; Priority Based Mobile Transaction Scheme Using Mobile Agents
Learning Center
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Priority Based Mobile Transaction Scheme Using Mobile Agents


The International Journal of Computer Science and Information Security (IJCSIS) is a reputable venue for publishing novel ideas, state-of-the-art research results and fundamental advances in all aspects of computer science and information & communication security. IJCSIS is a peer reviewed international journal with a key objective to provide the academic and industrial community a medium for presenting original research and applications related to Computer Science and Information Security. . The core vision of IJCSIS is to disseminate new knowledge and technology for the benefit of everyone ranging from the academic and professional research communities to industry practitioners in a range of topics in computer science & engineering in general and information & communication security, mobile & wireless networking, and wireless communication systems. It also provides a venue for high-calibre researchers, PhD students and professionals to submit on-going research and developments in these areas. . IJCSIS invites authors to submit their original and unpublished work that communicates current research on information assurance and security regarding both the theoretical and methodological aspects, as well as various applications in solving real world information security problems. . Frequency of Publication: MONTHLY ISSN: 1947-5500 [Copyright � 2011, IJCSIS, USA]

More Info
  • pg 1
									                                                          (IJCSIS) International Journal of Computer Science and Information Security,
                                                          Vol. 9, No. 1, January 2011

    Priority Based Mobile Transaction Scheme Using
                    Mobile Agents
                                               J.L.Walter Jeyakumar #1, R.S.Rajesh #2
                                            Department of Computer Science and Engineering,
                                                Manonmaniam Sundaranar University,
                                                   Tirunelveli, Tamilnadu, INDIA.

Abstract— We define a priority based mobile transaction                    due to disconnection should be minimized in mobile
scheme in which mobile users can share data stored in the cache            transactions. The low and variable bandwidth of wireless
of a mobile agent which is a special mobile node for coordinating          network together with the expensive transmission cost makes
the sharing process. This framework allows mobile affiliation              bandwidth consumption an important concern [2]. Correctness
work group to be formed dynamically with a mobile agent and                of transactions executed on both fixed and mobile hosts must
mobile hosts. Using short range wireless communication                     be ensured by the operations on shared data. Blocking of
technology, mobile users can simultaneously access the data from           mobile transactions due to long disconnection periods should
the cache of the mobile agent. Data Access Manager module at
                                                                           be minimized to reduce communication cost and to increase
the mobile agent enforces concurrency control using cache
invalidation technique. Four levels of priority are assigned to the        concurrency [3]. After disconnection, mobile host should be
requesting mobile nodes based on available energy and                      able to process transactions and commit locally. In Mobile
connectivity. This model supports disconnected mobile                      computing, there is always a competition for shared data since
computing by allowing mobile agent to move along with the                  it provides users with the ability to access information through
Mobile Hosts. The proposed Transaction frame work has been                 wireless connections that can be retained even while the user
simulated in J2ME and NS2 and performance of this scheme is                is moving. Further, mobile users are required to share their
compared with existing frame works.                                        data with others. This provides the possibility of concurrent
                                                                           access of data by mobile hosts which may result in data
Key words – Transaction, concurrency control, mobile database,
                                                                           inconsistency. Concurrency control methods have been used
cache invalidation, mobility
