EAGER: Using Machine Learning to Increase the Operational Efficiency of Large Distributed Systems

EAGER：利用机器学习提高大型分布式系统的运营效率

• 批准号: 1649087
• 负责人: Evgenia Smirni
• 金额: $ 30万
• 依托单位: College of William and Mary
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-09-01 至 2019-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1649087&HistoricalAwards=false
• 关键词: EAGER; Using; Machine; Learning; Increase

Large, distributed systems are nowadays ubiquitous and part of sustainable IT solutions to a broad range of customers and applications. Data centers in the private or public cloud and high performance computing systems are two examples of complex, highly distributed systems: the former are used by almost everyone on a daily basis, the latter are used by computational scientists for advancing science and engineering. High availability and reliability of these complex systems are important for the quality of user experience. Efficient management of such systems contributes to their availability and reliability, and relies on a priori knowledge of the timing of the collective demands of users and a priori knowledge of certain performance measures (e.g., usage, temperature, power) of various systems components.This project aims to provide a systematic methodology to improve the operational efficiency of complex, distributed systems by developing neural networks that can efficiently and accurately predict the incoming workload within fine and coarse time scales. Such workload prediction can dramatically improve the operational efficiency of data centers and high performance systems by driving proactive management strategies that specifically aim to enhance reliability. For datacenters, the focus is on actively reducing performance tickets that are automatically triggered by pro-actively managing virtual machine resizing and migration. For high performance computing systems the focus is on predicting hardware faults to autonomically improve the scheduler's efficiency, direct cooling, and improve performance and memory bandwidth.

如今，大型分布式系统是无处不在的，也是可持续IT解决方案的一部分，可用于广泛的客户和应用程序。 私有或公共云和高性能计算系统中的数据中心是复杂，高度分布的系统的两个示例：前者几乎每个人每天都使用，后者被计算科学家用于推进科学和工程。 这些复杂系统的高可用性和可靠性对于用户体验的质量很重要。 此类系统的有效管理有助于其可用性和可靠性，并依靠对用户集体需求的时间的先验知识以及对某些性能措施的先验知识（例如，各种系统组件的某些绩效措施（例如，用法，温度，温度，功率）的知识秤。这样的工作负载预测可以通过推动专门旨在提高可靠性的积极管理策略来大大提高数据中心和高性能系统的运行效率。 对于数据中心，重点是积极降低通过积极管理虚拟机调整和迁移而自动触发的性能门票。对于高性能计算系统，重点是预测硬件故障以自主提高调度程序的效率，直接冷却，并提高性能和内存带宽。

项目成果

Fault Site Pruning for Practical Reliability Analysis of GPGPU Applications

• DOI: 10.1109/micro.2018.00066
• 发表时间: 2018-10
• 期刊: 2018 51st Annual IEEE/ACM International Symposium on Microarchitecture (MICRO)
• 作者: Bin Nie;Lishan Yang;Adwait Jog;E. Smirni

Efficient Deep Neural Network Serving: Fast and Furious

• DOI: 10.1109/tnsm.2018.2808352
• 发表时间: 2018-02
• 期刊: IEEE Transactions on Network and Service Management
• 影响因子: 5.3
• 作者: Feng Yan;Yuxiong He;Olatunji Ruwase;E. Smirni

How to Supercharge the Amazon T2: Observations and Suggestions

• DOI: 10.1109/cloud.2017.43
• 发表时间: 2017-06
• 期刊: 2017 IEEE 10th International Conference on Cloud Computing (CLOUD)
• 作者: Feng Yan;Lihua Ren;Daniel J. Dubois;G. Casale;Jiawei Wen;E. Smirni

CEDULE: A Scheduling Framework for Burstable Performance in Cloud Computing

CEDULE：云计算中突发性能的调度框架

• DOI: 10.1109/icac.2018.00024
• 发表时间: 2018
• 期刊: 2018 IEEE International Conference on Autonomic Computing (ICAC
• 作者: Ali, Ahsan;Pinciroli, Riccardo;Yan, Feng;Smirni, Evgenia
• 通讯作者: Smirni, Evgenia

Machine Learning Models for GPU Error Prediction in a Large Scale HPC System

大规模 HPC 系统中 GPU 错误预测的机器学习模型

• DOI: 10.1109/dsn.2018.00022
• 发表时间: 2018
• 期刊: 2018 48th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN
• 作者: Nie, Bin;Xue, Ji;Gupta, Saurabh;Patel, Tirthak;Engelmann, Christian;Smirni, Evgenia;Tiwari, Devesh
• 通讯作者: Tiwari, Devesh