Jinyong Yoon, 2010. 10. 18..  Andrew File System  The Prototype  Changes for Performance  Effect of Changes for Performance  Comparison with A Remote-Open.

Slides:



Advertisements
Similar presentations
OSes: 15. Distributed File Systems 1 Operating Systems v Objectives –introduce issues such as naming, stateful and stateless, and replication Certificate.
Advertisements

Silberschatz, Galvin and Gagne  Operating System Concepts Chapter 16 Distributed-File Systems Background Naming and Transparency Remote File.
Silberschatz, Galvin and Gagne ©2009 Operating System Concepts – 8 th Edition, Lecture 23: Distributed-File Systems (Chapter 17)
Module 17: Distributed-File Systems Background Naming and Transparency Remote File Access Stateful versus Stateless Service File Replication Example Systems.
Silberschatz, Galvin and Gagne ©2009 Operating System Concepts – 8 th Edition, Chapter 17: Distributed-File Systems.
Chapter 17: Distributed-File Systems Silberschatz, Galvin and Gagne ©2005 AE4B33OSS Chapter 17 Distributed-File Systems Background Naming and Transparency.
CS-550: Distributed File Systems [SiS]1 Resource Management in Distributed Systems: Distributed File Systems.
Chapter 17: Distributed-File Systems Adapted to COP4610 by Robert van Engelen.
Andrew File System (AFS)
Distributed File Systems CS 3100 Distributed File Systems1.
G Robert Grimm New York University Disconnected Operation in the Coda File System.
Copyright © Clifford Neuman - UNIVERSITY OF SOUTHERN CALIFORNIA - INFORMATION SCIENCES INSTITUTE CS582: Distributed Systems Lecture 13, 14 -
Distributed File Systems
Coda file system: Disconnected operation By Wallis Chau May 7, 2003.
Other File Systems: AFS, Napster. 2 Recap NFS: –Server exposes one or more directories Client accesses them by mounting the directories –Stateless server.
Chapter 17: Distributed-File Systems Part Silberschatz, Galvin and Gagne ©2005 Operating System Concepts Chapter 17 Distributed-File Systems Chapter.
Distributed File Systems
Disconnected Operation In The Coda File System James J Kistler & M Satyanarayanan Carnegie Mellon University Presented By Prashanth L Anmol N M Yulong.
Chapter 17 Distributed File Systems By: Amar Deo( ) Ankur Patel( )
G Robert Grimm New York University Scale and Performance in Distributed File Systems: AFS and SpriteFS.
Caching. Andrew Security Andrew Scale and Performance Sprite Performance.
Jeff Chheng Jun Du.  Distributed file system  Designed for scalability, security, and high availability  Descendant of version 2 of Andrew File System.
PRASHANTHI NARAYAN NETTEM.
Distributed-File Systems
NFS. The Sun Network File System (NFS) An implementation and a specification of a software system for accessing remote files across LANs. The implementation.
University of Pennsylvania 11/21/00CSE 3801 Distributed File Systems CSE 380 Lecture Note 14 Insup Lee.
Lecture 23 The Andrew File System. NFS Architecture client File Server Local FS RPC.
Distributed File Systems Concepts & Overview. Goals and Criteria Goal: present to a user a coherent, efficient, and manageable system for long-term data.
Distributed File Systems 1 CS502 Spring 2006 Distributed Files Systems CS-502 Operating Systems Spring 2006.
Networked File System CS Introduction to Operating Systems.
Distributed Systems. Interprocess Communication (IPC) Processes are either independent or cooperating – Threads provide a gray area – Cooperating processes.
Advanced Operating Systems - Spring 2009 Lecture 21 – Monday April 6 st, 2009 Dan C. Marinescu Office: HEC 439 B. Office.
Distributed File Systems
1 10/2/2015 Chapter 16 Distributed-File Systems 此章不作上课内容 l Background l Naming and Transparency l Remote File Access l Stateful versus Stateless Service.
Distributed File Systems Case Studies: Sprite Coda.
Distributed File Systems Distributed file system (DFS) – a distributed implementation of the classical time-sharing model of a file system, where multiple.
Distributed file systems, Case studies n Sun’s NFS u history u virtual file system and mounting u NFS protocol u caching in NFS u V3 n Andrew File System.
Chapter 20 Distributed File Systems Copyright © 2008.
What is a Distributed File System?? Allows transparent access to remote files over a network. Examples: Network File System (NFS) by Sun Microsystems.
Introduction to DFS. Distributed File Systems A file system whose clients, servers and storage devices are dispersed among the machines of a distributed.
Dr. M. Munlin Network and Distributed System Structures 1 NETE0516 Operating Systems Instructor: ผ. ศ. ดร. หมัดอามีน หมัน หลิน Faculty of Information.
Mobile File System Byung Chul Tak. AFS  Andrew File System Distributed computing environment developed at CMU provides transparent access to remote shared.
Presented By: Samreen Tahir Coda is a network file system and a descendent of the Andrew File System 2. It was designed to be: Highly Highly secure Available.
A Low-bandwidth Network File System Athicha Muthitacharoen et al. Presented by Matt Miller September 12, 2002.
Caching in the Sprite Network File System Scale and Performance in a Distributed File System COMP 520 September 21, 2004.
Information Management NTU Distributed File Systems.
GLOBAL EDGE SOFTWERE LTD1 R EMOTE F ILE S HARING - Ardhanareesh Aradhyamath.
Silberschatz, Galvin and Gagne  Operating System Concepts Distributed File Systems Distributed file system (DFS) – a distributed implementation.
Lecture 25 The Andrew File System. NFS Architecture client File Server Local FS RPC.
Distributed File Systems Questions answered in this lecture: Why are distributed file systems useful? What is difficult about distributed file systems?
Silberschatz, Galvin and Gagne  Operating System Concepts Distributed File Systems Distributed file system (DFS) – a distributed implementation.
Silberschatz, Galvin and Gagne  Operating System Concepts Chapter 16 Distributed-File Systems Background Naming and Transparency Remote File.
Computer Science Lecture 19, page 1 CS677: Distributed OS Last Class: Fault tolerance Reliable communication –One-one communication –One-many communication.
Chapter 17: Distributed-File Systems
Andrew File System (AFS)
Scale and Performance in a Distributed File System
NFS and AFS Adapted from slides by Ed Lazowska, Hank Levy, Andrea and Remzi Arpaci-Dussea, Michael Swift.
Chapter 17: Distributed-File Systems
CSE 451: Operating Systems Winter Module 22 Distributed File Systems
Scale and Performance in a Distributed File System
Distributed File Systems
DISTRIBUTED FILE SYSTEMS
Distributed File Systems
CSE 451: Operating Systems Spring Module 21 Distributed File Systems
Distributed File Systems
Chapter 15: File System Internals
Outline Review of Quiz #1 Distributed File Systems 4/20/2019 COP5611.
Distributed File Systems
Chapter 17: Distributed-File Systems
Distributed File Systems
Presentation transcript:

