What goes around comes around. Many of the techniques being announced for today's SMP systems is the stuff that was being done on big MPP machines by NEC / Intel / Fujitsu / Cray / Thinking Machines in the 1990s.
It seems to me that the techniques that were in silicon then are being brought into kernels today. Crays ran a modified BSD4.3. All we need now is someone to say Hypercube.
Today's HPC is pretty messy. Driven by MPI and Fortran they have even pushed the Linux kernel out of contention by using OS-Bypass because the kernel is too slow getting bytes on the wire. In Plan9 we are experimenting with Syscall currying to reduce latency
I've drifted a bit because I wanted to use this quote :
"You want to make your way in the CS field? Simple. Calculate rough time of amnesia (hell, 10 years is plenty, probably 10 months is plenty), go to the dusty archives, dig out something fun, and go for it. It's worked for many people, and it can work for you." - Ron Minnich
oh if you liked that pdf here's another one about Plan9 on the Blue Gene
see http://www.amazon.co.uk/Advanced-Computer-Architecture-Proce...
It seems to me that the techniques that were in silicon then are being brought into kernels today. Crays ran a modified BSD4.3. All we need now is someone to say Hypercube.
Today's HPC is pretty messy. Driven by MPI and Fortran they have even pushed the Linux kernel out of contention by using OS-Bypass because the kernel is too slow getting bytes on the wire. In Plan9 we are experimenting with Syscall currying to reduce latency
http://iwp9.quanstro.net/papers/usecsys.pdf
I've drifted a bit because I wanted to use this quote :
"You want to make your way in the CS field? Simple. Calculate rough time of amnesia (hell, 10 years is plenty, probably 10 months is plenty), go to the dusty archives, dig out something fun, and go for it. It's worked for many people, and it can work for you." - Ron Minnich
oh if you liked that pdf here's another one about Plan9 on the Blue Gene
http://www.iwp9.org/papers/bluegene-20.pdf