Altera FPGAs accelerate algorithms in Texas academic compute resource

November 16, 2015 // By Graham Prophet
A new server cluster uses Altera FPGAs to help researchers and academia run complex algorithmic-based research.

Programmable logic technology from Altera is now in place inside an advanced server cluster at the Texas Advanced Computing Center (TACC) at The University Texas at Austin, which seeks to help researchers and academia to run complex algorithmic-based research. Microsoft Research shared more about the server cluster becoming available for research use at TACC, in a blog post today:

Altera field programmable gate arrays (FPGAs) are used as accelerators on the Microsoft Project Catapult board, and turn commodity servers into high-performance, power-efficient machines. Researchers are being invited to use TACC resources to run their software programs and algorithms, associated with, but not limited to, deep learning, convolutional neural networks (CNN), facial and object recognition, online search, and even genomic research. TACC also offers free use of boards to students, professors and the community in its ongoing effort to provide access to world-class high performance computing systems.

“TACC?s 14-year history is impressive, and we are pleased Altera technology will be the latest asset in the centre's compute arsenal,” said Erhaan Shaikh, vice president, Infrastructure Division, Altera Business Units. “By making these high-performance servers with our FPGAs inside available to researchers, we anticipate that much progress will be made in the fields of science, math, medicine and all areas requiring these powerful computations.”

The Texas Advanced Computing Center (TACC) designs and operates some of the world's most powerful computing resources. The centre's mission is to enable discoveries that advance science and society through the application of advanced computing technologies. To learn more about TACC, visit

Altera FPGAs are enabling higher speed data processing by providing customised high-bandwidth, low-latency connections to network and storage systems, as well as implementing compression, data filtering, and algorithmic acceleration.