| 1 |
On Wed, 2012-09-12 at 08:58 -0400, Ian Stakenvicius wrote: |
| 2 |
|
| 3 |
> So essentially what you're saying here is that it might be worthwhile |
| 4 |
> to look into parallelism as a whole and possibly come up with a |
| 5 |
> solution that combines 'emerge --jobs' and build-system parallelism |
| 6 |
> together to maximum benefit? |
| 7 |
|
| 8 |
Forget about jobs and load average, and just keep starting jobs all |
| 9 |
around until there is only 20% (or whatever tuneable amount) free memory |
| 10 |
left. As far as I can tell this is always the real bottleneck in the |
| 11 |
end. Once you hit swap overall throughput has to go down quite a bit. |
| 12 |
|
| 13 |
Hans |
Archives