In ‘High performance computer (HPC)’ resource environment master nodes is responsible to receive computational jobs submitted by users and transfer it to any FREE state compute nodes, How HPC knows which node is free which node is in use so ‘HPC job scheduler‘ helps HPC machine this task, There are several job scheduler available in the market here will talk about TORQUE, SLURM which one can use, Resource manager program called (PBS) Portable Batch System.
TORQUE, SLURM use to submit computational jobs through job script to perform actual computing operations.
PBS: Portable Batch System
PBS is a queueing system similar to what is implemented at supercomputer centers, Jobs are submitted to the queue that reflects the resources needed, and a scheduler decides which ones to run when nodes become available. These decisions are made on the basis of length of run, how long a job has been waiting, and fair sharing of resources among different users.
What is Serial programs in HPC
In term of HPC computation serial programs are fine for trivial applications with a very simple calculation that does not require much processing computing power. in case you are in a situation where you would like to to run high computational programs that are time demanding and needs much more processing power. In that cases, we may have to consider to go for parallel to achieve results faster.
What about Parallel Programs in HPC
A program that can be run on more than one node has distributed memory parallelization, A parallel job uses more than one core.
In case of Parallel computing involves having two or more processors to solve a single problem. Here multiple tasks are being executed by multiple CPUs. As we add more CPUs we can have more faster the tasks which can be done in less time.
In case of parallel programming we should remember that if the program that doesn’t have any form of Parallelization or Parallel API implementation can not take advantage of utilizing multiple CPUs, user has to wright his/her program accordingly.