Global cost/quality management across multiple applications
Global cost/quality management across multiple applications
复制标题
跨多个应用程序的全球成本/质量管理
DOI：
10.1145/3368089.3409721
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Kremer, Ulrich
中科院分区：
文献类型：
--
作者：
Liu, Liu;Isaacman, Sibren;Kremer, Ulrich
近似是一种优化应用程序结果质量与其资源使用之间的平衡的技术。已经针对单个应用程序场景研究了以质量换取性能的问题，但没有针对多个近似应用程序可能在同一台计算机上同时运行、通过共享计算机资源相互干扰的环境进行了研究。将现有的单一应用程序技术应用于这种多编程环境可能会导致配置空间大小爆炸，或导致整体应用程序质量较差。我们的新 RAPID-M 系统是第一个跨应用程序配置管理框架。它通过将各个应用程序的配置聚集到本地“相似性桶”中来减少问题的大小。全局的跨应用配置选择就是基于这些本地的桶空间。 RAPID-M 动态地将存储桶分配给应用程序，以便在尊重各个应用程序成本预算的同时最大化整体质量。分配存储桶后，可以在本地执行存储桶内的重新配置，而对全局选择的影响最小。使用六个可配置应用程序的实验结果表明，即使是复杂应用程序的大型配置空间也可以聚集到少量的存储桶中，从而使我们的六个应用程序的搜索空间大小减少多达 9 个数量级。 RAPID-M构建的性能成本模型平均预测误差≤3%。对于我们的应用程序执行跟踪，RAPID-M 动态选择配置，与其他可能的方法相比，预算违规率降低了 33.9%，平均预算超出率为 6.6%。 RAPID-M 成功地将执行次数提高了 22.75%，这意味着在高系统负载下全局输出质量提高了 1.52 倍。 RAPID-Mi 的总开销在应用程序执行时间的 ≤1% 内。
Approximation is a technique that optimizes the balance between application outcome quality and its resource usage. Trading quality for performance has been investigated for single application scenarios, but not for environments where multiple approximate applications may run concurrently on the same machine, interfering with each other by sharing machine resources. Applying existing, single application techniques to this multi-programming environment may lead to configuration space size explosion, or result in poor overall application quality outcomes.Our new RAPID-M system is the first cross-application con-figuration management framework. It reduces the problem size by clustering configurations of individual applications into local"similarity buckets". The global cross-applications configuration selection is based on these local bucket spaces. RAPID-M dynamically assigns buckets to applications such that overall quality is maximized while respecting individual application cost budgets.Once assigned a bucket, reconfigurations within buckets may be performed locally with minimal impact on global selections. Experimental results using six configurable applications show that even large configuration spaces of complex applications can be clustered into a small number of buckets, resulting in search space size reductions of up to 9 orders of magnitude for our six applications. RAPID-M constructs performance cost models with an average prediction error of ≤3%. For our application execution traces, RAPID-M dynamically selects configurations that lower the budget violation rate by 33.9% with an average budget exceeding rate of 6.6% as compared to other possible approaches. RAPID-M successfully finishes 22.75% more executions which translates to a 1.52X global output quality increase under high system loads. Theo verhead ofRAPID-Mis within≤1% of application execution times.
DOI：
10.1287/orms.2013.03.02
发表时间：
2019-08
期刊：
Volume 40, Number 3, June 2013
影响因子：
--
作者：
通讯作者：
--
影响因子：
19.5
作者：
Rubner, Y;Tomasi, C;Guibas, LJ
通讯作者：
Guibas, LJ
DOI：
10.1145/1452044.1452048
发表时间：
2008-12-01
影响因子：
1.3
作者：
Park, Sungwoo;Pfenning, Frank;Thrun, Sebastian
通讯作者：
Thrun, Sebastian