                                                                           to control concurrency. Due to limitations and restrictions of
                       I. INTRODUCTION                                     wireless communication channels, it is difficult to ensure
                                                                           consistency of data while sharing takes place.
    Mobile computing environment consists of Fixed Hosts
                                                                               In this paper, we present a priority based mobile
(FHs), Mobile Hosts (MHs) and Base stations or Mobile
                                                                           transaction frame work that allows mobile users to share data
Support Stations (MSSs). MH is connected to the Fixed
                                                                           cached in the Mobile Agent which is a special node for
network through MSS via wireless channels. The
                                                                           coordinating the sharing process. Whenever an MH enters into
Geographical area covered by a MSS is called a cell. Mobile
                                                                           a Mobile Agent area it can connect and access the data in the
Hosts are portable computers which move around in a cell.
                                                                           cache. But upon update request by a MH, updation is done at
When a MH enters into a new cell hand-off or hand-over takes
                                                                           the local cache and invalidation report is sent to all the mobile
place. MH communicates only with the MSS responsible for
                                                                           hosts which have already accessed the same data. This will
its cell. Transactions and data management functions are done
                                                                           force the mobile hosts to refresh their data values. This
using the data base servers installed at MSS. In mobile
                                                                           framework also provides the provision for transaction update
computing, it is necessary that a computation is not disrupted
                                                                           during disconnection. Data Access Manager (DAM) at the
while an MH is not connected. The part of the computation
                                                                           Mobile Agent will take care of concurrency control while
executing on an MH might continue executing concurrently
                                                                           sharing takes place. Concurrency control is enforced using
with the rest of the computations while the MH is moving and
                                                                           cache invalidation technique. In order to give priority to the
not connected to the network [8]. With the evolution of
                                                                           mobile nodes running on low power and with low
PCS( personal Communication System) and GSM(Global
                                                                           connectivity, four levels of priority are used. We also take into
System for Mobile communication), advanced wireless
                                                                           account the mobility of the Mobile Hosts that has a strong
communication services are being offered to the mobile users.
                                                                           impact on mobile applications [4].
Mobile Database System is a distributed client/server system
                                                                               The remaining part of this paper is organized as follows.
based on PCS or GSM in which clients can move around
                                                                           Section II summarizes the related research. Section III focuses
freely while performing their data processing activities in
                                                                           on the Mobile Agent based architecture. Section IV specifies
connected or disconnected mode. Frequent disconnections,
                                                                           the proposed framework for disconnected mobile computing.
mobility, limited battery power and resources pose new
                                                                           Section V gives the performance analysis. In Section VI, the
challenges to mobile computing environment. Frequent aborts

