File models and file accessing models

File models and file accessing models

1. Prepared By:
Mehta Ishani
1300407010030
File models and file accessing
models

2. Contents
2
File models
Unstructured and structured files
Mutable and immutable files
File accessing models
Accessing remote files
Unit of data transfer

3. Unstructured files
3
Simplest model
File – unstructured sequence of data
No substructure
Contents – un interpreted sequence of bytes
Unix, MS-Dos
Modern OS used this model because sharing of files is easier
in compared with structured file model
Since file has no structure then different applications can
interpret the contents of files in many different ways.

4. Structured files
4
Rarely used
File – ordered sequence of records
Files – different types, different size and different properties
Record – smallest unit of data that can be accessed
Two categories
Files with non indexed records
Files with indexed records

5. Structured files
5
Files with non indexed records
File records is accessed by specifying it’s position within file
For ex. Fifth record from beginning, second record from end
Files with indexed records
Records have one or more key fields that can be addressed by
specifying values
File is maintained as B-tree or other suitable data structure or
hash table to locate records quickly

6. File attributes
6
Information describing that files
Has name and value
Contains information such has owner, size, access
permission, date of creation, date of last modification and
date of last access
User can read the value of any attribute but can’t change
modify it
Maintained and used by directory service because they are
subject to different access control

7. Mutable files
7
Used by most existing os
Update performed on files overwrites on old contents to
produce new contents
File is represented as a single stored sequence and that is
altered by each update operation

8. Immutable files
8
Cedar files system
File can not be modified once it has been created except to
be deleted
 file versioning approach is used, a new version of file is
created when change is made rather than updating same file
In practice storage space may be reduced by keeping only
differences rather than created whole file again
Sharing is much easier because it supports caching and
replication which eliminates the problem of keeping multiple
consistent copies

9. Immutable files
9
Suffering from two issues
Increased use of disk space
Increased disk allocation activity
CFS uses keep parameter as the no. of most current version
of file to be retained.

10. Accessing remote files
10
One of the following model is used when request to access
remote file
Remote service model
Data catching model

11. Remote service model
11
Processing of client request is performed at server’s node
Client request is delivered to server and server machine
performs on it and returns replies to client
Request and replies transferred across network as message
File server interface and communication protocol must be
designed carefully so as to minimize the overhead of
generating the messages
Every remote file access results in traffic

12. Data catching model
12
Reduced the amount of network traffic by taking advantage
of locality feature
If requested data is not present locally then copied it from
server’s node to client node and catching there
LRU is used to keep the cache size bounded
Cache Consistency problem

13. Unit of data transfer
13
Refers to fraction of file data that is transferred to and from
client as a result of single read write operation
Four data transfer models
File level transfer model
Block level transfer model
Byte level transfer model
Record level transfer model

14. File level transfer model
14
When the operation required file data, the whole file is
moved
Advantages are
Efficient because network protocol overhead is required only
once
Better scalability because it requires fewer access to file server
and reduce server load and network traffic
Disk access routines on server can be better optimized
Offers degree of resiliency to server and network failure
Drawbacks is it requires sufficient storage space
Ex are amoeba, CFS, Andrew file system

15. Block level transfer model
15
file data transfer take place in units of data blocks
A file block is contiguous portion of file and fixed in length
Advantage is does not required large storage space
Drawback is more network traffic and more network
protocol overhead
Poor performance
Ex are Sun microsystem’s NFS, Apollo domain file system

16. Byte level transfer model
16
File data transfer take place in units of bytes
Provides maximum flexibility
Difficulty in cache management

17. Record level transfer model
17
Suitable with structured files
File data transfer take place in unit of records
Ex. RSS(research storage system)

18. 18