I have a little experience with SungridEngine and MPI (using OpenMPI).
Whats the different between these frameworks/API and JPPF?
I have a little experience with SungridEngine and MPI (using OpenMPI). Whats the different
Share
Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
I have a little experience with SungridEngine and MPI (using OpenMPI).
Whats the different between these frameworks/API and JPPF?
All three of these are somehow related to parallel computing, but on pretty different levels.
The Sun Grid Engine (SGE) is a queueing system. It is usually set up by the system administrator of a big computing site, and allows users to submit long-running computing “jobs”. SGE checks whether any computing nodes are unoccupied, and if they are, it starts the job on that machine, otherwise the job will have to wait in the queue until a machine becomes available. SGE mainly cares about correct distribution of the jobs. For a single user, SGE is of very limited use. SGE is often used in high performance computing to schedule the user jobs.
JPPF is a Java framework which can help an application developer to run and implement a parallel Java program. It allows a Java application to run independent parts of it on other machines in parallel. It is useful to split a computing-intensive Java application into several mostly independent parts (which are typically called “tasks”). Although I do not really know the framework, I guess that it is mostly used to distribute big business applications onto several computers.
MPI (Message Passing interface) is an API (mainly for C/FORTRAN, but bindings for other languages exist) that allows developers to write massively parallel applications. MPI is mostly intended for data-parallel applications, where all parallel jobs do the same operations, but on different data, and where the different jobs have to communicate a lot. It is used in high performance computing, where a single application may run on up to several thousands of processors for up to several days.