In-Place Associative Computing

 
Current Solution
GSI Technology's New APU Solution
 

GSI Technology is developing a new, patented Associative Processing Unit (APU) that changes the concept of computing from serial data processing—where data is moved back and forth between the processor and memory—to massive parallel data processing, compute, and search in-place directly in the memory array.

 

This in-place associative computing technology removes the bottleneck at the I/O between the processor and memory. Data is accessed by content and processed directly in place in the memory array without having to cross the I/O. The result is an orders of magnitude performance-over-power ratio improvement compared to conventional methods that use CPU and GPGPU (General Purpose GPU) along with DRAM.

 

Target applications include memory-bound sparse matrix-vector multiplication,
convolutional neural networks, image detection, signal detection, speech recognition,
recommender systems for e-commerce, and data mining tasks such as prediction, classification, and clustering.

For more information on this exciting new technology, please contact Pat Lasserre at patl@gsitechnology.com .
 

Presentations
Stanford EE Computer Systems Colloquium
In-Place Deep Learning
Interviews
Dr. Avidan Akerib Interview at
Global Predictive Analytics Conference
In-Place Associative Computing Interview
with RE•WORK
Application Notes
In-Place Computing:  Scaling to 1M Similarity Searches per Second
Articles
Review of 2017 Energy Consequences of Information Conference