Jinyong Yoon,

 Andrew File System  The Prototype  Changes for Performance  Effect of Changes for Performance  Comparison with A Remote-Open File System  Conclusion

 Developed at Carnegie Mellon University  Distributed file system by considerations of scale  Locality of file references  Present a homogeneous, location-transparent file name space to all the client workstations  Use 4.2 BSD  Server ▪ A set of trusted servers – Vice  Clients ▪ User level processes – Venus ▪ File system call hooking ▪ Contacts with servers only opens and closes for a whole-file transfer ▪ Caches files from Vice ▪ Store modified copies of files back on the servers

workstation Venus User Program Unix Kernel Disk Server Vice Unix Kernel Disk workstation Venus User Program Unix Kernel Disk workstation Venus User Program Unix Kernel Disk Server Vice Unix Kernel Disk Network

 Venus on the client with a dedicated process  Persistent process on the server  Each server stored the directory hierarchy  Mirroring the structure of the Vice files .admin directory – Vice file status info  Stub directory – location database  Vice-Venus interface by their full pathname  There’s no notion of a low-level name such as inode  Before using a cached file, Venus verifies its timestamp  Each open of a file thus resulted in at least one interaction with a server, even if the file were already in the cache and up to date

 stat primitive  To test for the presence of files  To obtain status information before opening files  Each stat call involved a cache validity check  Increase total running time and the load on servers  Dedicated Process  Excessive context switching overhead  Critical resource limits excess  High virtual memory paging demands

 Remote Procedure Call (RPC)  Simplification of implementation  Network related resources in the kernel to be exceeded  Location Database  Difficult to move users’ directories between servers  Etc.  Use Vice file without recompilation or relinking

 Benchmark  Command scripts that operates on a collection of files  70 files (source code of an application program)  200kb  Stand-alone Benchmark and 5 phases

 Skewed distribution of Vice calls  TestAuth – Validate cache entries  GetFileStat – Obtain status information about files absent from the cache

 Load unit  Load placed on a server by a single client workstation running this benchmark  A load unit – 5 Andrew users

 CPU/disk utilization profiling  Performance bottleneck is CPU  Frequently context switches  The time spent by the servers in traversing full pathnames

 Cache management  Previous ▪ Status(in virtual memory)/Data(in local disk) cache ▪ Interception only opening/closing operations ▪ Modifications to a cached files are reflected back to Vice when the file is closed  Callback - the server promises to notify it before allowing a modification ▪ This reduces cache validation traffic ▪ Each should maintain callback state information ▪ There is a potential for inconsistency

 Name resolution  Previous ▪ inode – unique, fixed-length ▪ pathname – one or more, variable-length ▪ namei routine – maps a pathname to an inode ▪ Each Vice pathname involves implicit namei operation ▪ CPU overhead on the servers  fid – unique, fixed-length, two-level name ▪ Map a component of a pathname to a fid ▪ Each 32 bit-Volume number, Vnode number, Uniquifuier ▪ Volume number: Identifying a Volume on one server ▪ Vnode number: Index into an file storage information array ▪ Uniquifuier: Allowing Reuse of Vnode number

 Communication and server process structure  Using Lightweight Processes (LWPs) instead of a single process  An LWP is bound to a particular client only for the duration of a single server operation.  Low-level storage representation  Access files by their inodes ▪ vnode on the servers ▪ inode on the clients

workstation User Program Unix Kernel Unix File System Unix file system calls -If D is in the cache and has a callback on it -If D is in the cache but has no callback on it -If D is not in the cache Non-local file operations Local Disk

 Scalability  19% slower than stand-alone workstation  Prototype is 70% slower

 Scalability

 Remote Open  The data in a file are not fetched en masse  Instead the remote site potentially participates in each individual read an write operation  File is actually opened on the remote site rather than the local site  NFS

 Advantage of remote-open file system  Low latency

 Scale impacts Andrew in areas besides performance and operability