                                                                      86                             http://sites.google.com/site/ijcsis/
                                                                                                     ISSN 1947-5500
                                                     (IJCSIS) International Journal of Computer Science and Information Security,
                                                     Vol. 9, No. 1, January 2011

discussion on the proposed model is presented. Finally,                 dynamicity of the life time of data was proposed. Here, life
Section VII concludes the paper.                                        span of data is predicted based on the probability of updation
                                                                        of data item. This method makes PLP of data item very close
                      II. RELATED WORK                                  to the actual valid life span of a data item.
    When simultaneous access to data is made at the server,                 In [10], a transaction model for supporting mobile
concurrency control techniques are employed to avoid data               collaborative works was proposed. This model makes use of
inconsistency. Conventional locking based concurrency                   Export-Import repository, which is a mobile sharing work
control methods like centralized Two Phase locking and                  space for sharing data states and data status. But in the Export-
distributed Two Phase locking are not suitable for mobile               Import repository based model, locking is the main technique
environment. In centralized two phase locking scheme [17,               which has the following disadvantages. (i) More bandwidth is
19], where one node is responsible for managing all locking             needed for request and reply since the locking and unlocking
activities, the problem of single point failure cannot be               requests have to be sent to the server. (ii) Disconnection of
avoided. The distributed two phase locking scheme used in               mobile host or a transaction failure will result in blocking of
[18], allows all nodes to serve as lock managers. But in the            other transactions for a long period. Our framework is better
event of data partition, this algorithm could degenerate into a         than the model which uses Export-Import repository for
centralized two phase scheme.          In conventional locking          sharing data since it minimizes message communication costs
scheme, the communication overhead that arises due to                   and data update costs to a larger extent. Also disconnected
locking and unlocking requests can create a serious                     mobile hosts are treated separately within a mobile affiliation
performance problem because of low capacity and limited                 by waiting for their reconnection. Transaction Management
resources in mobile environment [5]. Moreover, it makes                 solution proposed in [15], is for reducing energy consumption
mobile hosts to communicate with the server continuously to             at each MHs by allowing each MH to operate in three modes,
obtain and manage locks [6].                                            Active, Doze, and Sleep thus providing a balance of energy
    The timestamp approach for serializing the execution of             consumption among MHs.
concurrent transactions was developed for the purpose of
more flexibility, to eliminate the cost of locking, and to cater                   III. MOBILE AGENT BASED ARCHITECTURE
for distributed database systems [7, 13]. In timestamping, the              The proposed Mobile Agent based architecture model is
execution order of concurrent transactions is defined before            illustrated in Fig 1. The model consists of Mobile Hosts,
they begin their execution. The execution order is established
by associating a unique timestamp to every transaction. When
two transactions conflict over a data item, their timestamps are
used to enforce serialization by rolling back one of the
conflicting transactions. To exploit the dynamic aspects of
two phase locking and the static ordering of timestamping, a
number of concurrency control techniques were developed
using a combined approach. In mixed approach techniques
called Wound-wait and Wait-die [7, 16], the enforcement of
mutual exclusion among transactions is carried out using
locking while conflicts are resolved using timestamps.
    In [14], optimistic concurrency control scheme is used to
minimize locking overhead by delaying lock operation until
conflicting transactions are ready to commit. They rely on
efficiency in the hope that conflicts between transactions will
not occur. Without using lock during the execution of the
transactions, this scheme promotes deadlock free execution.
In optimistic concurrency control with dynamic time stamp
adjustment protocol, client side write operations are required.
But it may never be executed due to delay in execution of a
transaction [1]. In multi version transaction model [9], data is               - Disconnected Mobile Host    MSS – Mobile Support
made available as soon as a transaction commits at a mobile                    - Connected Mobile Agent      DBS – Data Base
host and another transaction can share this data. But data may                                                       Server
be locked for a longer time at a mobile host before the lock is                -    Mobile Affiliation       BSC – Base Station
released at the database server.                                                                                     Controller
    In [11], AVI (Absolute Validity Interval) was introduced                   -    Wireless LAN
for enforcing concurrency control without locking. AVI is the
valid life span of a data item. But it calculates AVI only                     -    Short range wireless
based on previous update interval. In [12], a method based on
PLP (Predicted Life Period), which takes care of the                            Fig.1. Mobile Agent Based Architecture model

                                                                   87                                http://sites.google.com/site/ijcsis/
                                                                                                     ISSN 1947-5500
                                                      (IJCSIS) International Journal of Computer Science and Information Security,
                                                      Vol. 9, No. 1, January 2011

