Task-based programming with OmpSs and its application

機(jī)譯：使用Omps及其應(yīng)用程序進(jìn)行基于任務(wù)的編程

代理獲取

本網(wǎng)站僅為用戶提供外文OA文獻(xiàn)查詢和代理獲取服務(wù)，本網(wǎng)站沒有原文。下單后我們將采用程序或人工為您竭誠獲取高質(zhì)量的原文，但由于OA文獻(xiàn)來源多樣且變更頻繁，仍可能出現(xiàn)獲取不到、文獻(xiàn)不完整或與標(biāo)題不符等情況，如果獲取不到我們將提供退款服務(wù)。請(qǐng)知悉。

頁面導(dǎo)航

摘要
著錄項(xiàng)
引文網(wǎng)絡(luò)
相似文獻(xiàn)
相關(guān)主題

摘要

OmpSs is a task-based programming model that aims to provide portability and flexibility for sequential codes while the performance is achieved by the dynamic exploitation of the parallelism at task level. OmpSs targets the programming of heterogeneous and multi-core architectures and offers asynchronous parallelism in the execution of the tasks. The main extension of OmpSs, now incorporated in the recent OpenMP 4.0 standard, is the concept of data dependences between tasks.\ud\udTasks in OmpSs are annotated with data directionality clauses that specify the data used by it, and how it will be used (read, write or read&write). This information is used during the execution by the underlying OmpSs runtime to control the synchronization of the different instances of tasks by creating a dependence graph that guarantees the proper order of execution. This mechanism provides a simple way to express the order in which tasks must be executed, without the need of adding explicit synchronization.\ud\udAdditionally, OmpSs syntax offers the flexibility to express that given tasks can be executed on heterogeneous target architectures (i.e., regular processors, GPUs, or FPGAs). The runtime is able to schedule and run these tasks, taking care of the required data transfers and synchronizations. OmpSs is a promising programming model for future exascale systems, with the potential to exploit unprecedented amounts of parallelism while coping with memory latency, network latency and load imbalance.\ud\udThe paper covers the basics of OmpSs and some recent new developments to support a family of embedded DSLs (eDSLs) on top of the compiler and runtime, including an prototype implementation of a Partial Differential Equations DSL.

機(jī)譯：OmpSs是一個(gè)基于任務(wù)的編程模型，旨在為順序代碼提供可移植性和靈活性，同時(shí)通過在任務(wù)級(jí)別動(dòng)態(tài)利用并行性來實(shí)現(xiàn)性能。 OmpSs針對(duì)異構(gòu)和多核體系結(jié)構(gòu)的編程，并在任務(wù)執(zhí)行中提供異步并行性。 OmpSs的主要擴(kuò)展現(xiàn)已包含在最新的OpenMP 4.0標(biāo)準(zhǔn)中，是任務(wù)之間的數(shù)據(jù)依賴關(guān)系的概念。\ ud \ udOmpSs中的任務(wù)用數(shù)據(jù)方向性子句注釋，該子句指定了它使用的數(shù)據(jù)以及如何使用它。（讀，寫或讀和寫）。基礎(chǔ)OmpSs運(yùn)行時(shí)在執(zhí)行過程中使用此信息，通過創(chuàng)建一個(gè)保證正確執(zhí)行順序的依賴圖來控制任務(wù)不同實(shí)例的同步。這種機(jī)制提供了一種表達(dá)任務(wù)執(zhí)行順序的簡(jiǎn)單方法，而無需添加顯式同步。\ ud \ ud此外，OmpSs語法還提供了靈活性，可以表達(dá)可以在異構(gòu)目標(biāo)體系結(jié)構(gòu)上執(zhí)行給定任務(wù)（即，常規(guī)處理器，GPU或FPGA）。運(yùn)行時(shí)能夠安排和運(yùn)行這些任務(wù)，并注意所需的數(shù)據(jù)傳輸和同步。 OmpSs是未來百億億次系統(tǒng)的有前途的編程模型，具有在處理內(nèi)存延遲，網(wǎng)絡(luò)延遲和負(fù)載不平衡的同時(shí)利用前所未有的并行度的潛力。\ ud \ ud本文介紹了OmpS的基礎(chǔ)知識(shí)以及支持AmpSs的一些最新進(jìn)展。編譯器和運(yùn)行時(shí)之上的嵌入式DSL（eDSL）系列，包括偏微分DSL的原型實(shí)現(xiàn)。