*QUESTION* Has anyone fiddled around with the smallest possible instance size to run Scalr in EC2? What about breaking it into MySQL and everything else?
*DETAIL* For our Dev and QA we use the Community Edition all-in-one sized as an m3.medium. With 175 instances running inside it, I'm getting a CPU graph that looks like the attached image. One interesting data point is on 7/7 when the server wasn't properly operating ZeroMQ. CPU usage drops to around 30%. This server will be growing to over 1,000 nodes by the end of the year and it looks like I need to start breaking this server into multiple parts. Nice notes here <https://scalr-wiki.atlassian.net/wiki/display/docs/Production+Deployment+Recommendations> on a distributed architecture. <https://scalr-wiki.atlassian.net/wiki/display/docs/Production+Deployment+Recommendations> <https://scalr-wiki.atlassian.net/wiki/display/docs/Production+Deployment+Recommendations> <https://lh3.googleusercontent.com/-OHuXAX8d9Ng/V4T7W7__XsI/AAAAAAAABO4/6pA7H-JPYqUbh_oj5ERGt49SkSb7vhi5gCLcB/s1600/m3.medium_cpu.png> I'm asking about this because we have built a second server to manage production infrastructure. The number of instances controlled by this second server will be under 50. Curious how small I can go? t2.medium t2.small? Two t2.small instances, one for MySQL and one for everything else? -- You received this message because you are subscribed to the Google Groups "scalr-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to scalr-discuss+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
