SLIDE 1
Bhaavyaa Kapoor Person # 5002 -6253 } Andrew File System } The - - PowerPoint PPT Presentation
Bhaavyaa Kapoor Person # 5002 -6253 } Andrew File System } The - - PowerPoint PPT Presentation
Bhaavyaa Kapoor Person # 5002 -6253 } Andrew File System } The Prototype } Changes for Performance } Effect of Changes for Performance } Comparison with A Remote-Open File System } Conclusion } Joined project of Carnegie
SLIDE 2
SLIDE 3
} Joined project of Carnegie Mellon University & IBM } Distributed scalable file system with homogeneous,
location-transparent file name space to all client workstations
} Motivation behind AFS: Ability to scale to serve
large no of clients without degradation in performance
} Clients and servers run the 4.2 Berkeley Software
Distribution(4.2 BSD)
SLIDE 4
Server: VICE ->Vast Integrated Computing Environment file system Client: Venus -> VIRTUE: Virtue is Reached Through Unix and Emacs
Workstation 1 Operating System
VI readme.txt VENUS
Open
Server 1
VICE
Operating System
- pen
1 3 Workstation 2 Operating System
VENUS
Read (cache)
2 2
Write (cache)
4
close
6
callback
7
Close
5
SLIDE 5
} Venus on the client with a dedicated, persistent
process
} Dedicated lock server process } Each server stored the directory hierarchy
- Mirroring the structure of the Vice files
- .admin directory – Vice file status info
- Stub directory – portions of the Vice name space on other
servers
- location database embedded in the file tree
} Vice-Venus interface named by their full pathname } Cached copies were verified with server
SLIDE 6
} The performance was significantly degraded
due to too much stat call involving cache validity check
} Network resources in the kernel frequently
exhausted due to excessive context switching
} Location Database: Difficult to move users’
directories between servers
} Unable to implement Disk quotas
SLIDE 7
} Cache management
- Cache Directories and symbolic links in addition
to files
- Callback reduces cache validation traffic
} Name resolution
- Fid: Notion of two-level names with unique fixed-
length
- 32-bit components – Volume number, Vnode
number, Uniquer
} Communication and server process structure
- Light weight process bound to a client for single
server operation
} Low level storage representation
- Access files by their inodes
SLIDE 8
} Scalability
- The data in the table below shows that AFS is19% slower
than stand-alone workstation
- Prototype is 70% slower
- Remarkable improvement in scalability
- CPU utilization increases from 8% at a load of 1 to 70% at
load of 20
SLIDE 9
} Remote Open
- Fundamental property : The data in a file are not fetched en
masse, instead the remote site potentially participates in each individual read an write operation It is called remote open since file is actually opened on the remote site rather than the local site
- Andrew under controlled conditions was compared to Sun
Microsystem's NFS
- NFS performance degrades rapidly at increasing loads
SLIDE 10
- At a load of 18, server CPU utilization saturates at 100% in NFS; in Andrew it is
38%
- Andrew’s scaling characteristics are superior to those of NFS
SLIDE 11
ñ
File Access Common name space from Different file names from
ñ
all workstations different workstations
ñ
File Location Automatic tracking by Mount points to files set by
ñ
Tracking file system processes administrators and users
ñ
and databases
ñ
Performance Client caching to reduce No local disk caching;
ñ
network load; callbacks limited cache consistency
ñ
to maintain cache consis-
ñ
tency
ñ
Andrew Average time of 210 Average time of 280
ñ
Benchmark seconds/client seconds/client
ñ
Scaling Maintains performance in Best in small to mid-size
ñ
capabilities small and very large installations
ñ
installations Security Kerberos mutual authen- Security based on
ñ
tication unencrypted user ID's
ñ ñ
AFS NFS
SLIDE 12
} AFS-1 - first prototype, used for statistics
gathering in preparation of AFS-2
} AFS-2 – thread per client, call-back for cache
consistency, local directory operations with write- through, introduction of volumes, read-only replication
} AFS-3 – commercialization of AFS by Transarc
Corp,; introduction of cells for a world wide unified name space, use of v-node interface, serious security using Kerberos, chunk-based transfer and caching
} 16 Aug 2000: IBM announces Open AFS
SLIDE 13