IDENTIFICATION OF EFFICIENT PEERS IN P2P COMPUTING SYSTEM FOR REAL TIME APPLICATIONS

International Journal of Peer to Peer Networks (IJP2P) Vol.3, No 6, November 2012
DOI : 10.5121/ijp2p.2012.3601 1
IDENTIFICATION OF EFFICIENT PEERS IN P2P
COMPUTING SYSTEM FOR REAL TIME
APPLICATIONS
Jigyasu Dubey1
and Vrinda Tokekar2
1
Department of Information Technology, Shri Vaishnav Institute of Technology &
Science, Indore, India
jigyasudube@yahoo.co.in
2
Institute of Engineering & Technology, Devi Ahilya Vishwavidyalaya, Indore, India
vrindatokekar@yahoo.com
ABSTRACT
Currently the Peer-to-Peer computing paradigm rises as an economic solution for the large scale
computation problems. However due to the dynamic nature of peers it is very difficult to use this type of
systems for the computations of real time applications. Strict deadline of scientific and real time
applications require predictable performance in such applications. We propose an algorithm to identify the
group of reliable peers, from the available peers on the Internet, for the processing of real time
application’s tasks. The algorithm is based on joint evaluation of peer properties like peer availability,
credibility, computation time and the turnaround time of the peer with respect to the task distributor peer.
Here we also define a method to calculate turnaround time (distance) on task distributor peers at
application level.
KEYWORDS
Peer, P2P Networks, P2P Computing, Real Time Application
1. INTRODUCTION
The Peer-to-Peer (P2P) distributed computing systems have attracted more and more research
efforts recently. Currently several P2P networks have been developed for various types of
applications such as distributed video encoding [1], file sharing [2][3], and media streaming [4].
When a P2P system is used for distributed computing great processing power can be achieved.
Applications such as distributed.net[5], and SETI[6][7] uses the idle CPU cycles of thousands of
computers connected to the Internet in order to break encryption codes and find signs of
intelligent life in outer space. These systems are not decentralized systems, because they have just
one single host distributing workload, whereas all other peers operate as workers. However, these
applications have shown that the combined computational power of even a few hundred peers is
often the only way to efficiently and economically solve very expensive computation task. To
ensure high throughput and good utilization of resources it is required to have proper resource
management. Due to the ad-hoc and dynamic nature of the P2P network paradigm, varying
resource availability and unpredictable latencies are present which causes number of challenges
in managing the computing resources and scheduling task execution across the systems [8].
Moreover the tasks in real time applications have deadline to be met which requires the
predictable performance of the computing system [9]. The P2P computing systems have lack of

