Capacity testing is a crucial part of best-practices for high uptime.
Scale Up To 7,000 tasks in 20 AWS hosts
We were having some reports of agent failures while launching large numbers of tasks, so I did my own research. I was able to learn quite a bit about AWS and the safe speed parameters around these tasks.
AWS instances don’t have the horsepower of bare metal, so it pays to be a bit conservative when configuring Marathon for that scenario. With its roots in large-scale datacenter operations, Mesos can present quite the firehose to these EC2 units and we need to be conscious of this.
In addition to this public plog post, a great deal of learning was transferred to the internal development, sales engineering, and support teams.
- Category: Technology
- Date: 02.29.2016
- Tech: Mesos, Docker, AWS, NGiNX