International Journal of Computer Science and Communication Vol. 2, No. 2, July-December 2011, pp. 315-318 THREAD BASED CACHE CONSISTENCY MODEL FOR MOBILE ENVIRONMENT G. Shanmugarathinam1 and K. ViveKanandan2 1 1 Lecturer,Research IT Scholar Dept., (External Ibri College Mode), of Technology, Bharathiar University, Sultanate India of Oman Research Work, Bharathiar University, Coimbatore. 2Professor, Bharathiyar University, India. ABSTRACT In this paper, we propose a cache consistency maintenance scheme called Thread Based Cache Consistency Model (TBCCM), for mobile environment. It lies on the following four key features. (i) Use of log at both server and mobile user’s cache maintains data consistency. (ii) Use of Thread Agent (TA) at both client & server. (iii) Use of log database at server, (iv) Use of Migration Thread Agent at serve. Since log is maintained in both server and client it is easy for server as well as client to make their data object reliable. The data object is reliable due to the usage of log comparison technique. Thread Agents at server and client make an internal request response operation(s) to access or update data object. This makes the medium to be liable. Log database at server maintains log of both client and server. Whenever there is need of update or access, log of client is checked along with log database at server. If there is need of access or update it precedes further process such as Security Level checking, synchronization Process etc., Migration Agent at server is activated upon updation of data object at server by certain client. It also seeks for other client(s) who are in need of the same data object that are updated currently. Keywords: Mobile Database, Wireless Networks Database Cache, Threads Agent, Low Bandwidth Networks. 1. INTRODUCTION there is an Updation at server, synchronization module In mobile client/server computing environments, begins to function in order to update new data object at mobile clients make access to their server to get Mobile client. Upon Updation to client cache a log is interested data and then are disconnected because of maintained in both server and client. Our process deals high cost of wireless communication or low bandwidth with disconnected approaches. Agent provides or low power consumption. Mobile clients usually keep optimization on the data object and services between their own local copies in order to reduce the overhead client and server. of communication with the server. The updates of the server database sometimes are subject to leading to 2. RELATED WORK invalidation of the cached map in mobile clients.[1],[2]. An evolution of data consistency becomes most required However it is not efficient to resend the entirely copied because of two major reasons. One is the limited map from the server to mobile clients for solving bandwidth of wireless channels and the other one is short invalidation. This paper proposes a Thread based update battery span. To conquer these two major problems many propagation method to propagate the server’s update have approached their techniques. But we are in situation into its corresponding mobile clients by sending only to choose upon which would be better in sense of update logs. Our approach overrides Invalidation performance, reliability and cost.In order to make data Report Scheme in previous algorithm(s). Instead of it reliable earlier algorithms have adapted various schemes. we use various Agents in server and client. Upon Some of them are Greedy dual-LU, Bit Sequences (BS), IR-Based Cache Invalidation Modal [2], [3]. All the above request/response an internal report is transferred specified approaches make an invalidation report for between Thread Agent (TA) at client and Thread Agent inconsistent data object. Based on the report the client (TA) at server. Migration Thread keeps track of has to update their data object in their cache. Here even frequently accessed data objects & users. Log based though the process is far better, it spends time in approach solves inconsistent data object. Because generating Invalidation report and checking the report whenever energy level or bandwidth is down, a log is generated. maintained in client cache. After certain period when client fetches adequate bandwidth or energy, Thread 3. THREAD BASED CACHE CONSISTENCY MODEL Agent allows to run a new Thread along with last (TBCCM) accessed data log. The log maintained in the server In this paper, we propose a cache consistency maintenance database is compared with recent client access log. If scheme called Thread Based Cache Consistency Model 316 International Journal of Computer Science and Communication (IJCSC) (TBCCM), for mobile environment. Strictly speaking, with Thread Agent (TA) at server. Each make a mutual TBCCM is a hybrid of stateless and stateful approaches understanding between them, upon it data object is made that it maintains log state information. On the other hand, to be valid in both client and server. Moreover maintaining unlike existing synchronous stateless approaches, log at server and client makes data object to be reliable. TBCCM does not require to produce an Invalidation The following subsections describe the proposed Report (IR). Since Thread Agent (TA) at client keeps track algorithm in detail. 