2
scheduling schemes which clearly analyze the resource requirement and predict total execution
time. Also they lack of mechanism for systematically recruiting the resources (peers) available in
the system. To recruit the reliable peers some earlier systems have focused on identifying good
peers [10], [11] based on the capability and reliability of peers. However they are not considering
the communication overhead presents in P2P network which is major cause behind the time
delay. It is very difficult to obtain accurate distance metrics for a dynamic, decentralized P2P
networks. The number of peers involved in such systems is very huge; hence usual techniques
such as ping and traceroute are not useful. The factor turnaround time reflects jointly processing
capability of peer as well as communication overheads involved. In this paper we define a task
unit and measure the distance between task distributor peer and task processors peers at the
application level in the form of turnaround time of a task unit. We proposed a method to identify
group of efficient peers in P2P computing System, to process real time applications tasks. The
method identifies the reliable peers from existing peers, for processing of real time application
tasks, on the basis of peer properties like peer credibility, peer computation time, and turnaround
time. The peer credibility is the probability that the result produced by a peer is correct. Peer
computation time represent the time when a peer actually executes the system’s computations in
the presence of peer autonomy failures.
2. RELATED WORK
P2P systems are different in both technological aspects and design/implementation issues.
Recently, considerable research effort is being done on several important issues related P2P
systems. In P2P computing networks, to minimize total processing time, Jingnan Yao and Laxmi
Bhuyan in [12] designed a distributed packet processing algorithm known as resource sharing
distributed packet processing algorithm known as resource sharing distributed packet processing
algorithm (RSDLP). In RSDLP algorithm the workload is distributed among peers by organizing
them into an efficient resource tree. Jingnan Yao, Jiam Zhou, et. al. in [9] proposed an load
sharing mechanism for computing real time jobs in P2P networks. The proposed mechanism
identifies most efficient resource pool with an optimized load scheduling. Author focus to explore
maximum network utilization by building a resource tree of maximum efficiency (MET).
Selection of peers is based on a combined evaluation of the available computation power and
communication bandwidth. However, for peer selection, peer credibility is also required to be
considered. Javier Celaya and Unai Arronategui in [13] proposed a new scalable scheduler for
workflows with deadlines in a P2P desktop grid. The scheduler is built upon a tree-based network
overlay with a distributed management of availability time intervals of resources. In proposed
scheduler, authors only consider the availability time intervals of resources, however for the
deadline driven tasks other parameters like resource credibility, actual participation time, and
communication bandwidth also need to be considered. Virginia Lo, et al. in [14] proposed a
system named cluster computing on the fly (CCOF) which harvest the CPU cycles from ordinary
users (Desktop PCs). They also proposed a wave scheduler which exploits the large blocks of
ideal time at night, to provide higher quality of service for deadline-driven jobs, using a
geographic based overlay to organize hosts by time zone. In this wave scheduler they explore the
possibilities to capture the CPU cycles from number of machines that lie completely idle at night.
It provides a higher guarantee of ongoing available cycles hence it is useful for deadline driven
tasks. The system provides the higher computation performance but due to using the peers from
same night time zone which belongs the same geographic location the reliability of the system
decreases. Dimitris Kamenopoulos, Iosif A Osman, et.al. in [15] presents a simple distance
measurement method in volunteer computing networks. The method is based on passive
monitoring of application level traffic. For the purpose of load balancing authors suggest that the
distances between peers in a volunteer computing network need to be considered. In [16] authors
proposed an algorithm to identify the reliable peer groups in P2P computing systems by using the

3
peer properties like peer availability, credibility and computation time. However for the real time
applications peer distance also be considered.
To identify the reliable peers for in P2P computing systems some earlier systems have focused on
good peers [10], [11] based on capability and reliability of peers where as some have focused on
distance value between peers. The real time and scientific applications also demand accurate
results. For such applications, in P2P computing systems it is required that peer selection is based
on peer credibility along with the above discussed peer properties.
3. P2P COMPUTING SYSTEMS - TOPOLOGY & CHARACTERISTICS
The P2P computing systems utilize the processing power of idle Desktop PCs presented at the
edge of the Internet as shown in figure 1. The P2P computing systems support only those
computations which can be divided into small tasks and which are embarrassingly parallel in
nature. The P2P computing systems consist of two major components- Task Distributor and Task
Processor [17]. A peer in a P2P computing system at a time can be a task distributor or a task
processor, but not both. A user on a peer provides the job in form of computation code with the
data which is to be processed. This peer becomes as a task distributor in the system and other
peers as task processors. The task which is provided by the user for processing must be able to
split in subtasks. The task distributor peer is responsible for splitting the task in to number of
small tasks and distributes these subtasks to the task processor peers in the system for processing.
The task distributor is also responsible for integrating the results which it receives from the
different task processors after the completion of task. The task processor peers receive the
computation code and data from the task distributor and return the results back to the task
distributor after processing the task.
Figure 1. P2P Computing System Model
The P2P computing systems act like virtual super computer with varying processing power by
utilizing the ideal CPU cycles of the desktop PCs connected to the Internet [18]. The number of
peers in a P2P computing system may range from less than 10 to hundreds of thousands. The pure
P2P computing system allows any peer to act as a task distributor or a task processor. For the
successful and efficient working of system, it is required that in a P2P computing system very few
peers act as task distributor, whereas the majority of the peers act as task processors. It is
expected and required that most of the peers in P2P computing system will take the responsibility
for processing the task, otherwise the computation take very long time. It is an important
characteristic of P2P computing system because it implies that the topology of P2P computing
system is primarily a forest like structure, where one-to-many type of communication occurs.

