Any tool that lowers the barrier to taking advantage of these [multi-core] processors is going to be critical. It's clear that this focus on memory bandwidth is going to go far beyond HPC. To me that's huge.
Josh Simons
Distinguished engineer at Sun Microsystems
Acumem is a leading provider of intelligent software technology
which analyzes and optimizes the computing performance in single- and
multi core environments.
Our goal is to contribute to our customers success in maximizing
the benefits of multicore technology, achieving the full potential of
their systems.
... HP's Multicore Toolkit used in
conjunction with Acumem SlowSpotter, offers customers a complete
multicore hardware and software solution that maximizes application
performance ...
Ed
Turkel
Manager
of HPC Product Marketing for HP's Scalable Computing and
Infrastructure Organization
At Pantor we use SlowSpotter to quickly gain a thorough understanding
of code performance qualities. We also use SlowSpotter on a regular
basis to alert us of harmful code and to quickly catch performance
regressions. It improves performance work productivity by offering
unique insight into these problems.
Our performance experts ... use [Acumem SlowSpotter] to find and fix multicore
performance bottlenecks in a wide range of data-intensive applications.
Bjorn Andersson
Director, HPC and Integrated Systems, SUN Microsystems
I first tried Acumem SlowSpotter on some highly optimized code. ...
No other tool would
have managed to pinpoint [the remaining] problem in such an obvious and intuitive
manner!
Jeffrey M. Birnbaum
Chief Architect, Merrill Lynch
Efficient multicore applications are fundamental to green computing and high performance. This requires tuning for deep memory hierarchies and thread interaction in addition to parallelization. Actually, sometimes a well-tuned sequential program can outperform a parallelized program. This process requires expert knowledge and many weeks of data gathering and human analysis using conventional performance tools. Acumems tools automatically analyze the efficiency of your applications and suggest necessary fixes to unleash multicore performance. What used to take experts weeks is now done in minutes at the click of a button. These language-agnostic tools do not require new languages or proprietary extensions. The resulting code increases performance across a wide range of architectures. In this session we will demonstrate the analysis and fix of popular open-source applications in minutes. We show repeated examples where parallelization alone completely misses the point. Performance improvements or energy savings of up to 30 times are demonstrated.