allinea - Now part of ARM

Performance Webinars Menu

Previous webinars available to view....

Optimizing Communication in High Performance Computing: The One Time You Want Everyone Talking Over Each Other!

In modern High Performance Computing (HPC), even with systems that have proprietary, fast interconnects, the overhead of communication brings an untrivial impact on software efficiency and the speed at which users can expect to see results.  By understanding more about communication, users can determine and address its impact on performance overall.

Optmize communication 180x100

View slides 

Discover and Fix I/O Performance Killers

Using industry leading tools from Allinea Software, this webinar reveals how IO performance issues are identified and investigated and how bad IO behavior can be significantly improved through deeper diagnostics. The session provides an overview of some of the complex issues around IO performance which can be resolved to great effect through end to end monitoring and optimization. 

Performance optimization   

 

Workload Imbalance: Tackling its Impact on Efficiency and Scale

Workload balance has become a central issue in high performance computing, as spotting the tell-tale signs of imbalance in your resources may help you gain more for your effort, accelerate run times dramatically and overcome barriers to scale. The session examines some common causes of imbalance and demonstrate valuable pointers for identification and further investigation. We will also share some examples of optimization made possible through analyzing workload related insight provided by cross-platform performance tools.

 workload image WEBINAR

 

Improving memory access patterns for optimization

What impact is memory bound code having on your HPC or scientific applications?  Without the right checks and ongoing tuning, your software run-times can suffer from unidentified bottlenecks that slow down applications. For example, where sections of code are memory bound, the CPU is starved of resource to complete its necessary compute efficiently. Paying attention to memory performance pays particular dividends in parallel applications, where there is contention for a shared resource. Optimizing for memory accesses can help reduce inefficiencies and reduce run-times.  This session demonstrates how high performance diagnostic, profiling and debugging tools can be used to enable more effective memory performance analysis, helping you define the most appropriate strategies for optimizing memory and overall performance.

 

Coming soon...

Core Utilization - How to Measure and Minimize Performance Impact

We're still finalizing the details of this event. Get on the mailing list for updates via our newsletter.

Vectorization - When and How to Use it for Optimum Performance

We're still finalizing the details of this event. Get on the mailing list for updates via our newsletter.

 

You can also take a software trial or email us at sales@allinea.com