topic Intel MKL 10.0 Update 4 is now available in Intel® oneAPI Math Kernel Library https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-10-0-Update-4-is-now-available/m-p/903355#M11451 <FONT face="Arial">Intel MKL 10.0 Update 4 is now available. Update 4 includes the following improvements:<BR /> </FONT><UL> <FONT face="Arial"> <LI>Performance Improvements in the BLAS: <UL> <LI> 32-bit improvements</LI> <UL> <LI> 40-50% improvement for (Z,C)GEMM on Quad-Core Intel Xeon processor 5300 series </LI><LI> 10% improvement for all GEMM code on Quad-Core Intel Xeon processor 5400 series </LI></UL> <LI> 64-bit improvements</LI> <UL> <LI>2.5-3% improvement for DGEMM on 1 thread on Quad-Core Intel Xeon processor 5400 series </LI><LI>50% improvement for SGEMM on the forthcoming Intel microarchitecture, code named Nehalem </LI><LI>3% improvement for CGEMM on 1 thread on the forthcoming Intel microarchitecture, code named Nehalem </LI><LI>2-3% improvement for ZGEMM on 1 thread on the forthcoming Intel microarchitecture, code named Nehalem </LI><LI>30% improvement for right-side cases of DTRSM on the forthcoming Intel microarchitecture, code named Nehalem </LI></UL></UL> </LI><LI> Improvements to the direct sparse solver (DSS/PARDISO): <UL> <LI> The performance of out-of-core PARDISO was improved by 35% on average.</LI></UL> </LI><LI> Performance improvement on the Intel microarchitecture, code named Nehalem:</LI> </FONT><UL> <FONT face="Arial"> <LI> 3-17% improvement for the following VML functions: Asin, Asinh, Acos, Acosh, Atan, Atan2, Atanh, Cbrt, CIS, Cos, Cosh, Conj, Div, ErfInv, Exp, Hypot, Inv, InvCbrt, InvSqrt, Ln, Log10, MulByConj, Sin, SinCos, Sinh, Sqrt, Tanh.</LI> <LI> 7-67% improvement for uniform random number generation.</LI> <LI> 3-10% improvement for VSL distribution generators based on Wichmann-Hill, Sobol, and Niederreiter BRNGs (64-bit only).</LI> </FONT></UL><FONT face="Arial"> <LI> On Linux*: When functions in Intel MKL are called from an MPI program they will be run on 1 thread by default (i.e., in the absence of explicit controls). </LI><LI> On Windows*: It is now possible to create DLLs from the static libraries. </LI></FONT></UL> <P><FONT face="Arial"><I>See the release notes for further information</I> </FONT></P><P><FONT face="Arial">To download, please login at the <A href="http://registrationcenter.intel.com">Intel Registration Center</A>. Intel MKL 10.0 Update 4 is available for Linux*, Windows*, and Mac OS* X.<BR /><BR /></FONT></P> Fri, 15 Aug 2008 21:54:20 GMT Todd_R_Intel 2008-08-15T21:54:20Z Intel MKL 10.0 Update 4 is now available https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-10-0-Update-4-is-now-available/m-p/903355#M11451 <FONT face="Arial">Intel MKL 10.0 Update 4 is now available. Update 4 includes the following improvements:<BR /> </FONT><UL> <FONT face="Arial"> <LI>Performance Improvements in the BLAS: <UL> <LI> 32-bit improvements</LI> <UL> <LI> 40-50% improvement for (Z,C)GEMM on Quad-Core Intel Xeon processor 5300 series </LI><LI> 10% improvement for all GEMM code on Quad-Core Intel Xeon processor 5400 series </LI></UL> <LI> 64-bit improvements</LI> <UL> <LI>2.5-3% improvement for DGEMM on 1 thread on Quad-Core Intel Xeon processor 5400 series </LI><LI>50% improvement for SGEMM on the forthcoming Intel microarchitecture, code named Nehalem </LI><LI>3% improvement for CGEMM on 1 thread on the forthcoming Intel microarchitecture, code named Nehalem </LI><LI>2-3% improvement for ZGEMM on 1 thread on the forthcoming Intel microarchitecture, code named Nehalem </LI><LI>30% improvement for right-side cases of DTRSM on the forthcoming Intel microarchitecture, code named Nehalem </LI></UL></UL> </LI><LI> Improvements to the direct sparse solver (DSS/PARDISO): <UL> <LI> The performance of out-of-core PARDISO was improved by 35% on average.</LI></UL> </LI><LI> Performance improvement on the Intel microarchitecture, code named Nehalem:</LI> </FONT><UL> <FONT face="Arial"> <LI> 3-17% improvement for the following VML functions: Asin, Asinh, Acos, Acosh, Atan, Atan2, Atanh, Cbrt, CIS, Cos, Cosh, Conj, Div, ErfInv, Exp, Hypot, Inv, InvCbrt, InvSqrt, Ln, Log10, MulByConj, Sin, SinCos, Sinh, Sqrt, Tanh.</LI> <LI> 7-67% improvement for uniform random number generation.</LI> <LI> 3-10% improvement for VSL distribution generators based on Wichmann-Hill, Sobol, and Niederreiter BRNGs (64-bit only).</LI> </FONT></UL><FONT face="Arial"> <LI> On Linux*: When functions in Intel MKL are called from an MPI program they will be run on 1 thread by default (i.e., in the absence of explicit controls). </LI><LI> On Windows*: It is now possible to create DLLs from the static libraries. </LI></FONT></UL> <P><FONT face="Arial"><I>See the release notes for further information</I> </FONT></P><P><FONT face="Arial">To download, please login at the <A href="http://registrationcenter.intel.com">Intel Registration Center</A>. Intel MKL 10.0 Update 4 is available for Linux*, Windows*, and Mac OS* X.<BR /><BR /></FONT></P> Fri, 15 Aug 2008 21:54:20 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-10-0-Update-4-is-now-available/m-p/903355#M11451 Todd_R_Intel 2008-08-15T21:54:20Z