M I S
Management Information Systems 2010/2011 Lecture … (10) 81 Transaction Processing Systems
Transaction processing systems (TPS's) were among the earliest computerized systems. Their primary purpose is to record, process, validate, and store transactions that take place in the various functional areas of a business for future retrieval and use.
Transaction processing systems are cross-functional information systems that process data resulting from the occurrence of business transactions, such as sales, purchases, deposits, withdrawals, refunds, and payments. A TPS is also acts as main link between the organization and external entities, such as customers' suppliers, distributors, and regulatory agencies
Types of Transaction Processing System (TPS's)
1.
2.
On-line system: involves a direct connection between operator and the TPS program. They provide immediate result and used to process a single transaction at a time. Ex: an order arrives by telephone call; it is processed at that moment and the result are produced. Batch-processing system: This is a second type of TPS, where transactions are grouped together and processed as a unit. Ex: cheque processing system in a bank. Types of Transactions:
1.
Internal Transactions: Those transactions, which are internal to the company and are related with the internal working of any organization. For example Recruitment Policy, Promotion Policy, Production policy etc. 82 2.
External Transactions: Those transactions, which are external to the organization and are related with the external sources, are regarded as External Transaction. For example sales, purchase etc. TPS Properties:
1.
2.
3.
4.
5.
Consistency: The transaction is a correct transformation of the state. This means that the transaction is a correct program. Isolation: Even though transactions execute concurrently, it appears to the outside observer as if they execute in some serial order. Isolation is required to guarantee consistent input, which is needed for a consistent program to provide consistent output. Reliability: TPS system is designed to ensure that all transactions are entered in sequential and systematic manner. Standardization: Transactions must be processed in the same way each time to maximize efficiency and effectiveness. Controlled Access: Since TPS also contains confidential matters or data; it acts as powerful tool for the organization. Hence access must be restricted. Objectives (Goals) of TPS
1. Process data generated by and about transactions.
2. Maintain a high degree of accuracy.
3. Ensure data and information integrity and accuracy.
4. Produce timely documents and reports.
5. Increase labor efficiency.
6. Help provide increased and enhanced service.
7. Help build and maintain customer loyalty.
8. Achieve competitive advantage.
83 Transaction Processing Activities
1. Data collection: Capturing data necessary for the transaction.
2. Data editing: Check validity and completeness of data.
3. Data correction: Correct the wrong data.
4. Data manipulation: Calculate, summarize, Process data.
5. Data storage: Update transactions (on Databases).
6. Document production and reports: Create end result reports.
FIGURE 18: Transaction Processing Activities
84 Major Characteristics of TPS
1. TPS handles data which shows the results of various activities on historical basis i.e., activities which have already happened.
2. It is relevant to all functional areas inside organization i.e. (production, marketing, finance and human resources) because each area has some kind of transaction.
3. TPS helps to assess the organizational performance.
4. Large amounts of data are processed.
5. The sources of data are mostly internal, and the output is intended mainly for an internal audience.
6. The TPS processes information on a regular basis: daily, weekly, monthly, annually etc.
7. It provides high processing speed to handle the high volume of data.
8. Input and output data are structured (i.e., standardized).
9. A high level of accuracy, data integrity, and security is needed which is provided by TPS.
Transactions on a Transaction-Oriented Database System
A transaction processing system plays a role as a mediator that accepts transaction requests from users, dispatches these requests to the database system, coordinates the execution of the involved transactions, and forwards transaction results to the original acquirers.
The common programming model for a transaction-oriented database system is the client-server model. Users or clients interact with the database system by submitting their transaction processes that consist of one or many database operations to the transaction processing system. The transaction Management Information Systems 2010/2011 Lecture … (10)
85 processing system will coordinate and manage the execution of these transaction processes by subsequently sending these database operations to the database system. The database system will carry out the actual execution of the submitted database operations. Finally, the transaction results that reflect the consistent states of the database system are returned to the clients.
FIGURE 19: Transaction-oriented database system
FIGURE 20: Dataflow of transaction-oriented database systems
TPSData BaseMessages to outside worldClients Management Information Systems 2010/2011 Lecture … (10) 86 To protect the integrity constraint of the database system, a set of essential components that includes a transaction manager, a scheduling manger and a log manger are deployed. Additional components such as communication manger or other resource managers can also be employed by the transaction processing system.
FIGURE 21: Transaction processing system components
The role of each transaction processing component is described as follows:
1.
2.
3.
Scheduling Manager. The scheduling manger manages the order of execution of the database operations. Usually, the scheduling manager makes use of concurrency control protocols, for example locking or timestamp protocols, in order to control the execution of transactions. Thus, the scheduling manger supports the isolation and consistency properties of transactions. Based on the applied concurrency control protocol, the scheduling manager will determine an execution order in which the submitted database operations will be carried out. For example, if a locking protocol is used, the scheduling manager will decide whether a lock request will be granted to the acquired transaction, or if a timestamp protocol is applied, the scheduling manager will assess if a submitted operation will be allowed to be carried out. Log Manager. The role of the log manager is to support the database system to recover from failures. The log manager keeps track of the changes of the database states by recording the history of transaction execution. Depending on the deployed recovery strategies, for example undo and/or redo, the log manger will record necessary information in a non-volatile logbook. The log manager ensures the atomicity and the durability properties of transactions. 
Comments
Post a Comment