Mobile Support Stations (MSS), and Data Base Servers (DBS).            energy and low connectivity is assigned the highest priority.
MSS is connected with a Base Station Controller (BSC) [20],            Other levels of priority are assigned according to the various
which coordinates the operations of MSS using its own stored           possibilities as given in Table I.
program. Unrestricted mobility is supported by wireless link
between MSS and Mobile Hosts. Each MSS serves one cell                                             TABLE I
whose size depends on the power of its MSS. Data Base                                         LEVELS OF PRIORITY
Servers are connected to the mobile system through wired
lines as separate nodes. Each DBS can be reached by any                  Status of        Energy             Connectivity        Priority
MSS and new DBSs can be connected and old ones can be                   an MH(Aij)      Availability(i)          (j)
taken out from the network without affecting mobile                         A00             Low                 Low                  1
communication. A DBS can communicate with a MH only via                     A01             Low                 High                 2
MSSs.                                                                       A10             High                Low                  3
    Mobile agent is a special mobile node which connects to                 A11             High                High                 4
the MSS to cache the frequently accessed data. Disconnected
Mobile Hosts can connect to the Mobile Agent using short
range wireless communication technologies to form mobile               B. Concurrency Control Mechanism
affiliation workgroup.
    Mobile hosts are allowed to access data from the cache.               When more number of mobile hosts are accessing data
When data request is made for the first time, data is retrieved        simultaneously the problem of data inconsistency arises. This
from the server and stored in the cache. Subsequent requests           problem can be solved if we use an efficient concurrency
are handled by the Data Access Manager module itself. When             control mechanism. When data request is made for the first
a mobile host requests for data update, after local updation of        time, data is retrieved from the server and stored in the cache.
the data item, invalidation report is sent to all the mobile hosts     Future requests for data are managed directly by the Data
that have already accessed the same data. This makes all the           Access Manager.
mobile hosts to refresh their data values. When a mobile host             Data Access Manager uses a suitable data item format to
is disconnected from the Mobile agent after updation request,          store data as quintuple [12] in the cache. It has (id, TLU, PLP,
the updation task is transferred to the Data Access Manager in         dataval, NT) where id denotes unique Id of the data item, TLU
the Mobile Agent. Data Access Manager module is used to                indicates time of Last Update, PLP is Predicted Life Period,
coordinate the operations in the cache.                                dataval is current value of the data item and NT denotes
     After disconnected from the server, Mobile Agent can              number of transactions that concurrently access the data item.
move along with the connected MHs and MHs can continue                     When Data Access Manager fetches data for the first time
their transaction execution. If data update at the server is           from the server, it sets TLU to current time, PLP to optimal
requested, mobile agent will wait for reconnection before              time based on the nature of data item and NT to 1. NT is
updation is made.                                                      incremented whenever a new data access request is made.
                                                                       Data in the cache becomes invalid, once it is updated in the
          IV. MOBILE T RANSACTION FRAMEWORK                            server. Life span of a data item is predicted using PLP. It
   When Mobile Hosts enter into the Mobile Agent area, they            makes use of the probability of updation as a basis for setting
connect to the Mobile Agent using short range wireless                 valid life span of a data item. In PLP interval, data item is
network technology to form Mobile Affiliation Work Group.              valid and all the mobile hosts can access same data item
Frequently accessed data are cached in the Mobile Agent.               concurrently.
Mobile Hosts can access the cached data in the Mobile Agent.               When a MH makes update request or PLP expires, the
The Data Access Manager module at the Mobile Agent is                  data item is invalidated. Now PLP is modified and
responsible for enforcing concurrency and cache invalidation.          invalidation report is sent. The predicted life period of data
                                                                       item is computed using the following formula given in [12].
A. Energy and Connectivity Evaluation                                  PLP=PPLP ± (p*PPLP)
                                                                       Where PPLP is Previous Predicted Life Period and p is
    Mobile Hosts all the time maintains its energy availability        predicted probability of updation of data item. p =
and connectivity. Connectivity is evaluated based on signal            Total_updates / NT. It is the ratio of data item update to data
strength. When signal strength goes below one fourth of total          item access. Since predicted probability of updation is based
strength, connectivity is considered as Low. When available            on recent past history of updation rate, it is highly probable
energy goes below 25% of total energy level, then energy               that PLP is very close to the actual validity interval of the data
availability is considered as Low. The status of an MH based           item.
on Energy Availability and Connectivity (Aij) can be A00 –
Low Energy & Low Connectivity, A01 – Low Energy & High                 C. Transaction Execution in the MH
Connectivity, A10–High Energy & Low Connectivity and A11
– High Energy & High Connectivity. When Data Access                     After connecting to the Mobile Agent , the MH intimates the
Manager receives a transaction request from a mobile host, it          status of Energy availability and Connectivity (Aij) of the MH
assigns a priority level using Aij. A mobile host with low             to DAM at Mobile Agent. Then, the execution of the

