What if? or Combining different ideas J. F. Pâris.

Slides:



Advertisements
Similar presentations
COS 461 Fall 1997 Replication u previous lectures: replication for performance u today: replication for availability and fault tolerance –availability:
Advertisements

Mendel Rosenblum and John K. Ousterhout Presented by Travis Bale 1.
The Zebra Striped Network Filesystem. Approach Increase throughput, reliability by striping file data across multiple servers Data from each client is.
Serverless Network File Systems. Network File Systems Allow sharing among independent file systems in a transparent manner Mounting a remote directory.
11-May-15CSE 542: Operating Systems1 File system trace papers The Zebra striped network file system. Hartman, J. H. and Ousterhout, J. K. SOSP '93. (ACM.
RAID Technology CS350 Computer Organization Section 2 Larkin Young Rob Deaderick Amos Painter Josh Ellis.
U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Emery Berger University of Massachusetts Amherst Operating Systems CMPSCI 377 Lecture.
Recovery 10/18/05. Implementing atomicity Note, when a transaction commits, the portion of the system implementing durability ensures the transaction’s.
Disconnected Operation in the Coda File System James J. Kistler and M. Satyanarayanan Carnegie Mellon University Presented by Deepak Mehtani.
Consistency in NFS and AFS. Network File System (NFS) Uses client caching to reduce network load Built on top of RPC Server cache: X Client A cache: XClient.
File System Implementation
File System Implementation
Other File Systems: LFS and NFS. 2 Log-Structured File Systems The trend: CPUs are faster, RAM & caches are bigger –So, a lot of reads do not require.
Copyright © Clifford Neuman - UNIVERSITY OF SOUTHERN CALIFORNIA - INFORMATION SCIENCES INSTITUTE CS582: Distributed Systems Lecture 15 - October.
Cse Feb-001 CSE 451 Section February 24, 2000 Project 3 – VM.
Distributed File System: Design Comparisons II Pei Cao Cisco Systems, Inc.
RAID Systems CS Introduction to Operating Systems.
Distributed File System: Design Comparisons II Pei Cao.
THE DESIGN AND IMPLEMENTATION OF A LOG-STRUCTURED FILE SYSTEM M. Rosenblum and J. K. Ousterhout University of California, Berkeley.
The Design and Implementation of a Log-Structured File System Presented by Carl Yao.
DESIGN AND IMPLEMENTATION OF THE SUN NETWORK FILESYSTEM R. Sandberg, D. Goldberg S. Kleinman, D. Walsh, R. Lyon Sun Microsystems.
Distributed File Systems Sarah Diesburg Operating Systems CS 3430.
Transactions and Reliability. File system components Disk management Naming Reliability  What are the reliability issues in file systems? Security.
RAID Shuli Han COSC 573 Presentation.
Distributed Systems. Interprocess Communication (IPC) Processes are either independent or cooperating – Threads provide a gray area – Cooperating processes.
CS 153 Design of Operating Systems Spring 2015 Lecture 17: Paging.
THE DESIGN AND IMPLEMENTATION OF A LOG-STRUCTURED FILE SYSTEM M. Rosenblum and J. K. Ousterhout University of California, Berkeley.
Distributed File Systems Overview  A file system is an abstract data type – an abstraction of a storage device.  A distributed file system is available.
CS 5204 (FALL 2005)1 Leases: An Efficient Fault Tolerant Mechanism for Distributed File Cache Consistency Gray and Cheriton By Farid Merchant Date: 9/21/05.
Data in the Cloud – I Parallel Databases The Google File System Parallel File Systems.
The Design and Implementation of Log-Structure File System M. Rosenblum and J. Ousterhout.
1 File Systems: Consistency Issues. 2 File Systems: Consistency Issues File systems maintains many data structures  Free list/bit vector  Directories.
MapReduce and GFS. Introduction r To understand Google’s file system let us look at the sort of processing that needs to be done r We will look at MapReduce.
Serverless Network File Systems Overview by Joseph Thompson.
Log-Structured File Systems
CS 153 Design of Operating Systems Spring 2015 Lecture 22: File system optimizations.
The concept of RAID in Databases By Junaid Ali Siddiqui.
COSC 3330/6308 Solutions to the Third Problem Set Jehan-François Pâris November 2012.
Disk Farms at Jefferson Lab Bryan Hess
ENERGY-EFFICIENCY AND STORAGE FLEXIBILITY IN THE BLUE FILE SYSTEM E. B. Nightingale and J. Flinn University of Michigan.
Silberschatz, Galvin and Gagne ©2009 Operating System Concepts – 8 th Edition File System Implementation.
Outline for Today Journaling vs. Soft Updates Administrative.
Distributed File Systems 11.2Process SaiRaj Bharath Yalamanchili.
Infrastructure for Data Warehouses. Basics Of Data Access Data Store Machine Memory Buffer Memory Cache Data Store Buffer Bus Structure.
Transactions and Reliability Andy Wang Operating Systems COP 4610 / CGS 5765.
JOURNALING VERSUS SOFT UPDATES: ASYNCHRONOUS META-DATA PROTECTION IN FILE SYSTEMS Margo I. Seltzer, Harvard Gregory R. Ganger, CMU M. Kirk McKusick Keith.
Embedded System Lab. 정영진 The Design and Implementation of a Log-Structured File System Mendel Rosenblum and John K. Ousterhout ACM Transactions.
Speculative Execution in a Distributed File System Ed Nightingale Peter Chen Jason Flinn University of Michigan Best Paper at SOSP 2005 Modified for CS739.
File System Performance CSE451 Andrew Whitaker. Ways to Improve Performance Access the disk less  Caching! Be smarter about accessing the disk  Turn.
Chapter Five Distributed file systems. 2 Contents Distributed file system design Distributed file system implementation Trends in distributed file systems.
Computer Science Lecture 19, page 1 CS677: Distributed OS Last Class: Fault tolerance Reliable communication –One-one communication –One-many communication.
RAID Technology By: Adarsha A,S 1BY08A03. Overview What is RAID Technology? What is RAID Technology? History of RAID History of RAID Techniques/Methods.
RAID TECHNOLOGY RASHMI ACHARYA CSE(A) RG NO
DISTRIBUTED FILE SYSTEM- ENHANCEMENT AND FURTHER DEVELOPMENT BY:- PALLAWI(10BIT0033)
CS Introduction to Operating Systems
Transactions and Reliability
Distributed File Systems
CSE-291 (Cloud Computing) Fall 2016
First Pop Quiz Answers COSC 6360 Fall 2017.
Solution to the Fourth COSC 6360 Quiz for Fall 2013
RAID RAID Mukesh N Tekwani
Overview Continuation from Monday (File system implementation)
A Redundant Global Storage Architecture
Printed on Monday, December 31, 2018 at 2:03 PM.
Cary G. Gray David R. Cheriton Stanford University
DESIGN AND IMPLEMENTATION OF THE SUN NETWORK FILESYSTEM
THE GOOGLE FILE SYSTEM.
RAID RAID Mukesh N Tekwani April 23, 2019
File System Performance
CSE 542: Operating Systems
Presentation transcript:

