国产bbaaaaa片,成年美女黄网站色视频免费,成年黄大片,а天堂中文最新一区二区三区,成人精品视频一区二区三区尤物

首頁(yè)> 外文OA文獻(xiàn) >Enabling resource sharing between transactional and batch workloads using dynamic application placement
【2h】

Enabling resource sharing between transactional and batch workloads using dynamic application placement

機(jī)譯:使用動(dòng)態(tài)應(yīng)用程序放置在事務(wù)和批處理工作負(fù)載之間啟用資源共享

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(méi)有原文。下單后我們將采用程序或人工為您竭誠(chéng)獲取高質(zhì)量的原文,但由于OA文獻(xiàn)來(lái)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。

摘要

We present a technique that enables existing middleware to fairly manage mixed workloads: batch jobs and transactional applications. The technique leverages a generic application placement controller, which dynamically allocates compute resources to application instances. The controller works towards a fairness goal while also trying to maximize individual workload performance. We use relative performance functions to drive the application placement controller. Such functions are derived from workload-specific performance models—in the case of transactional workloads, we use queuing theory to build the performance model. For batch workloads, we evaluate a candidate placement by calculating long-term estimates of the completion times that are achievable with that placement according to a scheduling policy. In this paper, we propose a lowest relative performance first scheduling policy as a way to also achieve fair resource allocation among batch jobs. Our technique permits collocation of the workload types on the same physical hardware, and leverages control mechanisms such as suspension and migration to perform online system reconfiguration. In our experiments we demonstrate that our technique maximizes mixed workload performance while providing service differentiation based on high-level performance goals.
機(jī)譯:我們提出了一種使現(xiàn)有中間件能夠公平管理混合工作負(fù)載的技術(shù):批處理作業(yè)和事務(wù)性應(yīng)用程序。該技術(shù)利用了通用的應(yīng)用程序放置控制器,該控制器動(dòng)態(tài)地將計(jì)算資源分配給應(yīng)用程序?qū)嵗???刂破鞒侥繕?biāo)努力,同時(shí)還嘗試最大化單個(gè)工作負(fù)載性能。我們使用相對(duì)性能函數(shù)來(lái)驅(qū)動(dòng)應(yīng)用程序放置控制器。此類功能源自特定于工作負(fù)載的性能模型-在事務(wù)性工作負(fù)載的情況下,我們使用排隊(duì)論來(lái)構(gòu)建性能模型。對(duì)于批處理工作負(fù)載,我們通過(guò)根據(jù)調(diào)度策略計(jì)算該放置可以實(shí)現(xiàn)的完成時(shí)間的長(zhǎng)期估計(jì),來(lái)評(píng)估候選放置。在本文中,我們提出了一種相對(duì)性能最低的優(yōu)先調(diào)度策略,以同時(shí)實(shí)現(xiàn)批處理作業(yè)之間的公平資源分配。我們的技術(shù)允許在相同的物理硬件上并置工作負(fù)載類型,并利用諸如暫停和遷移之類的控制機(jī)制來(lái)執(zhí)行在線系統(tǒng)重新配置。在我們的實(shí)驗(yàn)中,我們證明了我們的技術(shù)可以最大程度地提高混合工作負(fù)載的性能,同時(shí)根據(jù)高級(jí)性能目標(biāo)提供服務(wù)差異化。

著錄項(xiàng)

相似文獻(xiàn)

  • 外文文獻(xiàn)
  • 中文文獻(xiàn)
  • 專利
代理獲取

客服郵箱:kefu@zhangqiaokeyan.com

京公網(wǎng)安備:11010802029741號(hào) ICP備案號(hào):京ICP備15016152號(hào)-6 六維聯(lián)合信息科技 (北京) 有限公司?版權(quán)所有
  • 客服微信

  • 服務(wù)號(hào)