AndroidManifest -

javascript - Z-index in d3.js -