[Beowulf] Large amounts of data to store and process

Prentice Bisbal pbisbal at pppl.gov
Fri Mar 15 10:02:38 PDT 2019


On 3/15/19 3:35 AM, Benjamin Redling wrote:
> On 15.03.19 06:28, Jonathan Aquilina wrote:
>> I think what I was getting at is why not include the current HPC practices to every day desktops in the sense since we are reaching certain limits and have to write code to take advantage of more and more cores. Why not use MPI and the like to help distribute the software side of things to the cores?
>>
>> It could be my entire concept of MPI is way of or im misunderstanding completely how it works.
> Not my opinion -- I have none regarding MPI --  and one doesn't have to
> agree with the following article, but it gives a glimpse at what seem to
> be the weaknesses of MPI:
> https://www.dursi.ca/post/hpc-is-dying-and-mpi-is-killing-it.html
>
> Regards,
> Benjamin

Wow. That article is completely misguided. MPI doesn't work for data 
analytics and genomics because those are completely different problem 
types from what MPI was designed for. He's comparing apples to cats.

--
Prentice



More information about the Beowulf mailing list