                                                                  88                              http://sites.google.com/site/ijcsis/
                                                                                                  ISSN 1947-5500
                                                           (IJCSIS) International Journal of Computer Science and Information Security,
                                                           Vol. 9, No. 1, January 2011

transaction starts locally. If the operation is Data Read, the               Mobile Agent with the server is established. After
request is sent to DAM at the Mobile Agent. Otherwise, If the                reconnection with the server, update request is forwarded to
MH wants to update data, it first checks whether it is a                     the server. The server updates the data and sends invalidation
transaction update or not. If it is a transaction update, it will            confirmation along with the updated value. Once Data Access
check if the MH is about to be disconnected. If so, the                      Manager receives the confirmation, it updates the data in the
updation task is assigned to the Data Access Manager before                  cache. The data in the cache is invalidated if updation is made
disconnection. Otherwise, update request is sent to Data                     in the server or PLP expires.
Access Manager. The algorithm is given in Fig. 2.                                If transaction update is made by the Data Access Manager
                                                                             for the disconnected MH, the above procedure is followed
 Trans_Req_from_MH_to_MA(T, MH_ID )                                          except that at the end, DAM generates updation report and
// Transaction T is initiated by an MH whose ID is MH_ID //                  forwards it to the MH when it gets reconnected. The
Begin                                                                        algorithms are shown in Fig. 3 and 4.
    Connect to the Mobile Agent (MA’)
    Intimate the status of Energy availability and Connectivity (Aij)
        of the MH to DAM at MA’
    Start execution of the transaction T locally
    If Data Read                                                             Data_Access_Manager(T, MH_ID, Aij )
        Submit Read Request to DAM at MA’                                    // DAM module will be executed when MA receives a request from
    Else If Data Update                                                      transaction T with Requesting Mobile Host ID MH_ID, Status of
             If MH is about to be disconnected                               Energy availability and Connectivity Aij //
                 Assign updation task to DAM at MA’ and disconnect           Begin
             Else                                                                Schedule the transaction T by assigning priority to it by checking
                 Submit Update Request to DAM at MA’                                 Aij and place it in a priority queue
             End If                                                              Take the first transaction T1 from the queue
          Else if commit                                                         If Read Request
                    Commit                                                            If data is in the cache of MA
                Else                                                                       Update NT
                    Exit                                                              Else
                End if                                                                     If MA is disconnected
           End if                                                                              Wait for reconnection
                                                                                               If reconnected with server
   End if                                                                                          Fetch data from server and initialize quintuple
End                                                                                            End if
                                                                                           End if
                     Fig.2. MH Execution algorithm                                        Send data to MH
                                                                                      End if
                                                                                Else If Connected or disconnected Update request
                                                                                            Update locally and send invalidation report
D. Function of Data Access Manager and Server                                               If MA is disconnected
                                                                                                   While reconnection with the server is not got do
     After receiving a transaction request from an MH, the                                            Wait for reconnection
transaction is scheduled by assigning a priority to it by                                          End while
checking the energy availability and connectivity (Aij). Then,                             End if
the transaction is placed in a Priority queue. After scheduling,                           Forward update request to server
DAM will take the first transaction from the queue. When MH                                If Server Update / PLP expiration
makes a read request, if it is in the cache of the Mobile Agent,                                   Update quintuple
NT is incremented by one. Otherwise, if Mobile Agent is                                            Wait for MH request
connected to the server, it will fetch the data from the server                            End if
and initialize data item format. If Mobile Agent is                                       Check for any disconnected updates by MHs
disconnected, the read request will be put in the queue and it                                  if disconnected updates
will wait for Mobile Agent to get reconnection with the server.                                      Generate and forward update reports to MHs
Once reconnection is got, it will fetch data from the server and                                        when reconnection
initialize data item format.                                                                   End if
     When MH makes an update request, DAM updates data                                 Else
locally and invalidation report is sent to all the mobile hosts                            Wait for MH request
                                                                                       End if
that have already accessed the same data item. This forces all
                                                                                End if
the transactions to refresh their data values. In order to                    End
forward this update request to the server, it will check whether
Mobile Agent is connected to the server. If so, update request                                Fig.3. Data Access Manager Algorithm
is forwarded to the server. Otherwise, the update request will
be put in the queue and it will wait until reconnection of

                                                                        89                               http://sites.google.com/site/ijcsis/
                                                                                                         ISSN 1947-5500
                                                        (IJCSIS) International Journal of Computer Science and Information Security,
                                                        Vol. 9, No. 1, January 2011