4
The figure1 describes an abstract model of P2P computing system which consists of
heterogeneous peers interconnected via heterogeneous communication links. Each peer in the
diagram is capable of task processing. In P2P computing system peers can have different
processing power and they are interconnected with each other in an arbitrary fashion via links
with different bandwidths. Each peer has its own local task processing in the background. The
above discussed conditions lead towards delay in task completion time, which is intolerable for
real time applications. It also indispensable that result produced by each peer is correct. Due to
the dynamic nature of peers and available network latencies it is very difficult to use this type of
systems for computation of real time applications. The scientific and real time applications
require predictable performance because tasks in such applications have dead line to be met.
In this paper we proposed an algorithm to identify group of efficient and reliable peers from the
existing peers in P2P computing systems, for processing of real time application tasks. The
proposed algorithm is based on joint evaluation of peer properties: peer credibility, computation
time, and distance of the peer.
3.1. Peer Credibility
The peer credibility CP in [13] given as the probability that the result produced by a peer is
correct.
CP = CR / (ER + CR + IR) (1)
Here, ER represents the number of erroneous results, CR represents the number of correct results,
and IR represents the number of incomplete results. The term ER + CR + IR represents the number
of total tasks that a peer computes.
3.2. Peer Computation Time
The peer computation time (PCT) in [13] given as the expected Computation time when a peer
processes the system’s computations during IT.
PCT = IT X AP (2)
Where IT is peers ideal time and AP is peer availability. The Peer computation time represent the
time when a peer actually executes the system’s computations in the presence of peer autonomy
failures.
3.3. Distance of Peers
The peers in P2P computing system works in a completely passive manner hence need no
knowledge of distance information. For the purpose of load balancing, measuring any kind of
distance between peers is unnecessary, because a peer becomes a task distributor only when
actual workload occurs. The only parameter that needs to be considered is the distance between
task distributor peer and corresponding task processor peers.
The P2P computing system is designed for those applications that have large life time. A real time
task Ti is characterized by its arrival time ωi, execution deadline ωd, and size STi. A job is
considered successful if it can be executed before its deadline. In P2P computing system for the
same application, individual tasks are usually approximately equal in size and need equal
computational requirements. When a peer decides to become a task distributor and divide its
workload in to number of smaller tasks say t1, t2, t3, t4, ….,tn-1, tn , the task size has lower limit,
beyond which further division of a application makes no sense. In P2P computing system we call
this smallest possible task as “task unit”. In P2P computing system a task unit has following
properties-

5
 Any task assigned from a task distributor to a task processor consists of a positive integer
number of task units. For a same application total number of task unit is represented by n.
 We represent the size of task unit by ST. In P2P computing system all the task units have
equal size for a same application, i.e. ST1 = ST2 = ST3 = ST4 =………....= STn-1 = STn.
 The number of task units (n) for a given application is much larger than the number of
task processors (Tp) involved, i.e. n > Number of Tp.
From the above discussion we found that the size of task unit (ST) in P2P computing system is
constant and a static property. The distance between the task distributor and task processor is
dynamic and to minimize processing time we take the advantage of this parameter. The distance
D (TD, TP), where TD is task distributor peer and TP is a task processor peer, can be defined as the
time interval between the instance where TD dispatch a task unit to TP and the time the result of
this task unit is returned to TD. The distance D (TD, TP) is the turnaround time for a task unit. The
value of turnaround time D (TD, TP) for a task unit highly depends on parameters like network
bandwidth, task processor load and task processor peer’s processing capability. Suppose task
distributor peer (TD) dispatches a task unit tid to a task processor (TP) at time TS and receives result
of tid at time TC, than turnaround time D (TD, TP) is given as:
D (TD, TP) = TC - TS (3)
For simplicity we assume that TD dispatches single task unit at a time to the TP. But it may be the
case that TD dispatches two or more task unit at a time to TP. In this case, suppose TD dispatch n
task units to TP and received the cumulative result in time D (TD, TP), the measured turnaround
time for a task unit is D (TD, TP) / n. In P2P computing system a task distributor peer requires to
keep a set of touples (Task processor ID, turnaround time) for the load balancing related
decisions. There is no need to do anything on task processors, and also task distributor nodes does
not need to exchange distance information between them.
Figure 2. Task Distributor Model

