We just completed a move of a physical Data Center in to the Amazon Web Services Cloud. We decided that we were going to deploy the Solarwinds server agent for monitoring as it made monitoring much easier across VPN connections and between VPCs in the cloud. Once we went live, however, we noticed that the agent was consuming 500MB of Ram on every server. Because of the costs associated with increasing the RAM/CPU on servers, the Systems teams had tuned the servers to minimize requirements in the cloud and control costs. We were planning on a 50-60M hit using the Agents, but 500MB is far too much to support.
Is there any push to optimize the agent and try to get Memory usage down to something a little more palatable? The Agent satisfies a lot of requirements for server security and ease of use across security zones, and I really want to use it.