Server_Execution()                                                       given in table II. It is found that MHs with priority level 1 get
Begin                                                                    the least average response time while MHs with priority level
      Wait for connection                                                4 have got the highest average response time. The average
      If connection request                                              response time for the MHs increases with the increase in
         Connect authorized Mobile Hosts
                                                                         priority level. This is due to the fact that scheduler gives more
         If Data Read
               Send data item                                            emphasis to top priority requests. Hence the proposed model
         Else if Data Update                                             is suitable for mobile networks with poor energy and poor
                   Update data and send invalidation confirmation        connectivity MHs.
                     with updated data
              End if
         End if                                                                                      TABLE II
      End if                                                                 A VERAGE R ESPONSE TIME FOR M OBILE H OSTS WITH PRIORITIES
                                                                                  Priority level           1       2        3        4
                  Fig.4. Server execution algorithm

                 V. PERFORMANCE ANALYSIS                                     Avg. Response time           4.5     7.6     10.8      15
                                                                                   in Sec.
    Simulation for the framework is done in Pentium Dual
Core System @ 2.4 GHz with 3 GB RAM using J2ME and
NS2. A mobile network is simulated with 25 nodes and 4
Mobile Agents. Mobile nodes move using random walk
mobility model [21]. Response time is calculated as the time
taken to service the request made by the mobile host. The
response time is evaluated for the executed transactions for
the E-I repository model and the proposed scheme for both
non disconnected and disconnected cases. The results of the
analysis for the two cases are shown in Fig 4 and 5.

                       VI. DISCUSSION
    Fig. 4 shows the comparison of the performance of E-I
repository model [10] with the proposed model for non
disconnected case. As the number of transactions increases,
the response time increases steadily for both E-I model and
the proposed model. But the proposed model suffers from a
slight increase in response time compared to the E-I model,                Fig.4. Analysis of Response time for Transactions without disconnection
when the number of transactions exceeds 11. This is due to the
presence of Agent delay. It is also found that E-I model takes
more response time than the proposed model up to 10
transactions. This is due to the extra time involved in
communication overhead for locking mechanism of E-I
repository model. The cache in the Mobile Agent also
contributes to the low response time taken by the proposed
    Fig. 5 for transactions with disconnection case shows
same trend as non disconnected case discussed above. But it is
found that the disconnection support is provided by the
proposed model only up to 12 transactions. And there is a
steep increase in response time, when the number of
transactions exceeds 12. This is due to the communication
overhead associated with more number of updation tasks that
are transferred to the Data Access Manager in the Mobile
Agent during disconnection, as the number of transactions
    The response time for Mobile Hosts with each priority                   Fig.5. Analysis of Response time for Transactions with Disconnection
level as discussed above are evaluated for 25 cases. The
average response time is estimated for each priority level is

                                                                    90                                 http://sites.google.com/site/ijcsis/
                                                                                                       ISSN 1947-5500
                                                            (IJCSIS) International Journal of Computer Science and Information Security,
                                                            Vol. 9, No. 1, January 2011

                       VII.       CONCLUSION                                     [19] T. Ozsu and Valduriez, Principles of Distributed Database Systems,
                                                                                      Prentice Hall, Englewood Cliffs, NJ, 1999.
                                                                                 [20] Rajan Kurupillai, Mahi Dontamsetti and Fil J. Cosentino, Wireless
   In this paper, we have proposed a priority based transaction                       PCS, McGraw-Hill, New York, 1997.
scheme for disconnected mobile computing environment.                            [21] T. Camp, J. Boleng, and V. Davies, “A survey of mobility models for
                                                                                      ad-hoc network research,” Wireless Communications & Mobile
Mobile Agent can form a Mobile Affiliation work group with                            Computing (WCMC): Special issue on Mobile Ad Hoc Networking:
the disconnected Mobile Hosts using short range wireless                              Research, Trends and Applications, vol. 2, no. 5, pp. 483-502, 2002.
technology. The frequently accessed data are cached in the
Mobile Agent. This cached data can be accessed by the
mobile hosts when they get connected. Since the proposed
scheme uses priority based scheduling for transactions, we get
better results for mobile networks in which MHs have poor
energy and poor connectivity. When mobile hosts are
disconnected from the Mobile Agent, transaction update task
can be transferred to the Mobile Agent. By using a Mobile
Agent and concurrency control without locking for accessing
data, we claim that message communication costs and
database update costs are minimized to a larger extent.

