Collaborative Research: Programming Models and Storage System for High Performance Computation with Many-Core Processors

随着每颗芯片的核心数量的快速增加，存储系统正以惊人的速度下降，因此未来使用多核芯片构建的高端计算（HEC）系统面临的主要挑战是存储系统。需要大量数据集和/或高输入输出带宽的数据密集型应用将特别容易受到这些趋势的影响。从历史上看，HEC系统的存储架构在很大程度上受到底层操作系统（OS）中的文件系统接口的限制。本研究的重点是探索一种基于一次写树结构的新存储模型。本研究将探索存储系统用户的三种编程模型，所有这些模型都可以通过共享持久数据进行互操作：1)一个声明式编程模型，在这个模型中，任何数据结构都可以直接在存储系统中持久化，而不需要程序员的干预；2)一个强类型命令式编程模型，在这个模型中，类型系统扩展将被用来强制区分可以直接持久化的数据结构和那些不能持久化的数据结构；3)一个弱类型运行时接口，使低级C程序能够访问存储系统。

AbstractA major challenge for future High End Computing (HEC) systems built using many-core chips is the storage system since the available memory and bandwidth per processor core is starting to decline at an alarming rate, with the rapid increase in the number of cores per chip. Data-intensive applications that require large data sets and/or high input-output bandwidth will be especially vulnerable to these trends. Historically, the storage architecture of an HEC system has been constrained to a large degree by the filesystem interfaces in the underlying Operating System (OS). The specific focus of this research is on exploring a new storage model based on write-once tree structures. This research will explore three programming models for users of the storage system, all of which can inter-operate through shared persistent data: 1) a declarative programming model in which any data structure can be directly made persistent in the storage system, with no programmer intervention, 2) a strongly-typed imperative programming model in which a type system extension will be used to enforce a separation between data structures that can be directly made persistent and those that cannot, and 3) a weakly-typed runtime interface that enables low-level C programs to access the storage system.