ParaView on Multiple Processors

ParaView is, as the name implies, Parallel Aware and designed from the ground up for multiprocessor usage. It offers alot of flexibility in how this can be accomplished, and alot of options to make it work in almost any situation.


Use Multi-Core
The "Use Multi-Core" option in ParaView 3.12 and up

On some systems, you can simply launch ParaView normally and select the "Use Multi-Core" option in the Settings menu. Then, the next time you launch ParaView it will automatically distribute the server processes amongst all available cores using MPI. This is the simplest way, but unfortunately offers no flexibility in the number of cores or memory used (eg, a 16-core machine will always launch a 16-way job, utilizing all cores).

Otherwise, you will need to refer to the means below to establish a client-server connection to "LocalHost".


There are other more advanced configurations that can be used for Remote Visualization or separation of services (Data, Rendering, and User Interface). Each mode separates the three main components (User Interface, Computation, and Rendering) in different ways.

Configuration User Interface Computation Rendering
Basic Multiprocessor On the Client
ParaView Client-Server Mode On the Client On the Remote Server
ParaView Client-Distributed Server Mode On the Client On the Remote Cluster
ParaView Tiled-Display Mode On the Client (Not on the Tiled Display) On the Remote Cluster
ParaView in a CAVE On the Client (Not in the Cave) On the displays/nodes of the CAVE
ParaView Distributed Data-Distributed Render Mode On the Client On the Data Cluster On the Render Cluster

Back to ParaView