International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169

Volume: 4 Issue: 6 659 - 661

______________________________________________________________________________________
Design of Video Compression Approach in Grid Environment

Makhan Singh
Computer Science and Engineering, University Institute of Engineering & Technology, Panjab University, Chandigarh, India.
[email protected]

Abstract Grid offers an optimal solution to problems requiring large storage requiring large storage and/or processing power. Grid provides
direct access to computers, data, software and many other resources. Sharing between these resources is highly controlled and done with
consensus of both resource providers and consumers. Video compression is a lengthy and compute intensive task, involving compression of
video media from one format to another. Video compression refers t reducing the quantity of data used to represent digital video images, and its
a combination of spatial image compression and temporal motion compensation. Video compression system maintains high picture quality while
reducing its quantity by removing redundancies. In Grid computing the task is split up into smaller chunks and the resources of many computers
in a network can be applied to a single problem at the same time independently. These factors make the distribution of video compression viable
in Grid environment. This paper provides a discussion of design of video compression in Grid environment, Grid environment and its globus
toolkit 4.0.

Keywords-Grid computing, video compression, spatial image compression, temporal motion compensation, standards.
__________________________________________________*****_________________________________________________

I. INTRODUCTION III. VIDEO COMPRESSION APPROCH IN GRID

Grid computing is a technology that brings together a set of The steps required for setting up a Grid environment and
resources distributed over wide-area networks that can support solving the compute intensive problem of video compression
large-scale distributed applications. Grid computing is the act using Grid environment.
of sharing tasks over multiple computers. Grid computing
applies resources of many computers in a network to a single 1. Prepare a Grid Environment.
problem at the same time. Grid computing enables the 2. Each node must be synchronized with Grid node.
virtualization of distributed computing and data resources such 3. Each node must have Mplayer and mencoder at each
as processing, network bandwidth and storage capacity to Grid node.
create a single system image, granting users and applications Our main task is to perform video compression in Grid
seamless access to vast IT capabilities. Grid coordinates environment. A true video compression system would require
resource sharing and problem solving in dynamic, multi- enormous storage facility accoutered with high-speed
institutional, virtual organizations[1]. Grid focuses on large- access and replication facilities. A centralized storage in a
scale resource sharing, innovative applications and high scenario has its own demerits and infrastructure costs. Thus,
performance orientation. The idea of Grid computing pre-splitting and distributing media content across various
originated with Ian Foster, Carl Kesselman and Steve nodes with large storage capacities is a feasible solution.
Tuecke[2]. They got together to develop a toolkit to handle Centralized environment are often hit by performance
computation management, data movement, storage bottlenecks that arise as they doesnt scale enough while
management and other infrastructure that could handle large handling parallel connection-requests and maintaining sessions.
Grids without restricting themselves to specific hardware and On the other hand distributed environment is capable of
requirements. The technique is also exceptionally flexible. [2] efficiently handling the same.
Video is an integral part of multimedia. Video compression Hence the compute intensive problems can be efficiently
is a tradeoff between disk apace, video quality, and the cost of handled on a distributing environment. Grid computing is
hardware required to decompress the video in a reasonable considered to be a highly scalable solution where compute
time. This allows a more efficient storage and transmission of intensive tasks are split up into multiple low-cost systems so
data [3]. that the overload is minimized and cost reduced. Grid
computing works extremely well in parallel scenario where
II. GRID COMPUTING GLOBUS TOOLKIT 4.0 data can be split up in multiple nodes in Grid environment.
It is a collection of open-source components. Many of Grid proved to be an excellent solution for distributed video
components are based on existing standards whereas some compression. Parameters for video compression are cost, time
components are based on evolving standards. Version 4 is the required for the compression, cost of the software and quality
first version to support web service based implementations of of the compressed video[5].
many of its components[4]. A. Challenges of Grid Design
GT4 provides components in the following five categories:- A Grid environment designed for video compression is
1. Common runtime components characterized by the efficiency, resource utilization, fault
2. Security tolerance etc. it offers to its end users. Grid computing
3. Data management provides virtualization of a single system and complete
4. Information services control over the presentation session to its end users.
5. Execution management Challenges which are needed to be address are:-
659
IJRITCC | June 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 4 Issue: 6 659 - 661
______________________________________________________________________________________
1. Security- for supporting heterogeneous Grid architecture, IV. DESIGN AND PERFORMANCE ANALYSIS
there is a need to secure the Grid resources from In our experiment we have 3 nodes, each node can perform
unauthorized access. So we need secure standard methods video compression independently.
for authentication such as applying different passwords, Node B: - CA Server.
firewall configuration, CA server configuration is Node A: - PBS Cluster node.
required. Node C:- SGE Cluster node.
2. Network Establishment- As the network is already If video compression is performed on a single node it takes a
established in the laboratories we need to make a private long time. While if used distributed Grid the time of
network on the same infrastructure laid earlier. So we compression will be reduced. To compress any video without
need to plan a network on the existing network. changing its format the task of compression has been
3. Fault tolerance and Robustness- To support fault performed and three results have been taken:-
tolerance and robustness in the system there is a need to
script replication of jobs to idle resources and facilitate Node Actual Size after Time Avg. Time
fault tolerance and robustness in the system. . Size of Compression taken on taken on
4. Cost Effectiveness- Design need to be cost effective in video in single Grid in
terms of resources and financial structure. MB System Minutes
5. Bandwidth Availability- As the network link may not in
Minutes
always be consistent, one needs to manage the bandwidth
A 897 834 58 43
corresponding to network change and still maintain
B 897 834 56 48
quality of the media.
C 897 834 64 44
In addition to the above parameters, such a setup needs to be
Average time of compression (in Grid) = (T1+T2+T3)/3
highly user friendly to ensure ease of use and user satisfaction. = (43+48+44)/3
Our Video Compression Grid system uses distributed =45 min.
technique to take this problem under consideration and Average time of doing compression on a single node
develop an efficient solution. = (T1+T2+T3)/3
B. Different steps involved in Video Compression = (58+56+64)/3
=59.33 min
1. Test that each node of the Grid is properly connected Thus on average 14.33 minutes are saved in video
with each other and synchronized with CA server. compression.
2. Check that each node has required software.
3. First task is to select a video. Now using Mencoder we
will first split the video and with the help of python
the splitted video would be distributed.
4. Now further using mencoder compression will
performed and after the compression the performed,
each node will send the compressed video to the job
submission node. It will recombined into one video.
5. To test the quality of video we use mplayer to play
and test[3].
C. Challenges of Video Compression
A video compression system basically characterized by
flexibility it offers to its users. Video compression service must
provide its local or remote users complete control over the
presentation session[3].
Challenges which are needed to be address are:-

