Concertio Blog

Continuous Optimization of Compiler Flags

by Andrey Gelman | May 8, 2020 | Blog

One of the ways to improve code performance is to compile with the right flags. After all, compiler engineers have invested years and years into making compilers better, and in the process, they’ve added many flags that we can use. However, in order to achieve optimal... Read More

Turning General-Purpose CPUs into High-Performant Special-Purpose CPUs

by Uri Weiser | May 6, 2020 | Blog

During the last two decades, the amount of computing applications with different requirement characteristics has grown dramatically. Nevertheless, in most of the computing cases, we are still using the “same” General-Purpose computing engines (e.g. general-purpose... Read More

Using Averages, Percentiles, and Modes When Tuning Real-World Systems

by Omer Yehezkely | Apr 29, 2020 | Blog

Life isn’t always perfect… and neither is any system which is based on sampling of real-world systems. In many cases, the samples do not render a perfect normal distribution, but rather a skewed one. If we are to tune a system, whether manually or using an automatic... Read More

February 2020 Feature Updates

by Tomer Paz | Feb 13, 2020 | Blog

We have started the year off busy improving our products, releasing the following new features, and expanding to new use-cases! We hope you like these enhancements:    Knob refinement A new post-processing step that refines the recommended settings by filtering out... Read More

Automatically Accelerating TensorFlow Workloads on Intel CPUs

by Tomer Morad | Oct 23, 2018 | Blog

See how automated optimization increases ResNet50 performance by 1.77x One of the first thoughts that comes to mind about deep learning and AI is the hope that someday we might be able to develop cognition in computers, that our creations would be able to think on... Read More