What if? or Combining different ideas J. F. Pâris

Combining RAID-5 and LFS RAID-5 allows parallel writes thus increasing I/O bandwidth has problems with small writes LFS ensures that all writes are sequential adds I/O overhead for segment cleaning

Combining RAID-5 and LFS What would be the main advantage?

Combining RAID-5 and LFS What would be the main advantage? It would greatly reduce the small write penalty assuming that we can write the log one stripe at a time Was it already tried? Yes in the Zebra file system

Adding leases to NFS How would it work? When a process does a lookup for a file it gets a lease on that file for up to a few seconds During that time it does not need to revalidate its cache contents sends its updates to the server unless it has no local disk

Adding leases to NFS What would be the main advantage of this approach?

Adding leases to NFS What would be the main advantage of this approach? A big reduction of revalidation traffic: Server could handle more clients

Adding leases to NFS Would it not make the server stateful ?

Adding leases to NFS Would it not make the server stateful ? Yes

Adding leases to NFS Would it not complicate server recovery?

Adding leases to NFS Would it not complicate server recovery? NO, because the state information would only consist of short lived leases A server recovering after a crash will never have to deal with unexpired leases

Adding leases to NFS What would you do when a lease is broken?

Adding leases to NFS What would you do when a lease is broken? It would mean that somebody else is accessing the file We could either Return to standard NFS operation with more frequent cache refreshes Disable client caching for a while

Adding leases to AFS/Coda What would be the main advantage of this approach?

Adding leases to AFS/Coda What would be the main advantage of this approach? No more lost callbacks More faithful implementation of open-to- close semantics

Adding leases to AFS/Coda What would be the main disadvantage of this approach?

Adding leases to AFS/Coda What would be the main disadvantage of this approach? Leases would have to be renewed quite frequently