Re: [linux-audio-user] Re: slightly OT: supercomputer music

From: Alexandre DENIS <contact@email-addr-hidden>
Date: Wed Mar 08 2006 - 11:04:59 EET

Loki Davison wrote:

> Well, think about the name NUMA. If you don't have a deterministic
> memory access time then realtime is somewhat difficult.

On NUMA, remote memory access is done in constant time (the so-called
"NUMA factor"), so all memory accesses are bounded in time. Not
deterministic, but the high bound is known.

But even on cluster, MPI may be suitable for realtime audio. The typical
latency of MPI over a high performance network is bellow 10 usec which
should be enough for audio.

Received on Wed Mar 8 12:15:05 2006

This archive was generated by hypermail 2.1.8 : Wed Mar 08 2006 - 12:15:06 EET