I am just curious to know how the below situation is handled at architecture level of Memcached:
Assume we have configured 4 server to hold our key-value pairs that are cached and we have dedicated equal amount of memory for this purpose, say 1GB per server. During the course of our dynamic web page we cached many results into our memcache system by following good hashing function that is likely to distribute all the keys to all the server. At a given point in time, lets assume we have .5GB of data cached in each of the four server. For some reason, say we want to remove the fourth server from the system but still want to retain the key-value pairs stored in it to be accessible at high speed? How do we re-distribute it among other 3 servers (which still has 0.5GB of storage each)? I am new to memcached. Please let me know if my thinking in terms of system architecture, or scenario is incorrect... Will be waiting for the feedback from the mailing list :) -- --- You received this message because you are subscribed to the Google Groups "memcached" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
