distributed database has the characteristics of high scalability, high availability, low cost and performance improvement. How to build an appropriate data partition is the core problem for distributed database to sol...
详细信息
ISBN:
(纸本)9781665495080
distributed database has the characteristics of high scalability, high availability, low cost and performance improvement. How to build an appropriate data partition is the core problem for distributed database to solve the storage problem and improve the performance at the same time. By studying the data partition technology of distributed database, this paper gives the principles and design methods of database partition in distributed database design, and puts forward formulae of aggregation and balance that needs to be paid attention to in database partition, which provides a new idea for the transformation of database partition design from qualitative analysis to quantitative analysis. Moreover, why aggregation should be considered as one of the principles of data partitioning in distributed database is verified by experiments, and the application of database partition technology is illustrated by an example in telecom business support system.
Shared-nothing architecture has been widely used in distributed databases to achieve good scalability. While it offers superior performance for local transactions, the overhead of processing distributed transactions c...
详细信息
ISBN:
(纸本)9781450335317
Shared-nothing architecture has been widely used in distributed databases to achieve good scalability. While it offers superior performance for local transactions, the overhead of processing distributed transactions can degrade the system performance significantly. The key contributor to the degradation is the expensive two-phase commit (2PC) protocol used to ensure atomic commitment of distributed transactions. In this paper, we propose a transaction management scheme called LEAP to avoid the 2PC protocol within distributed transaction processing. Instead of processing a distributed transaction across multiple nodes, LEAP converts the distributed transaction into a local transaction. This benefits the processing locality and facilitates adaptive data repartitioning when there is a change in data access pattern. Based on LEAP, we develop an online transaction processing (OLTP) system, L-Store, and compare it with the state-of-the-art distributed in-memory OLTP system, H-Store, which relies on the 2PC protocol for distributed transaction processing, and HL-Store, a H-Store that has been modified to make use of LEAP. Results of an extensive experimental evaluation show that our LEAP-based engines are superior over H-Store by a wide margin, especially for workloads that exhibit locality-based data accesses.
This paper studies the construction and data mining of college sports database model. Aiming at the problems of low parallelism and unsystematic of current sports data processing, this paper proposes a mass of college...
详细信息
ISBN:
(纸本)9781450365772
This paper studies the construction and data mining of college sports database model. Aiming at the problems of low parallelism and unsystematic of current sports data processing, this paper proposes a mass of college sports data modeling and analysis method based on cloud computing to improve the management and analysis ability of college sports information, and constructs a distributed database model of college sports data. In the cloud computing environment, the database access model is designed, and the K-means data clustering method is used to mine the reliability of sports data, so as to realize the optimal information scheduling and retrieval analysis of college sports data. The simulation results show that the model has good real-time performance, high accuracy and reliability, and has a certain application prospect.
High performance low cost PC hardware, and high speed LAN/WAN technologies make distributed database(DDB) systems an attractive research area. Since Dynamic programming is not feasible for optimizing queries in a DDB,...
详细信息
ISBN:
(纸本)9781424450213
High performance low cost PC hardware, and high speed LAN/WAN technologies make distributed database(DDB) systems an attractive research area. Since Dynamic programming is not feasible for optimizing queries in a DDB, we propose a GA based query optimizer and compare its performance to random and optimal algorithms. We analyzed a set of possible GA parameters and determined that two-point truncate technique using GA gives the best results. New mutation and crossover operators have also been defined and experimentally analyzed. We performed experiments on a synthetic database with replicated relations, but no horizontal or vertical fragmentation. Network links are assumed to be gigabit Ethernet. Comparisons with optimal results found by exhaustive search show that our new GA formulation performs only 20% off the optimal results and we have achieved a 50% improvement over a previous GA based algorithm.
Relational database management systems (RDBMS) have served as the common choice for the storage of critical information in enterprises or organizations for decades. They are easily understood and applied, but traditio...
详细信息
ISBN:
(纸本)9781467383028
Relational database management systems (RDBMS) have served as the common choice for the storage of critical information in enterprises or organizations for decades. They are easily understood and applied, but traditional RDBMS exist some congenital shortcomings, such as expanding difficult. With the accelerating process of social informationization, the last few years have seen an explosive increase in the volume of data from various applications and they are likely to be stored in different sites. We have to face the uncomfortable fact that traditional RDBMS architectures have reached their limits and could no longer keep pace with the demand of modern society. In this paper, we introduce an innovative distributed database solution based on MySQL to meet the needs of horizontal scalability without affecting the specific implementation of application, which lets users think that it's a single MySQL database logically, actually being a distributed database system physically.
