Hi, http://apacheignite.gridgain.org/v1.5/docs/load-balancing
I found above documentation on load balancing but i am still not clear and have following queries 1) How load is balanced between server nodes when a lot of clients are trying to run queries or get / put values in a cache? do we consider each query / get / put as a task and by default use round robin scheduling? 2) How load is managed in more complex scenarios where data affinity also has to be considered and caches are replicated? if round robin scheduling is used on a cluster with 2 server node and if multiple clients try to fetch same cache value (by id) from a replicated cache then half of the clients will try fetching data from a server node which may not have it. 3) What are different ways in which i can override or customize load balancing? 4) who makes the decision that which server node to be used to fetch or put data in cache? is it client node? Any links to documentation or code explaining above points will be a great help. Regards, Vinay Sharma -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Data-grid-Load-balancing-tp2819.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.
