Proxy repository server

Cache repository contents if replication is not an option.

Although Plastic SCM is all about distributed development, we do understand there are situations in which going distributed is not an option for your team. If that's the case, you can stick with using a central repository server, but use a proxy server at some remote sites to speed up access.

The proxy server runs as service/daemon on any supported platform. It caches incoming revision data from the central server, in order to reduce network traffic on subsequent read operations. It is very simple to set up, and only requires some spare disk space to store the compressed revisions. (It has no database dependencies, in order to stay as slim as possible.)

How does it work?

Once the client is configured to use a proxy server, it sends requests for revision data to the proxy instead of the central server, greatly reducing network traffic. The proxy server determines whether it already has the requested data, and if not, passes the request on to the central server. The proxy server is able to handle concurrent requests, so it contacts the central server just once for each revision, even if several clients request it in parallel.

Once the proxy server downloads the data and caches it locally, it sends the data to the client. Subsequent requests for the same data obtain it directly from the cache.

Configuring a client to use the proxy server is simple: just specify the hostname and port:

configureproxyserver

If the client can't contact the proxy server, it transparently connects to the central server.