Implicit replication in a network file server

Bhide, Anupam; Elnozahy, E. N. Mootaz; Morgan, Stephen P.

doi:10.1109/mrd.1990.138251

“…There are several projects that provide high availability for NFS or VFS The HA-NFS system [5] uses a variant of the primary copy scheme with a write-ahead log to provide high-avai~abi! ?ty and atomicity, but the log is written directly to disk (or disks), which are shared by the primary and backup, and the system as a whole does not handle partitions (it runs on a local area network).…”

Section: Related Workmentioning

confidence: 99%

“…The Deceit system [22] also provides high availability within NFS but uses a different replication method, namely ISIS [6]; we expect to achieve better performance. (For example, Deceit has higher overhead than HA-NFS [5].) The Ficus system [14] supports high availability for VFS.…”

Section: Related Workmentioning

confidence: 99%

Replication in the Harp File System

Liskov¹,

Ghemawat²,

Gruber³

et al. 1981

View full text Add to dashboard Cite

Abstracttechnique [1,26,27]. In this method, client calls are directed to a single primary server, which communicates This paper describes the design and implementation of the with other backup servers and waits for them to respond Harp file system. Harp is a replicated Unix file system before replying to the client. The system masks failures by accessible via the VFS interface. It provides highly availperforming a failover algorithm in which an inaccessible able and reliable storage for files and guarantees that file server is removed from service. When a primary performs operations are executed atomically in spite of concurrency an operation, it must inform enough backups to guarantee and failures. It uses a novel variation of the primary copy that the effects of that operation will survive all subsequent replication technique that provides good performance befailovers. cause it allows us to trade disk accesses for network comHarp is one of the first implementations of a primary munication. Harp is intended to be used within a file sercopy scheme that runs on conventional hardware. It has vice in a distributed network; in our current implemensome novel features that allow it to perform well. The key tation, it is accessed via NFS. Preliminary performance performance issues are how to provide quick response for results indicate that Harp provides equal or better response user operations and how to provide good system capacity time and system capacity than an unreplicated implemen-(roughly, the number of operations the system can handle tation of NFS that uses Unix files directly.in some time period while still providing good response time). Harp achieves good performance by recording the

show abstract

Experiences with two high availability designs (replication techniques)

Bhide

¹

[1992 Proceedings] Second Workshop on the Management of Replicated Data

0

View full text Add to dashboard Cite

In this paper, we compare two replication schemes designed to provide high availability in an efficient manner: HA-NFS and ARM. Both schemes use the primary copy method for replica control.Both schemes were designed with the goal of minimizing the overheads during failure-free operation. In a primary copy scheme these overheads primarily consist of updating the secondary replicas. The two schemes were designed for different applications: ARM for providing high availability in a Shared Nothing database system, HA-NFS for providing high availability in an NFS file server environment. They also differ in that the HA-NFS scheme uses dual-ported disks to provide high availability, the ARM scheme uses replication over a network. In spite of the seemingly major differences, the schemes have the same key conceptual idea viz. propagating updates asynchronously to remote replicas. In addition to this idea, HA-NFS uses an unusual hardware arrangement in the form of dual-ported disks to further the lower the overhead of updating secondary replicas.In addition to discussing efficient updates, we will also discuss how each of the schemes handles the three key problems of replica control viz. ensuring one-copy serializability, preventing loss of updates of committed transactions and handling network partitions.

show abstract

Replication in the harp file system

Liskov

¹

,

Ghemawat

²

,

Gruber

³

et al. 1991

Proceedings of the Thirteenth ACM Symposium on Operating Systems Principles

View full text Add to dashboard Cite

Abstracttechnique [1,26,27]. In this method, client calls are directed to a single primary server, which communicates This paper describes the design and implementation of the with other backup servers and waits for them to respond Harp file system. Harp is a replicated Unix file system before replying to the client. The system masks failures by accessible via the VFS interface. It provides highly availperforming a failover algorithm in which an inaccessible able and reliable storage for files and guarantees that file server is removed from service. When a primary performs operations are executed atomically in spite of concurrency an operation, it must inform enough backups to guarantee and failures. It uses a novel variation of the primary copy that the effects of that operation will survive all subsequent replication technique that provides good performance befailovers. cause it allows us to trade disk accesses for network comHarp is one of the first implementations of a primary munication. Harp is intended to be used within a file sercopy scheme that runs on conventional hardware. It has vice in a distributed network; in our current implemensome novel features that allow it to perform well. The key tation, it is accessed via NFS. Preliminary performance performance issues are how to provide quick response for results indicate that Harp provides equal or better response user operations and how to provide good system capacity time and system capacity than an unreplicated implemen-(roughly, the number of operations the system can handle tation of NFS that uses Unix files directly.in some time period while still providing good response time). Harp achieves good performance by recording the

show abstract

Implicit replication in a network file server

Cited by 4 publications

References 4 publications

Replication in the Harp File System

Replication in the Harp File System

Experiences with two high availability designs (replication techniques)

Replication in the harp file system

Contact Info

Product

Resources

About