6
On a conceptual level, each task distributor peer needs to maintain only a single lookup table
named Turnaround-Time table as given in figure 2. The Turnaround-Time table contains task
processor ids as key and the value of estimated turnaround time. Whenever a task processor peer
returns a result, the corresponding distance value in the lookup table updated accordingly. To
measure the task unit turnaround time, the task distributor peer must require to record the time
when respective task is dispatched. In order to address this problem the task distributor peers
maintain a second lookup table named TaskID table (shown in figure 2) containing the task ids,
corresponding task start time and the task processor id to which task is dispatched. Whenever a
result is returned, its task id is matched against the respective entry in the lookup table and the
turnaround time is calculated by subtracting the time values stored there from the current time.
The following algorithm calculates the turnaround time of peers on the task distributor peer.
Algorithm for Distance Measurement on Task Distributor Peer
// Turnaround time between TD & TP = D (TD, TP)
// Task Processor ID = TPid
// Task ID = tid
// Start time of tid = TS
// End time of tid = TC
// Number of subtasks represented by tid = n
1. Start
2. Launch the peer into JXTA network;
3. if (task is available for processing)
4. Divide the task into multiple subtasks;
5. Create input pipe advertisement;
6. Build input pipe by using the pipe advertisement;
7. Broadcast a pipe advertisement to announce the task availability;
8. while (subtask is available for processing)
9. Wait for response;
10. if (response == result)
11. Extract TS from TaskID table corresponding to tid;
12. D (TD, TP) = TC – TS;
13. if (tid represents more than one task)
14. D (TD, TP) = D (TD, TP) / n;
15. if (TPid is not available in turnaround-time table)
16. Insert TPid and corresponding D(TD, TP) in turnaround-time table;
17. else
18. Update the value of D(TD, TP) in turnaround-time table;
19. end else
20. Store the result;

7
21. end if
22. else
23. Create output pipe;
24. In TaskID table enter values of tid, TS and TPid;
25. Send the subtask and data to the peer which responds via output pipe;
26. end else
27. end while
28. if (no more results awaited)
29. Integrate all the results of subtasks;
30. end if
31. Go to step 3;
32. end if
33. end
4. IDENTIFICATION OF GROUP OF RELIABLE PEERS
A P2P computing system is based on desktop PCs which are connected to the Internet. In such
systems it is assumed that each peer in addition of processing its local workload, has spare
computing cycles to share with other peers. The real time application task that required
computational resources, can initiate from any peer at any time with varying resource
requirements and time deadlines. The spare computing power available on each peer varies and
its availability time differs. The peers in the system can freely join and leave the system, in
between the computation, without any constraints. Such situations are known as peer autonomy
failure [13]. It leads to the delay and suspension of computations which are not acceptable for real
time application tasks. To avoid such conditions it is required to identify group of most effective
peers among the vast number of peers on the network, which process the real time application
tasks. To identify such group, choice of peers not only depends on the absolute processing power
that is available on a peer but also on the communication cost used to distribute the task and
reception of task results. When a high processing peer is connected via an extremely slow link or
when a fast link connects to a peer that is hardly available, it is required not to use such peers in
processing of real time application tasks. Thus, it may not be very helpful for real time
application task, to blindly include those peers that have either maximum processing power or
minimum communication bandwidth, in P2P computing system. Our aim is to identify the most
effective peers among the lots of peers available on the Internet, for the successful execution of
real time application task. In this section we proposed an algorithm to identify the group of most
efficient peers from the existing peers in P2P computing system.
From the above discussion, it is clear that performance of a P2P computing system is strongly
dependent on task execution time (Processing power of peer) as well as communication cost. The
successful execution of task on a peer is strongly relying on peer properties like –peer
computation time (PCT), peer credibility (CP), and distance between peers. The CP is probability
that result produced by a peer is correct. PCT represent the time when a peer actually execute
system’s computations in presence of peer autonomy failure. In our proposed algorithm, we use
above mentioned peer properties along with the turnaround time D (TD, TP), to identify the group

