Support Questions
Find answers, ask questions, and share your expertise

data volume to processing ratio

data volume to processing ratio


So if i have 10 nodes in a cluster each with 100 gb RAM and 100 gb allocated to local dir (caches).

So is the total data volume that can be processed by the cluster 2 TB ie 200 GB X 10?

also does the local dir serve as the swap space - ie when the RAM goes full data is then move to the local dir. Is that the function of local dir?

And if i try to process 3 TB data , then will the data be flushed first from RAM to cache and then from cache to /dev/null?

Appreciate the insights.