Virginians know about spellcheck, but it may not have permeated the rarefied bubble of UVa. Mr Jefferson probably spelled "research" that way back in the 19th century when planning his Academical Village. Or perhaps this was written by a thoughtless mere first-year.
Oh, I'm glad! I have several friends who are UVA grads (graduate and undergrad levels), and so I say it with love. Mockery is the sincerest form of flattery.
UVA is doing some great focused work in certain CS fields. Their student team recently won the National Collegiate Cyber Defense Competition in their first year competing [0].
It's interesting that most of the 10 finalist schools aren't the traditional CS powerhouses. If you wanted to study cyberdefense, what schools are best?
Here's a description of the "Automata Processor", a phrase I hadn't heard before [1]:
> Micron’s new Automata Processor (AP) architecture exploits the very high and natural level of parallelism found in DRAM technologies to achieve native-hardware implementation of nondeterministic
finite automata (NFAs). The use of DRAM technology to implement the NFA states provides high capacity and therefore provide extraordinary parallelism for pattern recognition. In this paper, we give an overview of AP's architecture, programming and applications.
It's better not to post home pages of things like institutions or magazines, but rather to submit the most interesting specific link that is available. We've found that discussion is only substantive when there's material to sink teeth into.
It's been around for at least 5 years with no real market traction. Processors in memory fix the Von Neuman bottleneck. Great for REGEX, none have floating point yet to make them useful for scientific code.
15 comments
[ 4.3 ms ] story [ 17.3 ms ] thread[0] http://raytheon.mediaroom.com/2018-04-15-University-of-Virgi...
> Micron’s new Automata Processor (AP) architecture exploits the very high and natural level of parallelism found in DRAM technologies to achieve native-hardware implementation of nondeterministic finite automata (NFAs). The use of DRAM technology to implement the NFA states provides high capacity and therefore provide extraordinary parallelism for pattern recognition. In this paper, we give an overview of AP's architecture, programming and applications.
[1] https://www.cs.virginia.edu/~skadron/Papers/wang_APoverview_...
Edit: since https://news.ycombinator.com/item?id=16987095 includes such a link, let's try changing the submission to that. Original url was https://engineering.virginia.edu/cap-research-0.