8
of most efficient peers. Suppose that total M number of peers available in the system. The average
peer credibility CV is given as:
Cav = (P1.CP + P2.CP +….. + PN.CP) / M (1)
The average peer computation PCV is given as:
PCav = (P1.PCT + P2.PCT + ……+ PN.PCT) / M (2)
The average turnaround time DV is given by:
Dav = {P1.D (TD, TP) + P2.D (TD, TP) + …..+ PN.D (TD, TP)} / M (3)
4.1 Protocol to Identify Most Efficient Peers
• A task distributor peer TD, which has real time task for processing sends task availability
message to all the peers connected to the system.
• A task processor peer TP, upon receiving a task availability message, checks its
availability, and in response sends a task request to TD.
• Among all the peers responded only those peers will be selected in group of efficient
peers if all the following conditions should be true for that peer:
 The value of Peer Credibility CP must be one or more than required threshold
value.
 The value of peer computation time PCT must be greater than or equal to the
average peer computation PCav.
 The value of turnaround time D (TD, TP) must be laser than or equal to the
average turnaround time Dav.
Those peers in the P2P computing system which are not satisfying the above conditions are not
allowed to process the real time applications task.
Algorithm of Peer Group Identification
// G1 : Group 1, G2 : Group 2, G3 : Group 3, G4 : Group 4
// PG1 : Peer Group 1, PG2 : Peer Group 2, PG3 : Peer Group 3, PG4 : Peer Group 4
// TP.CV : CV of TP
// → : become a member of.
1. Start
2. If (TP ε G1) then
3. If (TP. CV ≥ µ) then
4. TP → PG1;
5. Else
6. TP → PG3;
7. Else if (TP ε G2) then

9
8. If (TP. CV ≥ µ) then
9. TP → PG2;
10. Else
11. TP → PG4;
12. Else if (TP ε G3) then
13. TP → PG3;
14. Else
15. TP → PG4;
16. End
Figure 3. Peer Group Construction
The task processor peers are classified in different peer groups by the algorithm of peer group
identification as given above. First, the responded task processor peers are classified into G1, G2,
G3, and G4 classes depending on peer computation time PCT and peer distance D (TD, TP) as
shown in figure 3. Then, the classified task processor peers are classified into each peer group
according to the peer credibility CV. By the algorithm, peer groups are classified into four
categories PG1, PG2, PG3, and PG4 as shown in fig. here Dav is the average of distances and µ is
the desired credibility threshold.
 The peer group ‘PG1’ represent a peer group in which all the peers have low values of
D(TD, TP) and high values of PCT and CV. In group PG1 all the peers have high possibility
to successfully complete the task within the given deadline reliably because all the peers
have CP greater than the required threshold value of CP.
 The peer group ‘PG2’ has high values of PCT, CV, and D(TD, TP). It has a high probability
to produce correct results. However, it cannot complete the tasks within given deadline
because distance between peers is high.
 The peer group ‘PG3’ has high probability to complete the task within given deadline
because it has low values for D (TD, TP), but it has less probability to produce correct
results because the less values of CV and PCT.

