Matthew Dillon wrote a roundup post summarizing all the changes he’s made to DragonFly to improve SMP performance in the last few weeks. He’s removed almost all contention from DragonFly. This means better performance, scaling upward depending on the number of processors.
‘monster’, the system that builds all 20,000 items in dports, can complete the run in 15 hours. Compare this to the 2 weeks it used to take me to build the 12,000 packages in pkgsrc. This is admittedly on different hardware and different packaging systems, but it gives a sense of the scale of the improvement.
I think pkgsrc is still inherently non-parallellizable, so you’re composing apples to orangutans. Still interesting statistic, but probably not shining example of core system improvements in SMP.
s/composing/comparing/
Yeah, I know, but it’s the closest thing I have to a similar example of how far it’s come along.
We really need an always-running benchmark machine, starting in about 2007.
Can we please stop with this “pkgsrc doesn’t scale” myth? My latest full bulk build took 28h40min on moderately fast E3-1270v2. That should be nowhere near as powerful as monster as. So if the bulk build on a faster system takes an order of magnitude longer, it is a setup problem.