☰

Running Jobs and Using the Scheduler

You are not logged in.
Please Log In for full access to the web site.
Note that this link will take you to an external site (https://petrock.mit.edu) to authenticate, and then you will be redirected back to this page.

Review the concepts from the Distributed Applications module

1) Parallel Computing

In a shared memory application, all of the processors performing computation (Select all that apply)

read and write to memory that is only visible to them

read and write to memory visible to all of the processors

must be co-located on the same compute node

can be located anywhere on the HPC system

In a distributed memory application, all of the processors performing computation (Select all that apply)

read and write to memory that is only visible to them

read and write to memory visible to all of the processors

must be co-located on the same compute node

can be located anywhere on the HPC system

2) MPI

In a traditional distributed parallel computing model the processors share information via

shared memory on the compute node

the file system

by sending and receiving messages using a Message Passing Interface Library

In order to run an MPI application, which of the following is required? (Check all that apply.)

The code must be written with MPI functions

The code must be built with an MPI library

Your environment must include the same version of MPI that was used to build the code

The processing must use threads