In regards to OutputCache
, "client" caching simply means that cache-control headers and/or an expires header will be sent with the response, indicating that the client may cache the document. Typically, the client, especially if it's a web browser, will choose to do so. It then will not need to make a new request if the same resource is needed again. However, the browser may still occasionally make a HEAD request just to check if there's a new version of the resource.
"Server" caching means, still in regards to OutputCache
, that the server will cache the response locally, usually in memory. This means that as long as the cache is still valid, the server will not actually render the action again, but rather, will just serve up the cached resource, instead.
The main difference, then, between the two is that the server cache will be used for all requests for that resource, regardless of what client is currently making the request, while client cache will obviously just be limited to that one particular client. The server will not need to render the action again for that client, but will for the next client that comes along.
However, the default is Any
, which includes server and client caching (as well as other locations). In other words, server and client caching are not mutually exclusive, and usually you'd do both to minimize both the work the server needs to do and the amount of requests it needs to respond to.