Re:A 32nm 1000-Processor Array
by pjc50 on Monday June 20, 2016 @06:32AM
(#11937160)
Effectively you have to design for message-passing between microservices as a replacement for function calls. Not only do you have to step back from cache consistency but you have to step back from uniform memory access altogether and head towards a streaming architecture. It's very alien, but that's the only way to really use a system which has such a high ratio of processing power to external memory bandwidth/latency.
[ Parent
| on HN ]
There are no comments on this comment yet.