CA Threat Manager can scale in size to fit any organization. The following list provides scaling recommendations based on three organizational sizes.
For a small organization with fewer than 500 users residing in a single location a 100 MB network infrastructure offers sufficient bandwidth to maintain the CA TM processes.
Recommendation: Use a dedicated server to function as the CA TM Server. Use at least one additional server acting as both a Policy Proxy Server and a Redistribution Server.
Medium-sized organizations of 500 to 5,000 users with one or multiple office locations. Some organizations may even extend outside national borders.
Recommendation: A single CA TM Server should be located in the IT department. Use additional Policy Proxy Servers and Redistribution Servers located in regional offices to distribute policies and updates. This ensures a better overall view and management of the data in the organization.
Large organizations with over 5,000 users can be nationwide or global. In some industries there are various divisions with their own wide area networks (WANs) running within the enterprise.
Recommendation: A single CA TM Server should be located in the IT department at corporate headquarters. Regional offices should be identified as points from which distribution of policy and updates can be performed. These regions should have one or more Policy Proxy Servers and one or more Redistribution Servers.
For example, one or more servers can be designated as Redistribution Servers for clients in remote locations that are likely closer to the regional office than the IT department at corporate headquarters. This approach has obvious benefits, as remote clients have a local place to check for policy and content updates, reducing network resource consumption and latency.
Furthermore, redundancy is critical to the scalability of the processes. By identifying multiple servers in a location and taking advantage of intermediary locations between the remote clients and the corporate data center, updates to signatures, engines, policies, and so on can be more efficiently managed.