Good afternoon all,
Is there any documentation / recommendations as to the optimal Cluster size for an ArcGIS Monitor implementation?
e.g. We have an 8 Virtual processor VM with Monitor and PostgreSQL deployed, monitoring 12 servers across 3 environment.
Any advice, as always - gratefully received.
Many thanks
Matt
Solved! Go to Solution.
Note, the Server configuration Settings are related to user https calls to the server, not the agents that sent request to the monitored components. You should consider increasing the cluster size if there are many concurrent users connecting to the server, regardless of how big or small is your monitored environment. If the primary users are a handful of admins, the default should be sufficient. However, if we expose ArcGIS Monitor to a large number of concurrent users, e.g. we shared several analyses with a large group, and these users experience degraded performance, you should consider increasing the cluster size and potentially increase machine CPU resources. You might also need to increase PostgreSQL resources, e.g. CPU cores and shared_buffers, see https://www.postgresql.org/docs/9.1/runtime-config-resource.html
This cluster size
Note, the Server configuration Settings are related to user https calls to the server, not the agents that sent request to the monitored components. You should consider increasing the cluster size if there are many concurrent users connecting to the server, regardless of how big or small is your monitored environment. If the primary users are a handful of admins, the default should be sufficient. However, if we expose ArcGIS Monitor to a large number of concurrent users, e.g. we shared several analyses with a large group, and these users experience degraded performance, you should consider increasing the cluster size and potentially increase machine CPU resources. You might also need to increase PostgreSQL resources, e.g. CPU cores and shared_buffers, see https://www.postgresql.org/docs/9.1/runtime-config-resource.html
This cluster size
Thanks Andrew - greatly appreciated