A semantic file system is an information storage system that provides flexible associative access to the system's contents by automatically extracting attributes from files with file type specific transducers. Associative access is provided by a conservative extension to existing tree-structured file system protocols, and by protocols that are designed specifically for content based access. Compatibility with existing file system protocols is provided by introducing the concept of a virtual directory. Virtual directory names are interpreted as queries, and thus provide flexible associative access to files and directories in a manner compatible with existing software. Rapid attribute-based access to file system contents is implemented by automatic extraction and indexing of key properties of file system objects. The automatic indexing of files and directories is called "semantic" because user programmable transducers use information about the semantics of updated file system objects to extract the properties for indexing. Experimental results from a semantic file system implementation support the thesis that semantic file systems present a more effective storage abstraction than do traditional tree structured file systems for information sharing and command level programming.

This paper describes a C toolkit for easily extending the Unix file system. The toolkit exposes the NFS interface, allowing new file systems to be implemented portably at user level. A number of programs have implemented portable, user-level file systems. However, they have been plagued by low-performance, deadlock, restrictions on file system structure, and the need to reboot after software errors. The toolkit makes it easy to avoid the vast majority of these problems. Moreover, the toolkit also supports user-level access to existing file systems through the NFS interface---a heretofore rarely employed technique. NFS gives software an asynchronous, low-level interface to the file system that can greatly benefit the performance, security, and scalability of certain applications. The toolkit uses a new asynchronous I/O library that makes it tractable to build large, event-driven programs that never block.

This thesis presents a new architecture for information discovery based on a hierarchy of content routers that provide both browsing and search services to end users. Content routers catalog information servers, which may in turn be other content routers. The resulting hierarchy of content routers and leaf servers provides a rich set of services to end users for locating information, including query refinement and query routing. Query refinement helps a user improve a query fragment to describe the user's interests more precisely. Once a query has been refined and describes a manageable result set, query routing automatically forwards the query to relevant servers. These services make use of succinct descriptions of server contents called content labels. A unique contribution of this research is the demonstration of a scalable discovery architecture based on a hierarchical approach to routing.

We describe the first system that provides query based associative access to the contents of distributed information servers. Queries describe desired object attributes, and are automatically forwarded to servers that contain relevant information. In typical distributed information systems there are so many objects that underconstrained queries can produce large result sets and extraordinary processing costs. To deal with this scaling problem we use content labels to permit users to learn about available resources and to quickly formulate queries with adequate discriminatory power. We present experimental data that show that certain content label attributes can be automatically chosen. We have implemented associative access to a distributed set of information servers in the content routing system. A content routing system is organized as a network of servers called content routers that present a single query based image of a distributed information system. Experiments motivated by our video access service show that substantial performance benefits result when content routers are removed from the client-server path once an object of interest is found.

1 Introduction The Internet contains over one million hosts that provide file service and other information servers specializing in topics such as news, technical reports, biology, geography, and politics. The Internet's vast collection of servers can be viewed as a distributed database containing a wealth of information. Unfortunately, this information is relatively inaccessible because there is no mechanism for browsing and searching it associatively. The difficulty of providing associative access to a large number of distributed information servers lies primarily in problems of scale. The scale of the Internet, which is today only a fraction of its eventual size, is so great as to render infeasible any comprehensive indexing plan based on a single global index. In addition, the cost of distributing a query throughout the Internet is prohibitive. Finally, in very large scale systems, the number of results to a typical user query is incomprehensibly large. For these reasons, we expect that efficient associative access will require both content routing and query refinement. Content routing is the process of directing user queries to appropriate servers. Query refinement helps a user formulate meaningful queries. These query services can be implemented by a content routing system that will:

This paper describes a C++ toolkit for easily extending the Unix file system. The toolkit exposes the NFS interface, allowing new file systems to be implemented portably at user level. A number of programs have implemented portable, user-level file systems. However, they have been plagued by low-performance, deadlock, restrictions on file system structure, and the need to reboot after software errors. The toolkit makes it easy to avoid the vast majority of these problems. Moreover, the toolkit also supports user-level access to existing file systems through the NFS interface—a heretofore rarely employed technique. NFS gives software an asynchronous, low-level interface to the file system that can greatly benefit the performance, security, and scalability of certain applications. The toolkit uses a new asynchronous I/O library that makes it tractable to build large, event-driven programs that never block. 1

The Direct Access File System (DAFS) is an emerging industrial standard for network-attached storage. DAFS takes advantage of new user-level network interface standards. This enables a user-level file system structure in which client-side functionality for remote data access resides in a library rather than in the kernel. This structure addresses longstanding performance problems stemming from weak integration of bu#ering layers in the network transport, kernel-based file systems and applications. The benefits of this architecture include lightweight, portable and asynchronous access to network storage and improved application control over data movement, caching and prefetching.

This handout is about orderly ways of naming complicated collections of objects in a computer system. A basic technique for understanding a big system is to describe it as a collection of simple parts. Being able to name these parts is a necessary aspect of such a description, and often the most important aspect.

A semantic provides le system is an information storage system that exible associative access to the system's contents by automatically extracting attributes from les with le type speci c transducers. Associative access is provided by a conservative extension to existing tree-structured protocols, and by protocols that are designed speci le system cally for content based access. Compatiblity with existing le system protocols is provided by introducing the concept of a virtual directory. Virtual directory names are interpreted as queries, and thus provide exible associative access to les and directories in a manner compatible with existing software. Rapid attribute-based access to le system contents is implemented by automatic extraction and indexing of key properties of le system objects. The automatic indexing of les and directories is called \semantic " because user programmable transducers use information about the semantics of updated le system objects to extract the properties for indexing. Experimental results from a semantic le system implementation support the thesis that semantic le systems present a more e ective storage abstraction than do traditional tree structured le systems for information sharing and command level programming. 1

Using DNS SRV to Specify a Global File Namespace with NFS Version 4 The NFS version 4 (NFSv4) protocol provides a mechanism for a collection of NFS file servers to collaborate in providing an organization-wide file namespace. The DNS SRV Resource Record (RR) allows a simple way for an organization to publish the root of its file system namespace, even to clients that might not be intimately associated with such an organization. The DNS SRV RR can be used to join these organization-wide file namespaces together to allow construction of a global, uniform NFS file namespace. Status of This Memo This is an Internet Standards Track document. This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the