10
 The peer group ‘PG4’ has high D (TD, TP) and less value for CV and PCT. It has low
probability to complete task within given deadline as well as produce correct results.
5. CONCLUSIONS
In this paper first we proposed a method to calculate the turnaround time of a task unit in P2P
computing system at application level. We also give the algorithm for the proposed method. The
values of turnaround time gives the overall impact of network conditions because these values are
highly dependent on parameters like network bandwidth, task processor load and task processor
peer’s processing capability. The proposed method causes zero network traffic overhead, since it
uses normal application traffic to measure values of turnaround time. For each measurement the
task distributor require three indexed accesses to the lookup tables. However the proposed
method does cause a slight memory overhead on task distributor peer. The task distributor peer
must require storing a touple (task processor id, turnaround time) for every collaborated task
processor peer.
The proposed method is capable of identifying a reliable group of most efficient peers on which
execution of real time application task take placed. The peers are selected on the joint evaluation
of peer computation time (PCT), peer credibility (CP), and turnaround time D (TD, TP). The
algorithm classifies peers in to four peer groups. The peer group ‘PG1’ is best suited to carry out
the real time application tasks because peers in this group has the more credibility than the
desired threshold value and also the distance D (TD, TP) is also less.
REFERENCES
[1] V. Bharadwaj, D. Ghose, V. Mani, and T. G. Robertazzi, “Scheduling Divisible Loadas in Parallel
and Distributed Systems”, IEEE Computer Society Press, Los Almitos, California, 1996.
[2] S. Androutsellis-Theotokis and D. Spinellis, “A Survey of Peer-to-Peer File Sharing Technologies,”
White Paper, Electronic Trading Research Unit (ELTRUN), Athens University of Economics and
Business, 2002.
[3] Stefan Saroiu, P. Krishna Gummadi, and Steven D. Gribble, “A Measurement Study of Peer-to-Peer
File Sharing Systems,” In Proceedings of ACM/SPIE Multimedia Computing and Networking
(MMCN ’02), 2002.
[4] F. Chen, T. Repantis and V. Kalogeraki, “Cordinated Media Streaming and Transcoding in Peer-to-
Peer System”, Proceedings of the 19th InternationalParallel and Distributed ProcessingSymposium,
Denver, CO, USA, April 2005
[5] http://www.distributed.net
[6] Korpela, E., Werthimer, D., Anderson, D., Cobb, J., and Lebofsky, “SETI@home: Massively
Distributed Computing for SETI,” In Journal on Computing in Science and Engineering, Volume 3,
Issue 1, pp. 78 – 83, 2001.
[7] D. Anderson, J. Cobb, E. Korpela, M. Lebofsky, and D. Werthimer. SETI@home: An Experiment in
public-resource computing. Communicationsof the ACM, 45:56–61, 2002.
[8] Jigyasu Dubey, Dr. (Mrs.) Vrinda Tokekar, Anand Rajavat, “A Study of P2P Computing Networks”,
in proceedings of International Conference on Computer Engineering and Technology(ICCET’ 10),
pp 623-627, 13-14 Nov. 2010, Jodhpur, India, ISBN No 978-81-920748-1-8
[9] Jingnan Yao, Jian Zhou, Laxmi Bhuyan, “Computing Real Time Jobs in P2P Networks”,in
proceedings of 31st IEEE Conference on Local Computer Networks 2006, pp 107 – 114, 14-16 Nov.
2006.
[10] C. Liu, L. Yang, I. Foster and D. Angulo, “Design and Evaluation of a Resource Selection Framework
for Grid Applications”, 11th IEEE Symposium on High-Performance Distributed Computing, pp 63 –
72, 2002.
[11] A. Gupta, D. Agrawal and A. E. Abbadi, Distributed Resource Discovery in Large Scale Computing
Systems., IEEE Symposium on Applications and the Internet, 2005, pp 320 - 326.

