ForkTail: a black-box fork-join tail latency prediction model for user-facing datacenter workloads

ForkTail: a black-box fork-join tail latency prediction model for user-facing datacenter workloads
复制标题

ForkTail:用于面向用户的数据中心工作负载的黑盒分叉连接尾部延迟预测模型

DOI:
10.1145/3208040.3208058
复制
发表时间:
2018
期刊:
Proceedings of the 27th International Symposium on High-Performance Parallel and Distributed Computing
影响因子:
--
通讯作者:
Jiang, Hong
Jiang, Hong
中科院分区:
--
文献类型:
--
作者:
Nguyen, Minh;Alesawi, Sami;Li, Ning;Che, Hao;Jiang, Hong

文献摘要

参考文献

被引文献

相似文献

主要的面向用户的数据中心服务的工作流,包括Web搜索和社交网络,都是由各种Fork-Join结构支撑的。由于缺乏对Fork-Join结构的一般性能的理解,今天的网络中心经常诉诸于资源过度供应,在低资源利用率下操作,以满足此类服务的严格的尾部延迟服务水平目标(SLO)。因此,为了实现高资源利用率,同时满足严格的尾部延迟SLO,它是至关重要的是能够准确地预测尾部延迟广泛的Fork-Join结构的实际interests.In本文中,我们提出了ForkTail,一个黑盒Fork-Join尾部延迟预测模型,涵盖了广泛的Fork-Join结构。在ForkTail中,所有Fork节点都被视为黑盒,允许同构和非同构的情况,并且允许请求流中的不同请求派生不同数量的任务,这些任务分叉到不同数量的Fork节点。在中心极限定理的基础上,在重负载下的排队模型,我们能够达到一个高度计算有效的,经验表达式的尾延迟作为任务响应时间的均值和方差的函数。由于该表达式可以应用于任何粒度的请求子流,因此它可以用于合并环境中的服务的尾部延迟预测,其中不同的服务和应用可以共享相同的数据中心集群资源。我们基于基于模型和跟踪驱动的模拟的广泛测试结果,以及云环境中的真实案例研究表明,该表达式可以在80%和90%的负载水平下分别在20%和15%的预测误差内一致地预测尾部延迟。此外,我们的敏感性分析表明,这种错误可以很好地补偿不超过5%和3%的资源预留在这两个负载水平,分别。这一点,加上其极低的计算复杂性,使ForkTail成为面向用户的数据中心应用程序的离线和在线作业调度和资源配置的可行工具。
The workflows of the predominant user-facing datacenter services, including web searching and social networking, are underlaid by various Fork-Join structures. Due to the lack of understanding the performance of Fork-Join structures in general, today's datacenters often resort to resource overprovisioning, operating under low resource utilization, to meet stringent tail-latency service level objectives (SLOs) for such services. Hence, to achieve high resource utilization, while meeting stringent tail-latency SLOs, it is of paramount importance to be able to accurately predict the tail latency for a broad range of Fork-Join structures of practical interests.In this paper, we propose ForkTail, a black-box Fork-Join tail latency prediction model that covers a wide range of Fork-Join structures. In ForkTail, all Fork nodes are treated as black boxes, admitting both homogeneous and inhomogeneous cases, and different requests in the request flow are allowed to spawn different numbers of tasks forked to different numbers of Fork nodes. On the basis of the central limit theorem for queuing models under heavy load, we are able to arrive at a highly computational effective, empirical expression for the tail latency as a function of the means and variances of the task response times. Since this expression can be applied to request sub-flows at any granularities, it can be used for tail-latency prediction for services in a consolidated environment, where different services and applications may share the same datacenter cluster resources. Our extensive testing results based on model-based and trace-driven simulations, as well as a real-world case study in a cloud environment demonstrate that the expression can consistently predict the tail latency within 20% and 15% prediction errors at 80% and 90% load levels, respectively. Moreover, our sensitivity analysis demonstrates that such errors can be well compensated for with no more than 5% and 3% resource overprovisioning at these two load levels, respectively. This, together with its extremely low computational complexity, makes ForkTail a viable tool for both offline and online job scheduling and resource provisioning for user-facing datacenter applications.
并行队列中Fork/Join同步的混合解决方案
DOI: 10.1109/71.946659
发表时间: 2001
期刊: IEEE Trans. Parallel Distributed Syst.
影响因子: --
作者:
R. Chen
通讯作者: R. Chen
具有多个服务器的队列的大流量理论。
DOI: 10.2307/3212906
发表时间: 1974
影响因子: 1
作者:
J. Köllerström
通讯作者: J. Köllerström
DOI: 10.1145/2796314.2745859
发表时间: 2015-06
期刊: ACM SIGMETRICS Performance Evaluation Review
影响因子: --
作者:
Amr Rizk;Felix Poloczek;F. Ciucu
通讯作者: Amr Rizk;Felix Poloczek;F. Ciucu
DOI: 10.1109/71.730531
发表时间: 1998
期刊: IEEE Trans. Parallel Distributed Syst.
影响因子: --
作者:
S. Balsamo;L. Donatiello;N. Dijk
通讯作者: N. Dijk
DOI: --
发表时间: 2015
期刊: 2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing
影响因子: --
作者:
Z. Qiu;Juan F. Pérez
通讯作者: Juan F. Pérez