• Laeeth Isharc (16/16) Oct 29 2016 From Jan 2016:
• Kagamin (2/2) Oct 31 2016 On the other hand if you do more IO, you can have higher CPU load

Laeeth Isharc <laeeth-nospam nospamlaeeth.com> writes:

 From Jan 2016:

https://queue.acm.org/detail.cfm?id=2874238

For the entire careers of most practicing computer scientists, a 
fundamental observation has consistently held true: CPUs are 
significantly more performant and more expensive than I/O 
devices. The fact that CPUs can process data at extremely high 
rates, while simultaneously servicing multiple I/O devices, has 
had a sweeping impact on the design of both hardware and software 
for systems of all sizes, for pretty much as long as we've been 
building them.

This assumption, however, is in the process of being completely 
invalidated.
===
Even today, a Spark committer says that for Spark CPU is already 
often the bottleneck (for decompression):

https://www.oreilly.com/ideas/investigating-sparks-performance

Kagamin <spam here.lot> writes:

On the other hand if you do more IO, you can have higher CPU load 
due to compression and serialization.