distributed file system

1
Distributed file system

• Submitted by
• Neha gahlot

2
INTRODUCTION

• A distributed file system is a resource
  management component.
• It implements a common file system that can be
  shared by various computers in system.
• There are two important goals of file system as
  follows
• Network transparency
• High availability

3
ARCHITECTURE

• In DFS files can be store at any machine and the
  computation can be performed at any machine.
• When m/c needs to acess file stored on remote m/c
  , the remote m/c performs the necessary
  operation and return data.
• Two most important services present in DFS
• Name server
• Cache server

4

• Name server
• A process that maps names specified by clients to
  stored objects such as files and directories.
• Cache manager
• A cache manger is a process that implements file
  caching.
• In file caching copy of data stored at remote
  server is brought to client m/c when referenced.
• Cache managers can be present at both clients and
  file servers.
• Cache managers at the servers cache files in the
  main memory to reduce delays due to disk latency.
• If multiple clients are allowed to cache a file
  and modify it, the copies can become inconsistent.

5
Mechanism for building DFS

1. Mounting
2. Caching
3. Hints
4. Bulk data transfer
5. Encryption

6

• Mounting
• A mount mechanism allows the binding together of
  different filename spaces to form a single
  hierarchically structure name space.
• A name space (or a collection of files) can be
  bounded to or mounted at an internal node or a
  leaf node of a name space tree.
• A node onto which name space is mounted is called
  mount points.
• The kernel maintains a structure called the mount
  table, which maps mount points to appropriate
  storage devices.
• There are two approaches to maintain the mount
  information.
• Mount information can be maintained at clients. ?
  SUN NFS
• Mount information can be maintained at servers. ?
  Sprite file system

7

• Caching
• Caching is commonly employed in DFS to reduce
  delays in accessing of data.
• Caching exploits the temporal locality of
  reference exhibited by programs.
• Temporal locality of reference refer to the fact
  that file recently acessed is likely to be
  acessed again in near future.
• Hints
• An alternative approach is to treat the cached
  data as hints.
• The class of applications that can utilize hints
  are those which can recover after discovering
  that the cached data are invalid.

8

• Bulk Data Transfer
• Transferring data in bulk reduces the protocol
  processing overhead at both servers and clients.
• Bulk transfer amortizes the protocol processing
  overhead and disk seek time over many consecutive
  blocks of a file.
• Bulk transfers reduce file access overhead
  through obtaining a multiple number of blocks
  with a single seek by formatting and
  transmitting a multiple number of large packets
  in a single context switch and
• by reducing the number of acknowledgements that
  need to be met.

9

• Encryption
• Encryption is used for enforcing security in
  distributed system.
• In this two parties wish to communicate establish
  a key for coversation wid the help of
  authentication server.
• And this key never sent in plain text.