首页> 外文会议>Parallel and distributed processing >Distributing Peridoic Workload Uniformly Across Time to Achieve Better Service Quality
【24h】

Distributing Peridoic Workload Uniformly Across Time to Achieve Better Service Quality

机译:跨时间均匀地分配周期性工作负载,以提高服务质量

获取原文
获取原文并翻译 | 示例

摘要

A multimedia system consists of substantial amount of continuous media workload shceduled periodically at deterministic time points. Phased schedulingis achieved by shifting phase or the lag beteen the invocation times of independent periodic tasks. A proper phase configuraiton distributes workload uniformly over time an reduces task in-terference that may otherwise result in jitter, deadline miss, and logn response time. In our previous work, an algorithm is proposed that identifies optimal phase values for given periodic task set scheduled earliest deadiline first[1]. We now present another phase idnetification algorithm and evaluate its effectiveness. The algorithm in this paper is less accurate but faster than the previous algorithm. In addition, the new approach is applicable to wider set of problems.
机译:多媒体系统由在确定的时间点定期缩减的大量连续媒体工作负载组成。分阶段调度是通过将相位或延迟移到独立定期任务的调用时间之间来实现的。适当的阶段配置会随着时间的推移均匀地分配工作负载,从而减少任务干扰,否则可能会导致抖动,截止期限缺失和登录响应时间。在我们以前的工作中,提出了一种算法,该算法可以为给定的周期性任务集排定最早的最后期限优先识别最佳相位值[1]。现在,我们提出另一种阶段识别算法,并评估其有效性。本文算法的准确性较差,但比以前的算法要快。此外,新方法适用于更广泛的问题。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号