Vectorization and threading are critical to using such innovative hardware product such as the Intel Xeon Phi processor. Using tools early in the design and development processor that identify where vectorization can be used or improved will lead to increased performance of the overall application. Modern tools can be used to determine what might be blocking compiler vectorization and the potential gain from the work involved.
In this video, LLNL scientists discuss the challenges of debugging programs at scale on the Sequoia supercomputer, which has 1.6 million processors. “Bugs in parallel HPC applications are difficult to debug because errors propagate among compute nodes, programmers must debug thousands of nodes or more, and bugs might manifest only at large scale.”
Wisconsin-based Convergent Science, Inc. (CSI) is well-known for their advanced CFD software. As their development teams focus on speed, precision, and simplifying their workflows, the company says that debugging their code with Allinea DDT is providing a competitive advantage.
In this video from SC13, Mark O’Connor from Allinea Software describes how the company’s new Performance Report Tools provides a scalable and effective way to characterize and understand the performance of HPC application runs, without needing to instrument or recompile.
In this video from SC13, Guillaume Colin de Verdière from CEA in France describes the mission of the organization and how working closely with Allinea enables them to push the limits of HPC technology.
Today Allinea Software announced support for version 5.5 of the NVIDIA CUDA parallel programming toolkit. The new release includes debugging support for C++11, GNU 4.8 compilers, and ARMv7 architectures, which will soon power hybrid platforms with lower energy consumption for HPC.