1. Load distribution on server- To support multiple

connection requests from user, and facilitate minimum Figure 1. PLOTTING OF ORIGINAL TIME VERSUS COMPRESSION
response time. TIME
2. Media content management- It consists high storage
space, effective content management, replication After conducting around 12 runs as above the following
strategy etc. factors are affecting performance of compression:-
3. Rate control- For adapting to network, the system may 1) Size, video format, output format.
need to vary the transport and encoding rates. 2) Bandwidth.
4. Scalability and cost effectiveness- Design need to be 3) Grid Load, Grid Scheduling.
easily scalable and cost effective in terms of resources 4) Nodes local processing and availability.
and financial structure.
V. CONCLUSION
Grid computing is an efficient tool for sharing tasks over
multiple computers. Tasks can range from data storage to
complex calculations. Hence video compression can be done
660
IJRITCC | June 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 4 Issue: 6 659 - 661
______________________________________________________________________________________
more efficiently using Grid computing than on single system.
Resources can be better utilized and time complexity of a video
compression is reduced by distributing it on number of clients
using Grid computing. Hence its an ideal approach for
computing large computational tasks through Grid computing.
REFERENCES
[1] Jacob, Bart, et al. "Introduction to Grid computing." IBM
redbooks (2005).
[2] Foster, Ian, Carl Kesselman, and Steven Tuecke. "The anatomy
of the Grid: Enabling scalable virtual organizations."
International journal of high performance computing
applications 15.3 (2001): 200-222.
[3] http://www.mit.edu/~6.344/Spring2004/video_compression_200
4.pdf
[4] Foster, I. and Kesselman, C. Globus: A Toolkit-Based Grid
Architecture. Foster, I. and Kesselman, C. eds. The Grid:
Blueprint for a New Computing Infrastructure, Morgan
Kaufmann, 1999, 259-278.
[5] Talluri, Raj, et al. "A robust, scalable, object-based video
compression technique for very low bit-rate coding." IEEE
Transactions on Circuits and Systems for Video Technology 7.1
(1997): 221-233.

661
IJRITCC | June 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________