Branchless Computing

About the Event:

Have you ever written code as below?
-void f(bool b, long x, long& s) { if (b) s += x; }

Would you like to discover how much performance you left on the table? With a small change, that function could be made 2.5 times faster.

What about another code?
-if (a[i] && b[i]) do_something(); else do_something_else()

Did you know that under some not-so-exotic conditions, this line runs four times slower than it could?

Join us for “Branchless Computing: Why Conditions are Bad and What Can You Do About It?” lecture by Dr. Fedor Pikus, technical fellow and head of the Advanced Projects Team in Siemens Digital Industries Software. During the event, you will:

-learn how modern CPUs handle computations to ensure that the hardware is used efficiently;

-discover how conditional code disrupts the ideal flow of computations and the countermeasures employed by the CPU designers to retain good performance in the presence of such code.

Join us on October 18 at 17:00 in AUA’s room 308e (PAB). Prior registration is required. Please register here.

About the Speaker:

Fedor G Pikus is a technical fellow and head of the Advanced Projects Team in Siemens Digital Industries Software. His earlier positions included a senior software engineer at Google and chief scientist at Mentor Graphics (acquired by Siemens Software). He joined Mentor Graphics in 1998, when he made a switch from academic research in computational physics to the software industry. His responsibilities as a technical fellow include planning the long-term technical direction of Calibre products, directing and training engineers who work on these products, design and architecture of the software, and research in new design and software technologies. Fedor has over 25 patents and over 100 papers and conference presentations on physics, EDA, software design, and C++ language.

Language: English