3.1 System Architecture Figure 1: Thread Based Approach System Architecture Key Terms TL = Thread Listener SM = Security Module SynchM = Synchronization Module TM = Thread Monitor or Thread Manager MTA = Migrating Thread Agent Table 1 Summary of Thread Based Approach Components Components Client Server Thread Listener To listen for response, to synchronize To listen for requests, to synchronize and prompt with server. and prompt users for login authentication. Security Module To authenticate user identity and to To verify userID and to keep track of who control user access based on oranted can which database tables at the server for privileges. synchronization. Synch Module To work together with the synchronization Works together with the synchronization fixed server agent to propagate changes to module at the mobile client’s agent to and from the client table cache. propagate changes to and from the server. Thread Module This module keeps track upon Thread This module keeps track upon Thread creation, flow and maintenance. creation, flow and maintenance. DBListener To listen for database connection requests To listen for database connection requests from server. from client. Table Cache To store Data Base objects To store Data Base objects Thread Based Cache Consistency Model for Mobile Environment 317 3.1.1 Thread Agent at Server 4. ALGORITHM The Thread Agent (TAs) at server maintains as well as Algorithm presented below in figure 2 and 3 shows keeps on monitoring the frequent broadcast values and typical approach of managing data consistency in mobile frequent client cache access as shown in Figure 1 computing. We present two procedures MT New Whenever a value is read / written to server, it has to be Data ( ) and MT Update Data ( ) at server and each MU updated and to be broadcasted. During this process continuously executes the MT New Data ( ) or Thread Agent (TAs) maintains a Thread log which holds MT Update Data ( ). The Psudocodes MT New Data information about broadcast values, information of ( ) and MT Update Data ( ) and MU ( ) are shown below. mobile client who needs updated value. Algorithm for Server 3.1.2 Migration Thread Agent MT New Data ( ) Loop until Time t Whenever write operation is performed by mobile client Waits for Client Request to the server, a special Thread called “Migration Thread Fetch Request R from Client C Agent” will be activated upon write operation by client SM ( ) Checks is Client C is authenticated IF Authenticated == True to the server, which will be keep monitoring which client Allow Client C to access required Data Object d at Time t is performing the write operation to the server. It ELSE maintains a write log of both client & server. It also seeks Access Denied for Client C whether the updated data object is required by any other End of Loop MT Update Data ( ) mobile clients. If so it finds certain client(s) and it begins Data Object d to be updated to synchronize in order to keep data object at cache as SM ( ) Checks is Client C is authenticated more reliable. IF Authenticated == True Allow Client C to update required Data Object d at Time t 3.1.3 Thread Agent at Client ELSE Access Denied for Client C IF Update Data object d needed for Cn The Thread Agent (TAC) at client maintains as well as keeps on monitoring the frequent broadcast values and MTA ( ) Seeks for N number of Clients C IF C is Found (C = = n) Thread Agent (TAs) at server. Whenever a value is read SynchM ( ) Begins it operation(s) or written to server, it will be updated to server. Now ELSE the updated value will be broadcasted to the requested NO Clients Found for Update. mobile client. The Thread Agent (TAc) at client maintains End MTA ( ) a Thread log which holds information such as broadcast ELSE No Data Object to Update values, broad cast time, threadIDc, threadIDs, logIDc, and logIDs. Figure 2 3.1.4 Data Sequence Algorithm for Client For every data object with unique identifier x, the data MU ( ) { sequence for Mobile Server and Mobile user’s cache are SET Mode = Sleep / Awake; Sleep = 0; Wake = 1 IF Mode == 1 as follows: Loop Until Time t IF New Data object dx to be Read (dx, tx, tid,ut ,lid) where dx is the data object, tid is thread id, u is update time & l log id Send Request R to Server S t id Upon Receiving Ri from Client C, (dx, tsx, tsid, ust, lsid) where dx is the data object, tid is SM ( ) Checks is Client C, is authenticated thread id, ut is update time & lid log id IF Authenticated == True Allow Client C to access required Data Object d at Time t Table 2 x Update Cache C at Client C, Communication Message in TBCCM ELSE Block Client C, Until Permission is Granted.