Robot projects are often evolutionary dead ends, with the software and hardware they produce disappearing without trace afterwards. Common causes include dependencies on uncommon ...
Paul M. Fitzpatrick, Giorgio Metta, Lorenzo Natale
Performance improvement solely through transistor scaling is becoming more and more difficult, thus it is increasingly common to see domain specific accelerators used in conjunc...
Cache memories were invented to decouple fast processors from slow memories. However, this decoupling is only partial, and many researchers have attempted to improve cache use by p...
In earlier work, we showed that the one-sided communication model found in PGAS languages (such as UPC) offers significant advantages in communication efficiency by decoupling d...
Rajesh Nishtala, Paul Hargrove, Dan Bonachea, Kath...
The Merrimac supercomputer uses stream processors and a highradix network to achieve high performance at low cost and low power. The stream architecture matches the capabilities o...
Mattan Erez, Jung Ho Ahn, Ankit Garg, William J. D...