Thought Paper for Distributed System National College of

Thought Paper for Distributed System National College of Business Administration & Economics Lahore, Sub-campus Layyah 30th Sep, 2019.

Study of Concurrency Control Techniques in Distributed DBMS Research Publications

1. Abstract:

One of the most important tasks performed by any data base management system is concurrency control. Yes, without establishing a proper concurrency control it is not possible it is not possible to maintain the integrity of DBMS in concurrent environment. The concurrency control is used to integrity the data base by using an algorithm that just focus on maintaining and consistency of data using synchronized access.

It is more simple and easy to integrate the database in a centralized environment to synchronize the data among the various concurrent transactions'. But, this system becomes very complex as compared to centralized framework. When data is synchronizing and then distributed framework system copies all the given data according to given transaction, with in multiple fragments, and isolation performs calculation for all the complexities of algorithm operation.

Get quality help now

RhizMan

Verified writer

Proficient in: College

4.9 (247)

“ Rhizman is absolutely amazing at what he does . I highly recommend him if you need an assignment done ”

+84 relevant experts are online

Hire writer

These all concurrency control method are the most popular method and widely used to performing in both centralized and in distributed framework of DDBMS. Remember, this algorithm, based on two-phase locking that has best distributed concurrency technique. In this paper I put all necessary techniques to control concurrency by following various phase based locking techniques.

2. Introduction:

Well we all known database is consider the backbone for all the organization in the world to store and manages the data for later use. Also, with these database systems is resposnisble to perfrom and manage all the task of any organization in an efficient manner.

There are a lot of database applications but, there are some which are dependent of database system. Database system is implemented by using two approaches i.e. one is centralized and other is distributed approach. "A type of database which is implemented using distributed approach techniques is known as Distributed Database System as well". In a distributed database system all the records of a given data in any organization provide the same results. This is the technology which can be seen where combination of database is in the form of computer Network Technologies.

"However, on the other hand a type of database system which is combination of for all given with a given database organization is referring to be a Database Management System."

"From all the above combination of both DDB and DDBMS is known as Distributed database System (DDBMS)". By using these techniques the several functions is performed in order to get the accurate results and to manipulate, manage and get the data in proper from like concurrency control, recovery, security and transaction management system which provide a results after passing all these process. As we also known that DBMS is responsible to control the concurrency of given data which is a best mechanism to ensure the data in reliable from separate consistency and integrity of the database.

3. Concurrency Control

In a database management system, the multiple user at the same time access the data in a correct form but, at the back end level each user performing its individual task on dedicated system. In a case of concurrent data transaction it is assure that all the data is in schedule form because, every user get the same results to one serial transaction for the excitation of same data.

· The problems that are handled in DDMBS to control the concurrency of data are as follows;

· Most of user searches the data at the same time from at a number of distant sites,

· Database is results in the form of b fragmentation and replicated across all the given different website,

· Complexity occurs during the synchronization of programs that is executed from a different sites for the multiple users, and more,

· To get the satisfaction either database is implemented from all sites them Concurrency control techniques is performed.

4. Concurrency Control Techniques'

A lot of concurrency has been given by making proper researches so for as, results of correct database can access. All the given concurrency control is divided into various categories broadly such like that the first one is "Locking Based Protocols", and the second one is Timestamp-Ordering Based Protocols and the third one is Optimistic Protocols. The first two categories of concurrency control is belong to pessimistic approach. This approach is used where the high activity of database is involved and, pessimistic protocols have the low quality of databases in use. Then the implementation is performed in each database of this concurrency protocols/algorithm to get the data results in more simpler and cheaper environment. Yeah implantation techniques are based on where environment of centralized databases is simpler than that of distributed databases environment. The number of concurrency occurred in database is easy to implemented and easily extended to handle all the problems that present in distributed database. But, other than this, also there exits another concurrency control techniques that are not handle and classified it into other categories is as follows.

5. Pessimistic Techniques

The technique in which synchronization is performed by carrying all the transaction of given data from initial phase of execution of life cycle. This technique is suitable and mostly used for where databases have high quality of activities in database system. Also with this we can say that the Pessimistic Technique is mainly used a large number of conflicting is performed on the database system to reduce the performance without wastage of time. Moreover the Pessimistic Techniques is further divide into two categories;

1. Timestamp-Ordering Based Techniques

2. Two Phase Locking (2 PL) Based Technique

Two Phase Locking (2 PL) Based Technique

Two phase locking technique is used ensure the concurrency of data transaction in order to maintain the data integrity in databases system. In each technique every object in database is associated with a given variable to share the variables from all resources Is refers to be Lock. A lock is used to store the data availability in data object that is mutually exclusive. In two phase locking two operation are performed which is called phases of operations. The first phase is used to degrade or release the data in a lock only. However, the phase 2 perform more than one task at the same time like locking follows the some rules to perform lock operation. First one is to the operation cannot be performed before the lock operation is completed, the second one is conflict lock should not exit more than one time, and third one is transaction is not affected in data operations until and unless locks are obtained. The results of Tow phase Locking Based Technique is results in form of deadlock and starvations.

Timestamp-Ordering Based Techniques

These techniques in which transaction are checked ordered by order is called timestamp. The timestamp technique is used to generate the value that has global incremental counter variables to get the uniqueness of data. The transactions which have lower timestamp are easy to implement as compared to transactions that are implanted later. Therefore, this techniques is consider more compatible than that of Tow-phase Lock technique as ordering avoids the deadlocks situation that is more difficult to handle it. In timestamp two-approaches are followed that is totally based on this techniques i.e. the first one is wait and no wait and the second one is wound-wait.

Integrated

It is known by every system you are following have some advantages and disadvantages. From all the above techniques two-phase locking and timestamp ordering based techniques have their own relative pros and cons. In order to avoid this DDBMS introduced the integrated approach which is combination of both techniques and implemented at the same of execution. This approach provides the heterogeneous database and proves from all disadvantages connected to each other.

6. Optimistic Techniques

In this technique, the synchronization is performed till the execution of transaction until it becomes delayed or termination. This approach is mostly used to get the desirable results and maximize the degree of concurrency of data execution. Moreover, approach is reliable as a less number of transactions are used to avoid the conflict of data with each other. The technique performs the synchronization process by completing four phases such like that Read Phase, Compute Phase, Validate Phase and Write Phase. One thing is more beneficial in optimistic technique that is submission operation are not delayed even the higher degree of concurrency is involved in data. This is more suitable for such application where conflict occur that most of exits in rare cases. Optimistic approach is originally based of timestamp-ordering techniques.

7. Conclusion:

In a distributed Database management system several techniques are used to implement and proposed the data to control the concurrency of database. Mostly two techniques are useful to solve the concurrency of database that is Two-phase Locking Techniques and Timestamp-Ordering based Technique. Two-phase Locking Technique is used by Pessimistic methods in which concurrency may lead with number of problems like concurrency of lower degree, deadlock, starvation and cascading rollbacks also. However, on the other hand the optimistic techniques are more reliable than that of pessimistic methods without occurrence of deadlock t increase the degree of concurrency of database. After taking a complete view in detailed form of both techniques to hold the concurrency of database now, we can easily distinguish the difference between both methods which are the best. Hence the Pessimistic approach is mostly used for the distributed database system whereas, the optimistic approach are not more suitable or used than that of pessimistic because this method has low ratio activity.