This paper presents the design and evaluation of Pastry, a scalable, distributed object location and routing substrate for wide-area peer-to-peer ap- plications. Pastry performs application-level routing and object location in a po- tentially very large overlay network of nodes connected via the Internet. It can be used to support a variety of peer-to-peer applications, including global data storage, data sharing, group communication and naming. Each node in the Pastry network has a unique identifier (nodeId). When presented with a message and a key, a Pastry node efficiently routes the message to the node with a nodeId that is numerically closest to the key, among all currently live Pastry nodes. Each Pastry node keeps track of its immediate neighbors in the nodeId space, and notifies applications of new node arrivals, node failures and recoveries. Pastry takes into account network locality; it seeks to minimize the distance messages travel, according to a to scalar proximity metric like the number of IP routing hops. Pastry is completely decentralized, scalable, and self-organizing; it automatically adapts to the arrival, departure and failure of nodes. Experimental results obtained with a prototype implementation on an emulated network of up to 100,000 nodes confirm Pastry's scalability and efficiency, its ability to self-organize and adapt to node failures, and its good network locality properties.

/pdf/pastry-scalable-decentralized-object-location-and-routing-1u2equbbbp.pdf

Pastry: Scalable, Decentralized Object Location, and Routing for Large-Scale Peer-to-Peer Systems

A case for end system multicast

In today’s chaotic network, data and services are mobile and replicated widely for availability, durability, and locality. Components within this infrastructure interact in rich and complex ways, greatly stressing traditional approaches to name service and routing. This paper explores an alternative to traditional approaches called Tapestry. Tapestry is an overlay location and routing infrastructure that provides location-independent routing of messages directly to the closest copy of an object or service using only point-to-point links and without centralized resources. The routing and directory information within this infrastructure is purely soft state and easily repaired. Tapestry is self-administering, faulttolerant, and resilient under load. This paper presents the architecture and algorithms of Tapestry and explores their advantages through a number of experiments.

/pdf/tapestry-an-infrastructure-for-fault-tolerant-wide-area-3mweaivgsc.pdf

Tapestry: An Infrastructure for Fault-tolerant Wide-area Location and Routing

A Resilient Overlay Network (RON) is an architecture that allows distributed Internet applications to detect and recover from path outages and periods of degraded performance within several seconds, improving over today's wide-area routing protocols that take at least several minutes to recover. A RON is an application-layer overlay on top of the existing Internet routing substrate. The RON nodes monitor the functioning and quality of the Internet paths among themselves, and use this information to decide whether to route packets directly over the Internet or by way of other RON nodes, optimizing application-specific routing metrics.Results from two sets of measurements of a working RON deployed at sites scattered across the Internet demonstrate the benefits of our architecture. For instance, over a 64-hour sampling period in March 2001 across a twelve-node RON, there were 32 significant outages, each lasting over thirty minutes, over the 132 measured paths. RON's routing mechanism was able to detect, recover, and route around all of them, in less than twenty seconds on average, showing that its methods for fault detection and recovery work well at discovering alternate paths in the Internet. Furthermore, RON was able to improve the loss rate, latency, or throughput perceived by data transfers; for example, about 5% of the transfers doubled their TCP throughput and 5% of our transfers saw their loss probability reduced by 0.05. We found that forwarding packets via at most one intermediate RON node is sufficient to overcome faults and improve performance in most cases. These improvements, particularly in the area of fault detection and recovery, demonstrate the benefits of moving some of the control over routing into the hands of end-systems.

/pdf/resilient-overlay-networks-2see5sqvzl.pdf

Resilient overlay networks

/pdf/resilient-overlay-networks-e0tv6oh3hj.pdf

An intelligent file system implements flexible content-based associative access to file system data with a conservative extension of conventional tree structured file system naming semantics. Intelligent file systems are compatible with existing UNIX and NFS file systems and permit existing applications to take advantage of content based access to file system objects. Based upon our experience with an intelligent file system implementation we explore how an intelligent file system might be used in future operating systems to store structured objects.

Intelligent File Systems for Object Repositories

Information storage systems enable computational activities to preserve valuable information that can be shared by multiple users and for multiple purposes Automatic indexing and garbage collection are useful features of storage systems because they can make information sharing more convenient and effective Automatic indexing enables users and programmers to locate and access shared information without substantial prior agreement about information structure and index design, as is typically necessary in traditional file systems and databases Automatic garbage collection enables users and programmers to share information without requiring complete agreement on ownership and deallocation protocols for the shared objects
My thesis demonstrates, via two working systems, that update logs can be used to provide automatic garbage collection and indexing services in information storage systems The garbage collection system uses log-based replicating garbage collection, a new algorithm that enables concurrent compacting garbage collection of persistent storage In this garbage collector design, an update log is used to record client operations that require subsequent action by the collector The update log decouples the operations of application programs from the operations of the garbage collector The indexing file system uses an update log to decouple indexing operations from file modification operations Both the indexing system and the garbage collection system use update logs in novel ways to implement convergent consistency properties that promote convenient and efficient sharing of information (Copies available exclusively from MIT Libraries, Rm 14-0551, Cambridge, MA 02139-4301 Ph 617-253-5668; Fax 617-253-1690)

David K. Gifford

Papers

Intelligent File Systems for Object Repositories

Effective information sharing using update logs