Introducing Parallel Processing

Traditionally, software packages are written for serial computation:

• To be run on a single computer having a single Central Processing Unit (CPU),

• A problem is broken into a discrete series of instructions,

• Instructions are executed one after another,

• Only one instruction may execute at any moment in time.

Parallel processing enables the simultaneous use of multiple computer resources to solve a computational problem:

• To be run using multiple CPUs,

• A problem is broken into several parts that can be solved concurrently,

• Each part is further broken down to a series of instructions,

• Instructions from each part execute simultaneously on different CPUs.

Thanks to new technologies appearing at highly reduced costs (multi-core computers, clusters, …), Parallel Processing can succeed in performing distributed computing with a reasonable investment. This makes it possible to solve long-running and computationally intensive problems, especially when large independent data sets are involved. This results in a significant speed-up of simulation runs.

Figure 1: Parallel Processing with a distributed network

Figure 2: Parallel Processing with a Multi-Core Processor (here: 2 Dual-Core processors = 4 cores)

Parallel Processing allows an Simcenter Amesim™ software model with multiple cases to be run on different processors. For example, a simulation with 100 cases may be performed by 5 different processors simultaneously with each processor running 20 cases. At the end of the run, results from all of the cases are combined and sent back to the computer that originally submitted the job, the client computer.

Parallel Processing in Simcenter Amesim can be applied to:

• Multi-core processor: one processor composed of 2 or more cores (mostly 2 or 4 cores)

• Symmetric Multi Processor (SMP): one machine composed of several processors sharing the same memory (less than 32 processors)

• Massive Parallel Processor (MPP): machine composed of several processors having their own memory and connected with fast connection

• Cluster: several independent machines connected with simple network connections

Note

You are advised to use dedicated workstations for distributed computing. Using everyday computers such as employee desktops/laptops is not recommended since people have a tendency to turn off desktops/laptops after work.