HPC.NRW Advances Research and Education With State-Wide TotalView Licensing

The North Rhine-Westphalian Competence Network for High-Performance Computing (HPC.NRW) offers a broadly based contact and advice center for HPC users in the German state of North Rhine-Westphalia. To ensure the basic supply for locations without level 3 centers and the universities of applied sciences, the level 2 centers offer uniform, structured access for HPC users of all universities in North Rhine-Westphalia.

As part of the competence network, HPC.NRW aims to create a network of thematic clusters for low threshold training, consulting, and coaching offers. The goal is to use the high-performance computing and storage systems effectively and efficiently and also to support newly graduated and doctoral researchers. 

TotalView Helps HPC.NRW…

icon-process

Implement new ideas and findings faster

icon-code-scan

Debug highly-complex MPI programs

Detect potential problems in scientific programs

“The overall user experience of TotalView is well-suited for our requirements and we would recommend it for others.”

 

HPC Cluster System Requires Creative Solution for Complex Code

North Rhine-Westphalia has a differentiated ecosystem of HPC systems, with levels ranging from zero to three. Level three high-performance computers are at almost all university locations in the state. The mission is to establish a network of HPC throughout these universities, with structure, support, and training.

Previously HPC.NRW had used other debuggers, which were not easy to use on MPI programs. Often the choices were printf or similar statements in the code. With a robust HPC cluster system, HPC.NRW needed an efficient MPI debugger that could be used across the 12 data centers in North Rhine-Westphalia.

Additionally, the type of research conducted required highly-customized code that needed to be implemented quickly to advance science and technology. A more efficient debugging tool was needed to develop faster.
 

Site-Wide Licenses Support for 12 HPC.NRW Data Centers

TotalView’s site-wide license model provides HPC.NRW the flexibility to distribute licenses across their clusters and enable use of the debugger throughout their computing HPC systems. Now, with TotalView installed on systems such as the Noctua system at Paderborn Center for Parallel Computing, TotalView is available to all users of this system. 

In addition to the flexibility of a site-wide license, TotalView’s license model provides flexibility to efficiently provide TotalView licenses to users. With TotalView’s license, users can debug from small to large HPC jobs, consuming what they need for a license and returning them for reuse when done. The licensing gives a balanced usage of TotalView in HPC.NRW systems like the Noctua cluster system.

A site-wide, multi-site licensing solution offers a few other benefits over traditional team licensing. Unlike the team license, the shared license is installed and configured on the HPC cluster, reducing installation time. Licensing and maintenance costs are also lower compared to the team model. If HPC.NRW wants to expand their network, the licensing model allows the user base to be increased. The license also allows HPC.NRW to quickly get access to all new releases and software improvements.

Leveraging TotalView’s Training program to bring users up to speed quickly on HPC debugging, HPC.NRW have started to integrate TotalView into their HPC education and training. For long-term planning, HPC.NRW will leverage the TotalView training material and labs to make all researchers who develop scientific code on HPC.NRW systems familiar with TotalView. This will drastically improve the ability to find problems in their custom, highly-parallel and advanced codes. 
 

Quickly Solve Your Debugging Challenges

TotalView helps organizations like HPC.NRW easily scale and deploy a solution that allows developers to intuitively diagnose and understand complex code. See for yourself how TotalView will help you do the same.

START FREE TRIAL