Indeed, n^n+(n+1)^(n+1) numbers are usually post-processed by Rich Dickerson.

[QUOTE]Indeed, n^n+(n+1)^(n+1) numbers are usually post-processed by Rich Dickerson.[/QUOTE]Noted! Sorry about that!

:smile:

A quick question: Is it worth the trouble to get all of our computers to work on one post processing job at a time?

[QUOTE=Xyzzy;310618]Noted! Sorry about that!

:smile:

A quick question: Is it worth the trouble to get all of our computers to work on one post processing job at a time?[/QUOTE]

Almost certainly not; you'll be network-limited unless you have something a lot faster than gigabit Ethernet between the computers, and the linear algebra distribution is I think based on the assumption that all the CPUs working on the job are identical.

Parallel LA takes a lot more aggregate CPU time than running the job on a single computer; but it's useful because for really big jobs there often isn't a single computer large enough, and four weeks on 1024 cores is much more palatable than four years on eight cores even if it comes out as many more core-years.

[i]PS: would you be interested in sieving more of M929?[/i]

[QUOTE][I]PS: would you be interested in sieving more of M929?[/I][/QUOTE]We are still working 424-430M on one of our slower computers.

We are trying to get excited about sieving but it feels like watching wet paint dry on a humid day.

[QUOTE=Xyzzy;310622]WWe are trying to get excited about sieving but it feels like watching wet paint dry on a humid day.[/QUOTE]For most modern paints, humidity has little to do with paint drying. Oil-based paints, and most are these days, "dry" by oxidation of the oil.

Admittedly, the humidity might affect the comfort-level of the human observer.

[QUOTE]For most modern paints, humidity has little to do with paint drying. Oil-based paints, and most are these days, "dry" by oxidation of the oil.[/QUOTE]That is definitely very interesting! We will research this more. Thanks!

:rakes:

[QUOTE=debrouxl;310600]Indeed, n^n+(n+1)^(n+1) numbers are usually post-processed by Rich Dickerson.[/QUOTE]

I won't be able to get to it in the near future.
Mike, you are welcome to it if you still have an available box.

Xyxxy, it's yours. :smile:

If you wanted to get your hands dirty with OpenMPI, using two reasonably comparable computers connected to the same GigE switch would give you a noticeable speedup. Running two jobs independently, one on each computer, would be more efficient but perhaps not as much fun. :smile:

[QUOTE]Mike, you are welcome to it if you still have an available box.[/QUOTE]Thanks!

:smile:

[QUOTE]If you wanted to get your hands dirty with OpenMPI, using two reasonably comparable computers connected to the same GigE switch would give you a noticeable speedup.[/QUOTE]We have four identical computers in one room, but we do not have GigE cards or a GigE switch. We'll just stick to regular stuff for now.

:mike:

[QUOTE=pinhodecarlos;310065]Please reserve 1583_71_minus1 for me. Thank you.[/QUOTE]

Done.