|
From: | Marcus G. Daniels |
Subject: | Re: [Swarm-Modelling] ABMs on Graphical Processor Units |
Date: | Fri, 28 Dec 2007 11:20:00 -0700 |
User-agent: | Thunderbird 2.0.0.9 (X11/20071115) |
Roshan D'Souza wrote:
FPGAs have limited on chip memory and very low off-chip memory bandwidth as compared to GDDR3/4/5. Also, since the code is directly converted to hardware, I am not sure if the real-estate on the FPGAs will be enough for even simple ABMs.That's a reasonable parameter to optimize, but there's a wide range of memory latencies provided by computer systems today with different costs and benefits. Another way to avoid the memory wall is with concurrency (more threads than processors).
[Prev in Thread] | Current Thread | [Next in Thread] |