We propose an I/O classification architecture to close the widening semantic gap between computer systems and storage systems. By classifying I/O, a computer system can request that different classes of data be handled with different storage system policies. Specifically, when a storage system is first initialized, we assign performance policies to predefined classes, such as the filesystem journal. Then, online, we include a classifier with each I/O command (e.g., SCSI), thereby allowing the storage system to enforce the associated policy for each I/O that it receives. Our immediate application is caching. We present filesystem prototypes and a database proof-of-concept that classify all disk I/O — with very little modification to the filesystem, database, and operating system. We associate caching policies with various classes (e.g., large files shall be evicted before metadata and small files), and we show that end-to-end file system performance can be improved by over a factor of two, relative to conventional caches like LRU. And caching is simply one of many possible applications. As part of our ongoing work, we are exploring other classes, policies and storage system mechanisms that can be used to improve end-to-end performance, reliability and security.

All too often, decisions about whom to trust in computer systems are driven by the needs of system management rather than data security. In particular, data storage is often entrusted to people who have no role in creating or using the data---through outsourcing of data management, hiring of outside consultants to administer servers, or even collocation servers in physically insecure machine rooms to gain better network connectivity.

The Plutus file system introduced the notion of key rotation as a means to derive a sequence of temporally-related keys from the most recent key. In this paper we show that, despite natural intuition to the contrary, key rotation schemes cannot generically be used to key other cryptographic objects; in fact, keying an encryption scheme with the output of a key rotation scheme can yield a composite system that is insecure. To address these shortcomings, we introduce a new cryptographic object called a key regression scheme, and we propose three constructions that are provably secure under standard cryptographic assumptions. We implement key regression in a secure file system and empirically show that key regression can significantly reduce the bandwidth requirements of a content publisher under realistic workloads using lazy revocation. Our experiments also serve as the first empirical evaluation of either a key rotation or key

Cluster-based storage systems are popular for data-intensive applications and it is desirable yet challenging to provide incremental expansion and high availability while achieving scalability and strong consistency. This paper presents the design and implementation of a self-organizing storage cluster called Sorrento, which targets data-intensive workload with highly parallel requests and low write-sharing patterns. Sorrento automatically adapts to storage node joins and departures, and the system can be configured and maintained incrementally without interrupting its normal operation. Data location information is distributed across storage nodes using consistent hashing and the location protocol differentiates small and large data objects for access efficiency. It adopts versioning to achieve single-file serializability and replication consistency. In this paper, we present experimental results to demonstrate features and performance of Sorrento using microbenchmarks, application benchmarks, and application trace replay.

This paper describes the design, implementation and performance of a family of protocols for survivable, decentralized data storage. These protocols exploit storage-node versioning to efficiently achieve strong consistency semantics. These protocols allow erasure-codes to be used that achieve network and storage efficiency (and optionally data confidentiality in the face of server compromise). The protocol family is general in that its parameters accommodate a wide range of fault and timing assumptions, up to asynchrony and Byzantine faults of both storage-nodes and clients, with no changes to server implementation or client-server interface. Measurements of a prototype storage system using these protocols show that the protocol performs well under various system model assumptions, numbers of failures tolerated, and degrees of reader-writer concurrency.

This paper describes a consistency protocol that exploits versioning storage-nodes. The protocol provides linearizability with the possibility of read aborts in an asynchronous system that may suffer client and storage-node crash failures. The protocol supports both replication and erasure coding (which precludes post hoc repair of partial-writes), and avoids the excess work of two-phase commits. Versioning storagenodes allow the protocol to avoid excess communication in the common case of no write sharing and no failures of writing clients.

A local disk-based file system, LDFS, is an attractive way to speed up distributed applications. Local file access is much faster than accessing data on remote file servers through the network. LDFS is also scalable, as it does not rely on centralized file servers, and it exploits already existing resources (local disks) to provide storage. However, since individual workstations are less reliable and less available than file servers, LDFS must be made fault tolerant. We present an approach that integrates the LDFS with the distributed application. This is particularly suitable for mobile agent systems, because they can easily migrate to access remote files. LDFS avoids logging of individual file accesses, which are regenerated automatically from application messages. Our experiments show that the overhead of checkpointing with LDFS is generally smaller that with NFS, while access time to files decreases dramatically. Keywords: fault tolerant computing, stable storage, mobile agent distributed systems 1.

This paper gives a survey of all common transparent distributed systems. We distinguish between Distributed File Systems (DFS) and Distributed Operating Systems (DOS). Our overview is focussed on systems providing at least access or location transparency. The paper is organized as follows: The introduction offers definitions of the features of each transparent distributed system as well as the services it is able to provide. We also propose a catalog of criteria that enables us to compare different systems independently of implementation done. The main entries we make are heterogeneity of the system's environment, communication strategy, as well as naming and security issues. Finally, we examine the reliability and availability of the separate systems and the way these issues are achieved. The following section consists of the survey. The description of each system is organized as follows: First, we introduce the main goal the system was developed for, the classification of th...

Fault-tolerant storage systems spread data redundantly across a set of storage-nodes in an effort to preserve and provide access to data despite failures. One difficulty created by this architecture is the need for a consistent view, across storage-nodes, of the most recent update. Such consistency is made difficult by concurrent updates, partial updates made by clients that fail, and failures of storage-nodes. This thesis demonstrates a novel approach to achieving scalable, highly fault-tolerant storage systems by leveraging a set of efficient and scalable, strong consistency protocols enabled by storage-node versioning. Versions maintained by storage-nodes can be used to provide consistency, without the need for central serialization, and despite concurrency. Since versions are maintained for every update, even if a client fails part way through an update, concurrency exists during an update, the latest complete version of the data-item being accessed still exists in the system—it does not get destroyed by subsequent updates.

This paper presents a design principle that helps guide placement of functions among the