Language-based vectorization and parallelization using intrinsics, OpenMP, TBB and Cilk Plus

Abstract

The aim of this paper is to evaluate OpenMP, TBB and Cilk Plus as basic language-based tools for simple and efficient parallelization of recursively defined computational problems and other problems that need both task and data parallelization techniques. We show how to use these models of parallel programming to transform a source code of Adaptive Simpson's Integration to programs that can utilize multiple cores of modern processors. Using the example of Belman-Ford algorithm for solving single-source shortest path problems, we advise how to improve performance of data parallel algorithms by tuning data structures for better utilization of vector extensions of modern processors. Manual vectorization techniques based on Cilk array notation and intrinsics are presented. We also show how to simplify such optimization using Intel SIMD Data Layout Template containers.

Enhancing OpenMP Tasking Model: Performance and Portability

“…Conversely, all Cilk++, TBB and CUDA graphs require some refactoring from the code for different reasons: (a) Cilk++ does not provide data-flow dependencies, but full synchronizations instead; (b) TBB decouples the description of the graph from its execution, and requires specific functions for starting the graph and joining results; and (c) CUDA graphs provide a lowlevel API that forces programmers to manage data copies and point-to-point synchronizations. The performance comparison between these models is out of the scope of this paper, but several works have already tackled this topic showing performance results for OpenMP competitive to the other parallel models [12,17].…”

A survey on evaluating and optimizing performance of Intel Xeon Phi

“…Also, some works use OpenMP task pragmas for parallelization. 33,52,63 TA B L E 5 Parallelization language/library used on Phi MPI 4,9,12,16,22,24,25,28,35,36,43,44,50,54,59,60,63,66,76,82,86 Others Pthreads, 11,23,76,84,91,95 Intel TBB, 8,48 Cilk Plus, 48 OpenCL 49,100 Chatzikonstantis et al 28 study inferior-olivary nucleus (InfOli) simulation which is used in brain modeling. They accelerate the simulation using (i) MPI, (ii) OpenMP, and (iii) hybrid MPI+OpenMP.…”

“…,[12][13][14][16][17][18]20,21,24,[27][28][29][30][33][34][35][36][37]39,[42][43][44][45][46][47][48]50,52,55,57,59,63,66,84,86,90,92,93,95,99 IntelMKL 2,17,19,31,32,40,93,99 …”

High Performance Portable Solver for Tridiagonal Toeplitz Systems of Linear Equations

“…Our OpenMP (version 3.1) implementation of this method achieved very good speedup on Intel Xeon CPUs (up to 5.06) and Intel Xeon Phi (up to 29.45). While this approach can be further improved using more sophisticated vectorization techniques such as the use of intrinsics [ 1 , 8 , 11 ], it will result in a loss of portability between different architectures. OpenACC is a standard for accelerated computing [ 2 , 6 ].…”