Re:One trillion edges: graph processing at Facebook scale [pdf]
by MichaelGG on Thursday August 13, 2015 @02:58AM
(#10052642)
Not really on topic, but compare those numbers with the slowness of RAM and cache. A single shared variable that gets updated on one socket then another -- that can cost like 300 cycles, at least on Nehalem.
[ Parent
| on HN ]
Re:One trillion edges: graph processing at Facebook scale [pdf]
by cbsmith on Thursday August 13, 2015 @04:59AM
(#10052914)
In terms of throughput, the RAM is great. It's the latency thing. Which is why intelligent scheduling with embarrassingly parallel problems like this makes such a huge difference.