scholarly journals Automatic configuration of the Cassandra database using irace

2021 ◽  
Vol 7 ◽  
pp. e634
Author(s):  
Moisés Silva-Muñoz ◽  
Alberto Franzin ◽  
Hugues Bersini
Keyword(s):  
Simple Model ◽  
Statistical Models ◽  
Database Systems ◽  
General Purpose ◽  
Experimental Setup ◽  
Model Free ◽  
Nosql Database ◽  
Parameter Configuration ◽  
Key Factor ◽  
Best Parameter

Database systems play a central role in modern data-centered applications. Their performance is thus a key factor in the efficiency of data processing pipelines. Modern database systems expose several parameters that users and database administrators can configure to tailor the database settings to the specific application considered. While this task has traditionally been performed manually, in the last years several methods have been proposed to automatically find the best parameter configuration for a database. Many of these methods, however, use statistical models that require high amounts of data and fail to represent all the factors that impact the performance of a database, or implement complex algorithmic solutions. In this work we study the potential of a simple model-free general-purpose configuration tool to automatically find the best parameter configuration of a database. We use the irace configurator to automatically find the best parameter configuration for the Cassandra NoSQL database using the YCBS benchmark under different scenarios. We establish a reliable experimental setup and obtain speedups of up to 30% over the default configuration in terms of throughput, and we provide an analysis of the configurations obtained.

scholarly journals Performance Benchmarking of Key-Value Store NoSQL Databases

2018 ◽  
Vol 8 (6) ◽  
pp. 5333
Author(s):  
Omoruyi Osemwegie ◽  
Kennedy Okokpujie ◽  
Nsikan Nkordeh ◽  
Charles Ndujiuba ◽  
Samuel John ◽  
...  
Keyword(s):  
Data Storage ◽  
Web Applications ◽  
Query Language ◽  
Research Work ◽  
Database Systems ◽  
Nosql Databases ◽  
Performance Benchmarking ◽  
Nosql Database ◽  
Structured Query Language ◽  
Web Developers

<p>Increasing requirements for scalability and elasticity of data storage for web applications has made Not Structured Query Language NoSQL databases more invaluable to web developers. One of such NoSQL Database solutions is Redis. A budding alternative to Redis database is the SSDB database, which is also a key-value store but is disk-based. The aim of this research work is to benchmark both databases (Redis and SSDB) using the Yahoo Cloud Serving Benchmark (YCSB). YCSB is a platform that has been used to compare and benchmark similar NoSQL database systems. Both databases were given variable workloads to identify the throughput of all given operations. The results obtained shows that SSDB gives a better throughput for majority of operations to Redis’s performance.</p>

Fragment Re-Allocation Strategy Based on Hypergraph for NoSQL Database Systems

2016 ◽  
Vol 8 (3) ◽  
pp. 1-23 ◽  
Author(s):  
Zhikun Chen ◽  
Shuqiang Yang ◽  
Yunfei Shang ◽  
Yong Liu ◽  
Feng Wang ◽  
...  
Keyword(s):  
Database Systems ◽  
Communication Cost ◽  
Allocation Strategy ◽  
Hypergraph Partitioning ◽  
Computing Model ◽  
Operation Pattern ◽  
Nosql Database ◽  
High Scalability ◽  
Weighted Hypergraph ◽  
Allocation Strategies

NoSQL database is famed for the characteristics of high scalability, high availability, and high fault-tolerance. It is used to manage data for a lot of applications. The computing model has been transferred to “computing close to data”. Therefore, the location of fragment directly affects system's performance. Every site's load dynamical changes because of the increasing data and the ever-changing operation pattern. So system has to re-allocate fragment to improve system's performance. The general fragment re-allocation strategies of NoSQL database scatter the related fragments as possible to improve the operations' parallel degree. But those fragments may interact with each other in some application's operations. So the high parallel degree of operation may increase system's communication cost such as data are transferred by network. In this paper, the authors propose a fragment re-allocation strategy based on hypergraph. This strategy uses a weighted hypergraph to represent the fragments' access pattern of operations. A hypergraph partitioning algorithm is used to cluster fragments in the strategy. This strategy can improve system's performance according to reducing the communication cost while guaranteeing the parallel degree of operations. Experimental results confirm that the strategy will effectively contribute in solving fragment re-allocation problem in specific application environment of NoSQL database system, and it can improve system's performance.

scholarly journals Preconditioning for Sparse Linear Systems at the Dawn of the 21st Century: History, Current Developments, and Future Perspectives

2012 ◽  
Vol 2012 ◽  
pp. 1-49 ◽  
Author(s):  
Massimiliano Ferronato
Keyword(s):  
Linear Systems ◽  
Specific Problem ◽  
General Purpose ◽  
Sparse Linear Systems ◽  
System Matrix ◽  
Preconditioning Techniques ◽  
Key Factor ◽  
Recent Developments ◽  
Approximate Inverses ◽  
Linear Systems Of Equations

Iterative methods are currently the solvers of choice for large sparse linear systems of equations. However, it is well known that the key factor for accelerating, or even allowing for, convergence is the preconditioner. The research on preconditioning techniques has characterized the last two decades. Nowadays, there are a number of different options to be considered when choosing the most appropriate preconditioner for the specific problem at hand. The present work provides an overview of the most popular algorithms available today, emphasizing the respective merits and limitations. The overview is restricted to algebraic preconditioners, that is, general-purpose algorithms requiring the knowledge of the system matrix only, independently of the specific problem it arises from. Along with the traditional distinction between incomplete factorizations and approximate inverses, the most recent developments are considered, including the scalable multigrid and parallel approaches which represent the current frontier of research. A separate section devoted to saddle-point problems, which arise in many different applications, closes the paper.

Sign in / Sign up

Export Citation Format

Share Document