Advanced optimizations boost performance

The compilers and libraries in Intel® Composer XE offer advanced vectorization support, including support for Intel® Advanced Vector Extensions (Intel® AVX). The C/C++ optimizing compiler now includes Intel® Parallel Building Blocks (Intel® PBB) library, which expands the types of problems that can be solved with increased scale and reliability, while also simplifying the process.

Advanced Performance Features

• Intel® Parallel Building Blocks (Intel® PBB) and OpenMP* 3.0 simplifies implementing efficient parallel code.

• Optimized libraries provide math functions in Intel® Math Kernel Library and functions in many other domains like compression, crypto, and image processing in Intel® Integrated Performance Primitives provide automatic parallelization and performance.

• High-Performance Parallel Optimizer (HPO) offers an improved ability to analyze, optimize, and parallelize more loop nests.

• Interprocedural Optimization (IPO) dramatically improves performance of small- or medium-sized functions that are used frequently, especially programs that contain calls within loops.

• Profile-Guided Optimization (PGO) improves application performance by reducing instruction-cache thrashing, reorganizing code layout, shrinking code size, and reducing branch mispredictions.

  ?

Simplified Parallelism

Intel® Parallel Building Blocks is a set of comprehensive parallel development models that support multiple approaches to parallelism. Mix and match the models that suit your unique parallel implementation needs.

• Intel® Cilk™ Plus is the simplest way to add parallelism. It's a Intel® C/C++ compiler extension with only 3 keywords that simplifies implementing simple loop and task parallel applications. It offers superior functionality by combining vectorization features with high-level loop-type data parallelism and tasking.




• Intel® Threading Building Blocks is a C++ template library solution to implement loop and task parallelism. It includes scalable memory allocation, load-balancing, work-stealing task scheduling, a thread-safe pipeline and concurrent containers, high-level parallel algorithms, and numerous synchronization primitives.




• Intel® Array Building Blocks (In beta now) is for software developers who write compute-intensive, data parallel algorithms. It produces scalable, portable, and deterministic parallel implementations from a single high-level, maintainable, and application-oriented specification of the desired computation.

New Fortran Features

Intel® Fortran Compiler XE 12.0 also includes several advances, among them more complete support for the Fortran 2003 standard and some support for Fortran 2008 standards, including Co-Array Fortran, vector optimizations with Intel AVX, and help in auto-parallelization, for improved performance and parallelism on the latest multicore processors.

The Fortran compiler is available in Intel® Composer XE