 [1] Ho-Jin Choi, Byeong-Soo Jeong, “A Timestamped Based Optimistic
     Concurrency Control for Handling Mobile Transactions,” ICCSA
     2006, LNCS 3981, pp. 796-805, 2006.
 [2] Serrano-Alvarado, P., C. Roncancio, and M. E. Adiba, “A Survey of
     Mobile Transactions,” Distributed and Parallel Databases, Vol. 16,
     No. 2, pp. 193-230, 2004.
 [3] Sanjay Kumar Madria, Mukesh Mohania, Sourav S. Bhowmick,
     Bharath Bhargava, “Mobile data and transaction management,”
     Elsevier Information Sciences [4], pp. 279-309, 2002.
 [4] Dunham, M. H., and V. Kumar, “Impact of Mobility on Transaction
     Management,” MobiDE, pp. 14-21, 1999.
 [5] Vijay Kumar, Mobile Database Systems, WILEY INTERSCIENCE,
 [6] Victor C.S., Kwok wa Lam and Son, S.H., “Concurrency Control
     Using Timestamp Ordering in Broadcast Environments,” The
     Computer Journal, Vol.45 No.4, pp. 410-422, 2002.
 [7] P.A Bernstein, V. Hadzilacos and N. Goodman, Concurrency control
     and Recovery in Database Systems, Addison Wesley, 1987.
 [8] Chrysanthis P.K., “Transaction processing in a mobile computing
     environment,” in Proc. IEEE Workshop on Advances in Parallel and
     Distributed Systems, October 1993, pp. 77–82.
 [9] Madria, S. K., M. Baseer, and S. S. Bhowmick, “A Multiversion
     Transaction Model to Improve Data Availability in Mobile
     Computing,” CoopIS/DOA/ODBASE, pp. 322-338, 2002.
[10] Le, H. N., and M. Nygård, “A transaction model for Supporting mobile
     Collaborative Works,” IEEE, pp. 347-355, 2007.
[11] Salman Abdul Moiz, Mohammed Khaja Nizamuddin,”Concurrency
     Control without Locking in Mobile Environments,” IEEE, pp. 1336-
     1339, 2008.
[12] Miraclin Joyce Pamila J.C and Thanuskodi K, “Framework for
     transaction management in mobile computing environment,” ICGST-
     CNIR Journal, pp. 19-24, 2009.
[13] V. Kumar, Performance of concurrency control mechanisms in
     Database Systems, Prentice Hall, Englewood Cliff, NJ, 1996.
[14] H.T. Kung and John T. Robinson, “On optimistic methods of
     concurrency control,” ACM Transaction of Database Systems, Vol. 6,
     No.2, 1981.
[15] Le Gruenwald, Shankar M. Banik, Chuo N. Lau, “Managing real time
     database transactions in mobile ad-hoc networks,” Springer, 27-54,
[16] D.J. Rosencrantz, R.E. Sterns, and P.M. Lewis, “System level
     concurrency control for distributed database systems,” ACM
     Transaction of Database Systems, Vol. 3, No. 2, 1978.
[17] P.A. Alsberg and J.D. Day, “A principle for resilient sharing of
     distributed resources,” in Proc. 2nd International Conference on
     Software Engineering, 1976.
[18] A Borr, “High Performance SQL through Low Level System
     Integration,” in Proc. ACM SIGMOD Conference, June 1988.

                                                                            91                                http://sites.google.com/site/ijcsis/
                                                                                                              ISSN 1947-5500

To top