Spark is an in-memory engine and attempts to do computation in-memory.
Tachyon is memory-centeric distributed storage, OK, but how would that help
ran Spark faster?

Reply via email to