11
[12] Jingnan Yao and Laxmi Bhuyan, “Distributed Packet Processing in P2P Networks”, Published in
proceedings of IEEE Global Telecommunications Conference, GLOBECOM'05, Volume 1, pp. 142-
147, 28 Nov to 2 Dec. 2005
[13] Jaier Celaya, Unai Arronategui, “Distributed Scheduler of Workflows with Deadlines in a P2P
Desktop Grid”, Published in 18th Euromicro Conference on Parallel, Distributed and Network-based
Processing (PDP) 2010, pp. 69 – 73.
[14] Virginia Lo, Daniel Zappala, Dayi Zhou, Yuhong Liu, and Shanyu Zhao, “Cluster Computing on the
Fly: P2P Scheduling of Idle Cycles in the Internet”, In Proc. 3rd International Workshop on Peer-to-
Peer System (IPTPS 2004). San Diego, Feb. 2004.
[15] D. Kamenopoulos, I.A. Osman, P.Tsanakas, “Distance measurement in volunteer computing
networks: a completely decentralized approach”, Published in proceedings of Third IEEE
International Symposium on Network Computing and Applications, 2004. (NCA 2004), pp. - 399 -
404
[16] Jigyasu Dubey, Dr. (Mrs.) Vrinda Tokekar, “Identification of Reliable Peer Groups in Peer-to-Peer
Computing Systems”, published in proceedings of Third International Conference on Advances in
Communication, Network, and Computing – CNC 2012, LNICST pp.233-237, Bangalore, India.
[17] Jigyasu Dubey, Vrinda Tokekar, “P2PCS-A Pure Peer-to-Peer Computing System for Large Scale
Computation Problems”, In proceedings of IEEE International Conference on Computational
Intelligence and Communication Networks (CICN2011), pp. 582-585, Gwalior, India, Oct-201.
[18] Jigyasu Dubey, Dr. (Mrs.) Vrinda Tokekar, “A Framework for Pure Peer-to-Peer Computing
System”, Published in Proceedings of Ninth IEEE and IFIP International Conference Wireless and
Optical Communications Networks WOCN2012 Next Generation Internet, held during 20th, 21st and
22nd of September 2012, Indore, India, 978-1-4673-1987-4.
Authors
Jigyasu Dubey
Associate professor
Department of Information Technology
Shri Vaishnav Institute of Technology and Science Indore, (M. P.), India
Ph. D (Pursuing) (Computer Engineering) from DAVV, Indore
M.E. (Computer Engineering) in 2007 from DAVV, Indore
B.E. (Computer Science and Engineering) in 2000 from Vikaram University, Ujjain
Areas of Interest:
Computer Networks, Software Engineering, Object Oriented Analysis and Design
Dr. (Mrs.) Vrinda Tokekar
Professor & Head,
Department of Information & Technology,
Institute of Engineering & Technology,
Devi Ahilya University, Indore (M.P.) India
Ph. D. (Computer Engg.) in 2007 from DAVV, Indore
M.E. (Computer Engg.) in 1992 from DAVV, Indore
B.E. (Hons.) EEE, BITS Pilani in 1984,
Areas of Interest:
Computer Networks, Distributed Computing,, Security in Wireless Networks, e-Governance,
Multimedia Communication, Software Engineering

IDENTIFICATION OF EFFICIENT PEERS IN P2P COMPUTING SYSTEM FOR REAL TIME APPLICATIONS

More Related Content

What's hot (17)

Similar to IDENTIFICATION OF EFFICIENT PEERS IN P2P COMPUTING SYSTEM FOR REAL TIME APPLICATIONS (20)

More from ijp2p (20)

Recently uploaded (20)

IDENTIFICATION OF EFFICIENT PEERS IN P2P COMPUTING SYSTEM FOR REAL TIME APPLICATIONS