Using parallelism techniques to improve sequential and multi-core sorting performance

نویسنده

  • Alexandros V. Gerbessiotis
چکیده

We propose new sequential sorting operations by adapting techniques and methods used for designing parallel sorting algorithms. Although the norm is to parallelize a sequential algorithm to improve performance, we adapt a contrarian approach: we employ parallel computing techniques to speed up sequential sorting. Our methods can also work for multi-core sorting with minor adjustments that do not necessarily require full parallelization of the original sequential algorithm. The proposed approach leads to the development of asymptotically efficient deterministic and randomized sorting operations whose practical sequential and multi-core performance, as witnessed by an experimental study, matches or surpasses existing optimized sorting algorithm implementations. We utilize parallel sorting techniques such as deterministic regular sampling and random oversampling. We extend the notion of deterministic regular sampling into deterministic regular oversampling for sequential and multi-core sorting and demonstrate its potential. We then show how these techniques can be used for sequential sorting and also lead to better multi-core sorting algorithm performance as witnessed by the undertaken experimental study.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Parallel computing techniques for sequential and multi-core sorting

We adapt techniques employed in the design of parallel sorting algorithms to propose new sequential and multi-core sorting operations. The proposed approach is used to develop asymptotically efficient deterministic and randomized sorting operations whose practical sequential and multi-core performance, as witnessed by an experimental study, matches or surpasses existing optimized sorting algori...

متن کامل

A high-performance sorting algorithm for multicore single-instruction multiple-data processors

Many sorting algorithms have been studied in the past, but there are only a few algorithms that can effectively exploit both SIMD instructions and thread-level parallelism. In this paper, we propose a new high-performance sorting algorithm, called Aligned-Access sort (AA-sort), for exploiting both the SIMD instructions and thread-level parallelism available on today's multicore processors. Our ...

متن کامل

A parallelism-motivated sequential sorting framework

We employ techniques developed and used in the design of parallel sorting algorithms to propose a new framework for sequential sorting. This framework is then used to design new deterministic and randomized sorting methods whose asymptotic worst-case running time can match the existing lower bound for sorting, yet their practical performance, as witnessed by an experimental study, surpasses exi...

متن کامل

A Comparative Study on Performance Benefits of Multi-core CPUs using OpenMP

Achieving scalable parallelism from general programs was not successful to this point. To extract parallelism from programs has become the key focus of interest on multi-core CPUs. There are many techniques and programming models such as MPI, CUDA and OpenMP adopted in order to exploit more performance. But there is an urge to find the best parallel programming techniques for the benefit of per...

متن کامل

The Performance Potential for Single Application Heterogeneous Systems∗

A consideration of Amdahl’s Law [9] suggests a single-chip multiprocessor with asymmetric cores is a promising way to improve performance [16]. In this paper, we conduct a limit study of the potential benefit of the tighter integration of a fast sequential core designed for instruction level parallelism (e.g., an out-oforder superscalar) and a large number of smaller cores designed for thread-l...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:
  • CoRR

دوره abs/1608.08648  شماره 

صفحات  -

تاریخ انتشار 2016