• Intel® Trace Analyzer and Collector 7.0
• Intel MPI Library 3.1
• Intel MKL Cluster Edition 10.0
• Intel® MPI Benchmarks 3.1

• Call to the function "MPI_Init_thread" to determine the MPI threading environment. The MPI threading environment is reported to the user each time an Intel MPI Benchmark application is executed.
• Call to the function "MPI_Get_version" to report the version of the MPI library implementation that the three benchmark applications are linking to.
• "Alltoallv" benchmark
• Command-line flag "-h[elp]" to display the calling sequence for each benchmark application.
• Removal of outdated makefile templates. Now there are three complete makefiles called Makefile, make_ict, and make_mpich.
• Better user argument checking, clean message and break on most invalid user arguments.

• The parallel Intel Debugger is now capable of debugging an MPI application that uses Intel MPI Library 3.1. For 8.1 releases of the Intel® compilers, please make sure that you have installed version 8.1-23 or higher of the Intel Debugger. For the 9.1 releases of the Intel compilers, please make sure that you have installed version 9.1-23 or higher of the Intel Debugger.
• Chapter 10 of the Intel Cluster Toolkit 3.1 User's Guide discusses use of the Intel Debugger with Intel MPI Library 3.1.

• DAPL intra-node communication mode
• Further optimized collectives
• Intelligent process pinning
• Scalable job start-up protocol

• Additional thread-safe libraries at the MPI_THREAD_SERIALIZED level
• Simplified process management by integration with the leading job schedulers
• Support for Etnus TotalView*, DDT*, and Intel debuggers

• Intel® C++/Fortran Compiler 9.1 for Linux
• GNU Fortran 95 Compiler for Linux, v4.0 and later

• SUSE Linux Enterprise Edition 10.0

• Comparison feature that provides side-by-side comparison of two trace files and/or regions
• Addition of Counter Timeline to help you analyze counter data collected through provided PAPI and OS modules or through manual use of the Intel Trace Collector API
• Improved usability
  • Font customization
  • Undo operation
• Binary instrumentation on IA-32 and Intel® 64 architectures

Further performance optimizations for the Intel® Xeon® processor 5100
Additional optimizations further improve performance over Intel MKL 8.1 on the Intel Xeon processor 5100 sequence

Enhanced Fast Fourier Transform (FFT) performance
Optimizations in Intel MKL 9.0 have boosted FFT performance to the highest levels ever.

Additional Threading Support in Sparse BLAS
The following sparse BLAS triangular solvers were threaded with OpenMP*.

• mkl_dcsrmm - Level 3 triangular solver for the compressed sparse row format
• mkl_dcscmm - Level 3 triangular solver for the compressed sparse column format
• mkl_dcoomm - Level 3 triangular solver for the coordinate format

New VSL functionality

• The library provides fast/accurate mode of distribution generation.
• The library supports generation of quasi-random sequences based on user-defined parameters of quasi-random number generators.

• VML now supports complex single and double precision data for the following functions: Acos, Acosh, Asin, Asinh, Atan, Atanh, Cos, Cosh, Exp, Ln, Log10, Pow, Powx, Sin, Sinh, Sqrt, Tan, Tanh.
• New nearest integer functions for single and double precision: Trunc, Ceil, Floor, Round, NearbyInt, Rint.
• Three more functions for single and double precision arguments were added: ErfInv, Hypot and Modf.
• VML functions are now threaded with OpenMP

Every purchase of an Intel® Software Development Product includes a year of support services, which provides access to Intel® Premier Support and all product releases during that time. Intel Premier Support gives you online access to our expert engineering support staff, technical notes, application notes, and documentation. Install the product, and then register to get support and product update information.