You are to implement a full-blown (commercial-quality) distributed Peer-to-Peer file sharing system, DGOOGLE, in which users (peers) are capable of efficiently searching for content in the system using a Google-like text search engine. You will need to implement two differe...

Hazelcast is a clustering and highly scalable data distribution platform for Java. *Features:* * Distributed implementations of `java.util.{Queue, Set, List, Map}` * Distributed implementation of `java.util.concurrency.locks.Lock` * Distributed implementation of `java.util.concurrent.Ex...

Project 1. Cooperative Caching in Mobile P2P Databases (two groups each with 5 students) The objective of this project is to design and evaluate cooperative caching schemes for query processing in mobile P2P databases. The recommended schemes are described in Appendix A. Students may come up with...

== What is SWAMP? == SWAMP (the Script Workflow Analysis for MultiProcessing) is a system which aims to relocate scientists' existing data-analyzing shell-scripts to be executed safely and correctly on the data server. SWAMP was conceived from geoscientists' growing discontent over the hassle of d...

*DNFS是“Distributed Network File System”的简称。它是一种运行于Linux集群之上的可扩展分布式文件系统。* =系统架构= DNFS是一个3方架构，包括了元数据服务器（MDS）、对象存储设备（OSD）和客户端（CLIENT）。其中元数据服务器负责系统中文件的元数据管理及对客户端请求的调度；对象存储设备负责文件数据的保存；客户端是用户访问DNFS的入口，在这里，用户看到的DNFS就是一个文件系统。 MDS是系统的核心。所有来自CLIENT的请求都要通过MDS，MDS把CLIENT端写入的文件的元数据写入本地（同名的）元数据文件中，并且把该文件分条...

= VGlass = == Purpose == A general C-like language to write distributed programs freely. == Features == * *inline context conversion*. Sometimes an algorithm have to be executed across, say, clients and servers. If we separate the progress into two parts, and keep correspondence using networ...

=Online Post Certification for Global Computations= Unix middleware based on kaapi, introduces parallelism to any user function dealing with integers. The middle-ware computes a given function modulo several prime values which can be performed in parallel in a grid, introducing parallelism to seq...

=Introduction= Dikeva is a distributed data store aiming to be as simple and easy to deploy as possible. Using Berkeley DB as data storage and written in Python for easy deployment. =What are the requirements for dikeva?= The only requirement at the moment is a correct Python 2.X installation....

==概述== 一个分布式的key-value存储解决方案，本项目最初为参加ecug III时的topic，后由于工作原因疏于维护，更新。最近得于空闲，下定决心要完善此项目，遂重新命名为tengkong，取自中文“腾空”的读音 本项目基本原理得益于amazon的dynamo实现相关介绍。 欢迎有兴趣朋友，参与本项目。 ==特性== * 分布式数据持久存储 * 数据高可用性，容错性强 * 快速的读写操作(99.9%请求再100ms内完成) * 小规模集群(推荐1024节点) * 方便的加入删除节点 * 友好的管理界面 * 易安装部署 * 基...