ITR: Self-* Storage Systems

ITR：自助*存储系统

• 批准号: 0326453
• 负责人: Gregory Ganger
• 金额: $ 150万
• 依托单位: Carnegie-Mellon University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-10-15 至 2008-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0326453&HistoricalAwards=false
• 关键词: ITR; Self; Storage; Systems

Human administration of storage systems (including capacity planning,back-up, load balancing, etc.) is a large and growing issue in modernIT infrastructures; reported ratios of 1 sysadmin per 1-10TB of storageare scary with multi-petabyte data centers on the horizon.This research program defines and explores a new storage architecturethat integrates automated management functions and simplifies the humanadministrative task. We refer to conforming systems as self-* storagesystems in an attempt to capture many recent buzzwords in a singlemeta-buzzword; self-* systems should be self-configuring, self-organizing,self-tuning, self-healing, self-managing, etc. Human administratorsprovide only muscle for component addition and guidance on acceptablerisk levels and current levels of satisfaction. Borrowing organizationalideas from corporate structure and technologies from AI and controlsystems, self-* storage should simplify storage administration, increasesystem robustness, and simplify system construction.

存储系统的人工管理(包括容量规划、备份、负载平衡等)在现代IT基础架构中是一个巨大且日益严重的问题；报告的每1-10TB存储的系统管理员比率对于即将出现的数PB数据中心来说是可怕的。此研究计划定义并探索了一种新的存储体系结构，该体系结构集成了自动化管理功能并简化了人工管理任务。我们将一致性系统称为自*存储系统，以试图在一个单一的流行语中捕获许多最新的流行语；自*系统应该是自配置、自组织、自调优、自修复、自管理等。人力管理员只为组件添加提供力量，并就可接受的风险水平和当前的满意度水平提供指导。借鉴公司结构的组织思想以及人工智能和控制系统的技术，自*存储应该简化存储管理，增强系统的健壮性，并简化系统建设。