CSR: SHF: SMALL: Efficient, Low-Latency Networked Storage

CSR：SHF：小型：高效、低延迟的网络存储

• 批准号: 1320071
• 负责人: Christopher Stewart
• 金额: $ 40万
• 依托单位: Ohio State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-10-01 至 2017-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1320071&HistoricalAwards=false
• 关键词: CSR; SHF; SMALL; Efficient; Low

From key-value stores to distributed file systems to distributed databases, networked storage underpins modern Internet services. Networked storage allows programmers to separate logic and data, enables high throughput scale out, and takes advantage of increasingly fast datacenter networks. However, in the big data era, networked storage faces a new challenge: The amount of data accessed per user request is growing rapidly; outpacing processor speeds and DRAM capacity. Increasingly, user-perceived response times are dominated by the slowest storage accesses, i.e., the 99th percentile tails. Networked storage is notorious for fat tailed response times.We are developing networked storage systems that are 1) always fast and 2) cost efficient. A key approach is to understand and selectively use replication for predictability (or cloning). In this approach, clients issue redundant storage accesses against independent hardware resources. The first to respond provides the result. Replication for predictability reduces client-perceived variability, leading to always fast response times. Our implementations are especially cost effective at scale. To lower costs, we study the root causes of slow response times, saving resources by focusing on common causes. We also trade quality---e.g., slightly degraded search engine results--- for lower hardware costs when appropriate. For broader impact, the PI will work to transfer the technology to national and local companies.

从键值存储到分布式文件系统到分布式数据库，网络存储基于现代互联网服务。网络存储使程序员可以分开逻辑和数据，使高吞吐量缩放量表，并利用日益快速的数据中心网络。 但是，在大数据时代，网络存储面临着一个新的挑战：每个用户请求访问的数据量正在迅速增长；超过处理器的速度和DRAM容量。用户感知的响应时间越来越多地由最慢的存储访问（即第99个百分位数尾巴）主导。 网络存储在脂肪尾部的响应时间中臭名昭著。我们正在开发1）始终快速和2）成本效率的网络存储系统。 一种关键方法是理解并有选择地使用复制来可预测性（或克隆）。 在这种方法中，客户会针对独立的硬件资源发布冗余存储访问。第一个响应提供了结果。 可预测性的复制可降低客户感知的可变性，从而始终导致快速响应时间。 我们的实现在大规模上特别有效。 为了降低成本，我们研究了缓慢响应时间的根本原因，从而通过关注共同原因来节省资源。 我们还交易质量 - 例如，搜索引擎结果略有退化 - 在适当的情况下以较低的硬件成本。 为了获得更广泛的影响，PI将致力于将技术转移到国家和本地公司。

项目成果

Scaling Analytics through Approximate and Distributed Computing

通过近似和分布式计算扩展分析

• 发表时间: 2017
• 期刊: OSU Masters Thesis
• 作者: Chakrabarti, Aniket

Obtaining and Managing Answer Quality for Online Data-Intensive Services

获取和管理在线数据密集型服务的答案质量

• 发表时间: 2017
• 期刊: ACM Transactions on Modeling and Performance Evaluation of Computing Systems
• 影响因子: 0.6
• 作者: Kelley, Jaimie;Stewart, Christopher;Morris, Nathaniel;Tiwari, Devesh;He, Yuxiong;Elnikety, Sameh
• 通讯作者: Elnikety, Sameh

Resource Allocation using Adaptive Characterization of Online, Data-Intensive Workloads

• 发表时间: 2017
• 作者: Jaimie Kelley