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...

The Logical Disk (LD) defines a new interface to disk storage that separates file management and disk management by using logical block numbers and block lists. The LD interface is designed to support multiple file systems and to allow multiple implementations, both of which are important given the increasing use of kernels that support multiple operating system personalities. A log-structured implementation of LD (LLD) demonstrates that LD can be implemented efficiently. LLD adds about 5% to 10% to the purchase cost of a disk for the main memory it requires. Combining LLD with an existing file system results in a log-structured file system that exhibits the same performance characteristics as the Sprite log-structured file system.

Delta compression, which consists of compactly encoding one le version as the result of changes to another, can improve eciency in the use of network and disk resources. Delta compression techniques are readily available and can result in compression factors of ve to ten on typical data. Managing delta-compressed storage, however, is a dicult task. I will present a system that attempts to isolate the complexity of delta-compressed storage management by separating the task of version labeling from performance issues. I will show how the system integrates delta-compressed transport with delta-compressed storage. Existing tools for managing delta-compressed storage suer from weak le system support. Lack of transaction support is responsible for inecient application behavior. The only atomic operation in the traditional le system forces unnecessary disk activity due to copying costs. I will demonstrate that transaction support can improve application performance and extensibility wit...

le application-specific use of I/O devices. The architecture is applied to several representative classes of device including network interfaces, network connected peripherals, disk drives and framestores. Of these, disks and framestores are of particular interest since they must be shared at a very fine granularity but have traditionally been presented to the application via a window system or file-system with a high-level and coarse-grained interface. A device driver for the framestore is presented which abstracts the device at a low level and is therefore able to provide each client with guaranteed bandwidth to the framebuffer. The design and implementation of a novel client-rendering window system is then presented which uses this driver to enable rendering code to be safely migrated into a shared library within the client. A low-level abstraction of a standard disk drive is also described which efficiently supports a wide variety of file systems, and other applications requiring

This paper presents the design and implementation of a transaction manager embedded in a log-structured file system [11]. Measurements show that transaction support on a log-structured file system offers a 10% performance improvement over transaction support on a conventional, read-optimized file system. When the transaction manager is embedded in the log-structured file system, the resulting performance is comparable to that of a more traditional, user-level system. The performance results also indicate that embedding transactions in the file system need not impact the performance of nontransaction applications. 1. Introduction Traditionally, transaction support has been provided by database management systems running as user-level processes. These systems maintain their own cache of frequently accessed data pages and perform their own scheduling, allowing multiple processes to access a single database. When transaction support is provided as an operating system service, concurrency ...

Abstract not found

Transactional flash (TxFlash) is a novel solid-state drive (SSD) that uses flash memory and exports a transactional interface (WriteAtomic) to the higher-level software. The copy-on-write nature of the flash translation layer and the fast random access makes flash memory the right medium to support such an interface. We further develop a novel commit protocol called cyclic commit for TxFlash; the protocol has been specified formally and model checked. Our evaluation, both on a simulator and an emulator on top of a real SSD, shows that TxFlash does not increase the flash firmware complexity significantly and provides transactional features with very small overheads (less than 1%), thereby making file systems easier to build. It further shows that the new cyclic commit protocol significantly outperforms traditional commit for small transactions (95 % improvement in transaction throughput) and completely eliminates the space overhead due to commit records. 1

Today, software errors are the leading cause of outages in fault tolerant systems. System availability can be improved despite software errors by fast error detection and recovery techniques that minimize total downtime after an outage. This dissertation analyzes software errors in three commercial systems and describes the implementation and evaluation of several techniques for early error detection and fast recovery in a database management system (DBMS). The software error study examines errors reported by customers in three IBM systems programs: the MVS operating system and the IMS DBMS and DB2 DBMS. The study classifies errors by the type of coding mistake and the circumstances in the customer's environment that caused the error to arise. It observes a higher availability impact from addressing errors, such as uninitialized pointers, than software errors as a whole. It also details the frequencies and types of addressing errors and characterizes the damage they do. The error detec...

We have implemented an integrated and configurable file system called the Pegasus file-system (PFS) and a trace-driven file-system simulator called Patsy. Patsy is used for off-line analysis of file-system algorithms, PFS is used for on-line file-system data storage. Algorithms are first analyzed in Patsy and when we are satisfied with the performance results, migrated into PFS for on-line usage. Since Patsy and PFS are derived from a common cut-and-paste file-system framework, this migration proceeds smoothly. We have found this integration quite useful: algorithm bottlenecks have been found through Patsy that could have led to performance degradations in PFS. Off-line simulators are simpler to analyze compared to on-line file-systems because a work load can repeatedly be replayed on the same off-line simulator. This is almost impossible in on-line file-systems since it is hard to provide similar conditions for each experiment run. Since simulator and file-system are integrated (hence...

We present the design of a log-structured tertiary storage system (LTS). The advantage of this approach is that it allows the system to hide the details of jukebox robotics and media characteristics behind a uniform, random access, block-oriented interface. It also allows the system to avoid media mount operations for writes, giving write performance similar to that of secondary storage. 1