In this paper, we propose a new paradigm for network file system design, serverless network file systems. While traditional network file systems rely on a central server machine, a serverless system utilizes workstations cooperating as peers to provide all file system services. Any machine in the system can store, cache, or control any block of data. Our approach uses this location independence, in combination with fast local area networks, to provide better performance and scalability than traditional file systems. Further, because any machine in the system can assume the responsibilities of a failed component, our serverless design also provides high availability via redundant data storage. To demonstrate our approach, we have implemented a prototype serverless network file system called xFS. Preliminary performance measurements suggest that our architecture achieves its goal of scalability. For instance, in a 32-node xFS system with 32 active clients, each client receives nearly as much read or write throughput as it would see if it were the only active client.

Zebra is a network file system that increases throughput by striping file data across multiple servers. Rather than striping each file separately, Zebra forms all the new data from each client into a single stream, which it then stripes using an approach similar to a log-structured file system. This provides high performance for writes of small files as well as for reads and writes of large files. Zebra also writes parity information in each stripe in the style of RAID disk arrays; this increases storage costs slightly but allows the system to continue operation even while a single storage server is unavailable. A prototype implementation of Zebra, built in the Sprite operating system, provides 4-5 times the throughput of the standard Sprite file system or NFS for large files and a 15 % to 300 % improvement for writing small files. 1

Configuring redundant disk arrays is a black art. To configure an array properly, a system administrator must understand the details of both the array and the workload it will support. Incorrect understanding of either, or changes in the workload over time, can lead to poor performance. We present a solution to this problem: a two-level storage hierarchy implemented inside a single diskarray controller. In the upper level of this hierarchy, two copies of active data are stored to provide full redundancy and excellent performance. In the lower level, RAID 5 parity protection is used to provide excellent storage cost for inactive data, at somewhat lower performance. The technology we describe in this paper, known as HP AutoRAID, automatically and transparently manages migration of data blocks between these two levels as access patterns change. The result is a fully redundant storage system that is extremely easy to use, is suitable for a wide variety of workloads, is largely insensitive to dynamic workload changes, and performs much better than disk arrays with comparable numbers of spindles and much larger amounts of front-end RAM cache. Because the implementation of the HP AutoRAID technology is almost entirely in software, the additional hardware cost for these benefits is very small. We describe the HP AutoRAID technology in detail, provide performance data for an embodiment of it in a storage array, and summarize the results of simulation studies used to choose algorithms implemented in the array.

Mobile computers typically spin down their hard disk after a fixed period of inactivity. If this threshold is too long, the disk wastes energy; if it is too short, the delay due to spinning the disk up again frustrates the user. Usage patterns change over time, so a single fixed threshold may not be appropriate at all times. Also, different users may have varying priorities with respect to trading off energy conservation against performance. We describe a method for varying the spin-down threshold dynamically by adapting to the user's access patterns and priorities. Adaptive spin-downcan in some circumstances reduce by up to 50% the number of disk spin-ups that are deemed by the user to be inconvenient, while only moderately increasing energy consumption. 1 Introduction In today's mobile computers, the hard disk is typically spun down after a fixed period of inactivity in order to conserve energy. When the disk is next accessed, it is spun up again, which can cause a delay of a few s...

Despite impressive advances in file system throughput resulting from technologies such as high-bandwidth networks and disk arrays, file system latency has not improved and in many cases has become worse. Consequently, file system I/O remains one of the major bottlenecks to operating system performance [10]. This paper investigates an automated predictive approach towards reducing file latency. Automatic Prefetching uses past file accesses to predict future file system requests. The objective is to provide data in advance of the request for the data, effectively masking access latencies. We have designed and implement a system to measure the performance benefits of automatic prefetching. Our current results, obtained from a trace-driven simulation, show that prefetching results in as much as a 280 % improvement over LRU especially for smaller caches. Alternatively, prefetching can reduce cache size by up to 50%.

Research results [ROSE91] demonstrate that a log-structured file system (LFS) offers the potential for dramatically improved write performance, faster recovery time, and faster file creation and deletion than traditional UNIX file systems. This paper presents a redesign and implementation of the Sprite [ROSE91] log-structured file system that is more robust and integrated into the vnode interface [KLEI86]. Measurements show its performance to be superior to the 4BSD Fast File System (FFS) in a variety of benchmarks and not significantly less than FFS in any test. Unfortunately, an enhanced version of FFS (with read and write clustering) [MCVO91] provides comparable and sometimes superior performance to our LFS. However, LFS can be extended to provide additional functionality such as embedded transactions and versioning, not easily implemented in traditional file systems. 1. Introduction Early UNIX file systems used a small, fixed block size and made no attempt to optimize block place...

With the advent and subsequent popularity of portable computers, power management of system components has become an important issue. Current portable computers implement anumber of power reduction techniques to achieve a longer battery life. Included among these is spinning down a disk during long periods of inactivity. In this paper, we perform a quantitative analysis of the potential costs and bene ts of spinning down the disk drive as apower reduction technique. Our conclusion is that almost all the energy consumed by a disk drive can be eliminated with little loss in performance. Although on current hardware, reliability can be impacted by our policies, the next generation of disk drives will use technology (such as dynamic head loading) which is virtually una ected by repeated spinups. We found that the optimal spindown delay time, the amount of time the disk idles before it is spun down, is 2 seconds. This di ers signi cantly from the 3-5 minutes in current practice by industry. We will show in this paper the e ect of varying the spindown delay onpower consumption � one conclusion is that a 3-5 minute delay results in only half of the potential bene t of spinning down a disk. 1

Many people have observed that computer systems spend much of their time idle, and various schemes have been proposed to use this idle time productively. The commonest approach is to off-load activity from busy periods to less-busy ones in order to improve system responsiveness. In addition, speculative work can be performed in idle periods in the hopes that it will be needed later at times of higher utilization, or non-renewable resource like battery power can be conserved by disabling unused resources. We found opportunities to exploit idle time in our work on storage systems, and after a few attempts to tackle specific instances of it in ad hoc ways, began to investigate general mechanisms that could be applied to this problem. Our results include a taxonomy of idle-time detection algorithms, metrics for evaluating them, and an evaluation of a number of idleness predictors that we generated from our taxonomy. 1. Introduction Resource usage is often bursty: periods of high utilizat...

One of the major challenges of post-PC computing is the need to reduce energy consumption, thereby extending the lifetime of the batteries that p ower these mobile devices. Memory is a particularly important tar get for e orts to improve energy e ciency. Memory technolo gy is becoming available that o ers power management featur es such as the ability to put individual chips in any one of several di erent power modes. In this paper we explor e the interaction of page plac ement with static and dynamic hardware policies to exploit these emer ginghardwar efeatur es. In p articular, we c onsider p age allo cation p olicies that ancbe employed by an informed operating system to complement the hardware power management strategies. We perform experiments using two complementary simulation envir onments: a tracedriven simulator with workload traces that are representative of mobile computing and an execution-driven simulator with a detaile d processor/memory model and a more memoryintensive set of benchmarks (SPEC2000). Our r esults make a compelling case for a cooperative hardwar e/software approach for exploiting power-aware memory, with down to as little as 45 % of the Energy Delay for the best static policy and 1 % to 20 % of the Ener gyDelay for a traditional fullpower memory. 1.

This paper describes the architecture of eNVy, a large non-volatile main memory storage system built primarily with Flash memory. eNVy presents its storage space as a linear, memory mapped array rather than as an emulated disk in order to provide an efficient and easy to use software interface. Flash memories...