Hi,
I was just wondering if anyone had any experience of the new Intel Cluster Math Kernel Libs and how they compared with the regular MKL in a clustered environment.
I guess there are pros and cons of each?
Any practical feedback experience appreciated.
Jim
I was just wondering if anyone had any experience of the new Intel Cluster Math Kernel Libs and how they compared with the regular MKL in a clustered environment.
I guess there are pros and cons of each?
Any practical feedback experience appreciated.
Jim
連結已複製
1 回應
Hi Jim,
Intel Cluster MKL is a superset of MKL, If you're using ScaLAPACK, you definitely want Cluster MKL. That's the biggest advantage of Cluster MKL over regular MKL.Cluster MKL contains ScaLAPACK, BLACS, and optimized PBLAS libraries. You can download everything you need to use ScaLAPACK from Netlib but the build is complicated and the performance won't be as good.
Henry
Message Edited by hagabb on 01-21-2005 09:55 AM