A Hybrid Concurrency Control Strategy for Mobile Database Systems

Moiz, Salman Abdul

doi:10.9734/air/2015/12406

Cited by 2 publications

(7 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Scientists as early as 1981, such as Bernstein and Goodman (1981), for example, introduced and discussed more than 48 principle methods and 20 concurrency control algorithms. Recent work regarding various algorithms, methods, and strategies targeting concurrency control issues can be found in a wide range of research such as Moiz (2015), Akintola et al (2005), Bakura and Mohammed (2014), Batra and Kapil (2010), Haapasalo et al (2008), and Herlihy and Weihl (1991). Works such as Batra andKapil (2010), Carey andLivny (1988), Geschwent (1994), and Kanungo and Rustom (2015) conducted surveys of concurrency algorithms and compared them against various parameters.…”

Section: Related Workmentioning

confidence: 99%

A two-part multi-algorithm concurrency control optimization strategy for distributed database systems

Shebka¹

2022

Int. j. adv. appl. sci.

View full text Add to dashboard Cite

In this paper, we propose a novel holistic approach to address the issues of concurrency control after an exhaustive examination of the problem and the various forms it can transpire. The proposed strategy was formulated depending on different perspectives that are based on exploring a wide range of algorithms, methods, and strategies proposed in practice and theory that attempted to address the problem and its forms, but only partially succeeded in doing so. Here we proposed a two-part holistic strategy to optimize concurrency control in distributed environments that address a wide range of concurrency control anomalies by taking advantage of several concurrency control algorithms' strengths while minimizing their weaknesses. The novelty of our approach transpires from two interconnected parts that can be applied regardless of the type of distributed database environment. The first is a structured tier-based data classification system based on data sensitivity with respect to serializability requirements and ranges from strict to very relaxed forms of serializability constraints. The second is a concurrency management algorithm that allocates the appropriate concurrency control algorithm to each transaction depending on the type of transaction and/or type of data being accessed from the aforementioned tier-based classification method. Our proposed method also incorporates a priority allocation mechanism within the concurrency management algorithm. Priority is allocated to different tier transactions depending on the tier's level, which in turn reflects data importance and sensitivity. Although our proposed strategy remains an algorithmic approach as we encountered various challenges regarding performance testing of a novel multi-algorithm approach for handling concurrency control in distributed database systems. However, future work involves testing the performance of our proposed strategy either through real-time systems after considerable adjustments or by constructing an appropriate customized simulation framework. Finally, the potentials of the strategy presented here are very promising, hence, we recommend as we are also optimistic that other scholars are encouraged to further exploit the concept of using multiple concurrency control algorithms within the same distributed database environment.

show abstract

Section: Related Workmentioning

confidence: 99%

A two-part multi-algorithm concurrency control optimization strategy for distributed database systems

Shebka¹

2022

Int. j. adv. appl. sci.

View full text Add to dashboard Cite

show abstract

“…For each transaction model, discussion of the literature has been organized according to the basic types of concurrency control protocol (as discussed before). Mobile concurrency control 16 54,61-72-74 Hybrid concurrency control 9 43,48,50,[75][76][77][78][79][80] To perform an effective, detailed, and realistic comparison, we focused on research contributions made with respect to real time and mobile transactions for the period 2000-2021. We discussed the most significant research contributions and also performed a detailed comparative analysis.…”

Section: Landscape Of Literature On Concurrency Controlmentioning

confidence: 99%

“…Deadlock is a very common issue that occur in concurrency control protocols. The protocols 31,36,43,51,52,57,61,62,67,68,75,76,80 do not have a deadlock issue. The other protocols have not been tested for deadlocks.…”

Section: Comparison Wrt Other Featuresmentioning

confidence: 99%

“…The other protocols have not been tested for deadlocks. The protocols 57,65,80 avoid starvation issue. There is a need to review the rest of the protocols to see if they can avoid starvation or not.…”

Section: Comparison Wrt Other Featuresmentioning

confidence: 99%

“…A transaction is starving if it is repeatedly aborted and fails to commit. 43,65,80 This problem is quite probable in optimistic concurrency control methods. The research work on starvation is done with respect to two view-points, starvation detection and prevention.…”

Section: Problem Of Starving Transactions and Cyclic Restartsmentioning

confidence: 99%

See 2 more Smart Citations

Concurrency control for real‐time and mobile transactions: Historical view, challenges, and evolution of practices

Chaudhry

Yousaf

2021

Concurrency and Computation

View full text Add to dashboard Cite

Recently, mobile computing has changed the way that spatial data and GIS are processed. Unlike wired and stand-alone GIS, now the trend has been switched from offline to real-time data processing using location aware services, such as GPS technology. The increased usage of location aware services in multiuser real-time environment has made transaction management incredibly significant. If the simultaneous query operations on the same data item are not handled intelligently then this results in data inconsistency issue. Concurrency control protocol is one of the primary aspects that helps in overcoming this issue a in multiuser environment. To the best of our knowledge, the impact of technological advancements on concurrency control has not been thoroughly studied in the literature. In this article, we explored the literature on concurrency control algorithms in depth with respect to real-time applications and the applications with moving objects. We defined a taxonomy of concurrency control solutions and assessed the maturity of these solutions in the light of characteristics of real-time and mobile environment. We compared the most recent developments made in the literature and presented meaningful insights. Challenges are also identified and discussed, which can assist in doing research in this domain in future. K E Y W O R D S concurrency control, mobile, real time 1 INTRODUCTION IoT (Internet of Things) emerged as a promising technology that has revolutionized the way of living. GPS-based devices and sensors are commonly used in majority of the applications. The convergence of multiple domains such as real-time GIS, mobile computing, and IoT has made it necessary to solve the computing problem considering the requirements of each domain. Mobile GIS domain integrates several technologies: GPS, mobile computing, and GIS. In recent times, the data processing environment deals with varying kinds of sensors. The real-time data collected by sensors need to be processed in timely manner to ensure efficient data management. The sensors can be categorized into two types depending on their mobility: fixed and agile/mobile sensors. Fixed sensors are spatially fixed, whereas mobile sensors change their location with respect to time. Mobile sensors collect data on geographical grounds and this data can be referred to as spatiotemporal data. This has given rise to a new research domain of real-time mobile data processing and management. Thus, real-time processing domain emerged with respect to time depending on the nature of sensor technology and evolution of IoT. The future processing systems are expected to meet the requirements of spatiotemporal and real-time data, leading toward a new domain of real-time GIS. Real-time data can be used for both static and dynamic processing. For example, in the first aspect, mobile sensors can process and analyze the real-time spatiotemporal data for dynamic processing. On the other hand, real-time data can be collected and saved in a log for future offline analysis (static processing).

show abstract

A Hybrid Concurrency Control Strategy for Mobile Database Systems

Cited by 2 publications

References 8 publications

A two-part multi-algorithm concurrency control optimization strategy for distributed database systems

A two-part multi-algorithm concurrency control optimization strategy for distributed database systems

Concurrency control for real‐time and mobile transactions: Historical view, challenges, and evolution of practices

Contact Info

Product

Resources

About