I am studying distributed computing these days, and was curious to see how Memcached works. The Wiki says :
memcached is a general-purpose distributed memory caching system
A look at the architecture explains memory caching, but I fail to understand the way "distributed computing" works here. The servers are not even aware of each other – what about fault tolerance !
Can someone please help me get a "flow" of how data flows in such system ? And how does the client understand which server holds the required data (in case of a cache miss)
Thanks !
I have been trying to figure out the very same thing. As best as I can figure, distributed in this sense does not imply replication as one would expect with modern, distributed, cloud-like applications. In the sense that memcached is distributed, it seems to simply mean it is not necessarily local.
A similar question was asked, specifically detailing a use case with PHP, with an answer that addresses the memcached claim of being distributed.