In this paper, we present an optimistic concurrency control solution. The proposed solution represents an excellent blossom in the concurrency control field. It deals with the concurrency control anomalies, and, simul...
详细信息
ISBN:
(纸本)9781467358255
In this paper, we present an optimistic concurrency control solution. The proposed solution represents an excellent blossom in the concurrency control field. It deals with the concurrency control anomalies, and, simultaneously, assures the reliability of the data before read-write transactions and after successfully committed. It can be used within the distributed database to track data logs and roll back processes to overcome distributed database anomalies. The method is based on commit timestamps for validation and an integer flag that is incremented each time a successful update on the record is committed. We show that this method overcomes inefficient taken methods (i.e. unnecessary restarts) in order to improve the performance and to increase productivity. Moreover, the method includes information about the time-stamping of transactions and flagging successfully updated data in the main-memory as it is proposed. Finally, the implemented work finds a need for an adaptive and an integrated concurrency control method in distributed database systems. Thus, a new optimistic concurrency control method is presented in this paper where it is expected to produce correct and reliable results.
distributed processing is an efficient way to improve performance of a database management system significantly. Distribution of data involves fragmentation, replication and allocation process. Previous research works...
详细信息
ISBN:
(纸本)9781467348362;9781467348331
distributed processing is an efficient way to improve performance of a database management system significantly. Distribution of data involves fragmentation, replication and allocation process. Previous research works provided fragmentation solution based on empirical data which are not applicable at the initial stage of a distributed database. In this paper we have presented a fragmentation technique that can be applied at the initial stage when no experimental data are present as well as in later stages of a distributed database system for partitioning the relations. Scalability of our proposed technique also investigated for different situation those may arise in practical cases of a distributed database. Experimental results show that our technique can solve initial fragmentation problem of distributed database system properly also compete with other non initial fragmentation techniques quite good in later stages.
In this paper, we propose an authorization model for distributed databases. Multiple object granularity of authorizations, such as global relations, fragments and attributes, are supported. Administrative privilege ca...
详细信息
ISBN:
(纸本)3540258787
In this paper, we propose an authorization model for distributed databases. Multiple object granularity of authorizations, such as global relations, fragments and attributes, are supported. Administrative privilege can be delegated from one subject to another to provide decentralized authorization administration. Authorization propagations along both the relation fragmentation tree and the subject group-subgroup hierarchical tree are also considered. Further more, conflict resolution policy is provided that supports well controlled delegations and exceptions. Overall the system provides a very flexible framework for specifying and evaluating the authorizations in distributed database systems.
Since the era of big data is coming, the first important problem is how to enhance the speed of database query. For the query optimization of distributed database, the speed of query depends on the data transfer and o...
详细信息
ISBN:
(纸本)9781467395878
Since the era of big data is coming, the first important problem is how to enhance the speed of database query. For the query optimization of distributed database, the speed of query depends on the data transfer and order of join. The cost model minimizing communication cost is the emphasis of research. Parallel Genetic Algorithm-Max-Min Ant System was proposed to seek a best query execution plan, which combines faster convergence of Genetic Algorithm, globally search ability of Max-Min Ant System and parallel property of both them. The experiment results show that the proposed algorithm is effective for query processing of multi-join, and plays important role in improving the performance of distributed database.
In a distributed database system, data replicas are placed at different locations to achieve high data availability in the presence of link failures. With majority voting protocol, a location is survived for read/writ...
详细信息
ISBN:
(纸本)9781424431748
In a distributed database system, data replicas are placed at different locations to achieve high data availability in the presence of link failures. With majority voting protocol, a location is survived for read/write operations if and only if it is accessible to more titan half of the replicas. The problem is to find out the optimal placements for a given number of data replicas in a ring network. When the number of replicas is odd, it was conjectured by Hu et al. that every uniform placement is optimal, which is proved by Shekhar and Wu later. However, when the number of replicas is even, it was pointed out by Hu el al. that uniform placements are not optimal and the optimal placement problem may be very complicated. In this paper we study the optimal placement problem in a ring network, with majority voting protocol and even number of replicas, and give a complete characterization of optimal placements when the number of replicas is not too large compared with the number of locations.
暂无评论