Machine learning‐based load distribution and balancing in heterogeneous database management systems

2021 ◽  
Author(s):  
Anes Abdennebi ◽  
Anıl Elakaş ◽  
Fatih Taşyaran ◽  
Erdinç Öztürk ◽  
Kamer Kaya ◽  
...  
Keyword(s):  
Machine Learning ◽  
Load Distribution ◽  
Database Management ◽  
Database Management Systems ◽  
Management Systems ◽  
Heterogeneous Database

An inquiry into machine learning-based automatic configuration tuning services on real-world database management systems

2021 ◽  
Vol 14 (7) ◽  
pp. 1241-1253
Author(s):  
Dana Van Aken ◽  
Dongsheng Yang ◽  
Sebastien Brillard ◽  
Ari Fiorino ◽  
Bohan Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Database Management ◽  
Database Management Systems ◽  
Management Systems ◽  
Production Environment ◽  
Considerable Effort ◽  
Behavior Setting ◽  
Grade Configurations ◽  
Tuning Methods ◽  
Automated Tuning

Modern database management systems (DBMS) expose dozens of configurable knobs that control their runtime behavior. Setting these knobs correctly for an application's workload can improve the performance and efficiency of the DBMS. But because of their complexity, tuning a DBMS often requires considerable effort from experienced database administrators (DBAs). Recent work on automated tuning methods using machine learning (ML) have shown to achieve better performance compared with expert DBAs. These ML-based methods, however, were evaluated on synthetic workloads with limited tuning opportunities, and thus it is unknown whether they provide the same benefit in a production environment. To better understand ML-based tuning, we conducted a thorough evaluation of ML-based DBMS knob tuning methods on an enterprise database application. We use the OtterTune tuning service to compare three state-of-the-art ML algorithms on an Oracle installation with a real workload trace. Our results with OtterTune show that these algorithms generate knob configurations that improve performance by 45% over enterprise-grade configurations. We also identify deployment and measurement issues that were overlooked by previous research in automated DBMS tuning services.

scholarly journals Machine Learning for Database Management Systems

2020 ◽  
Vol 9 (08) ◽  
pp. 25132-25147
Author(s):  
Sai Tanishq N
Keyword(s):  
Machine Learning ◽  
Problem Solving ◽  
Database Management ◽  
Database Management Systems ◽  
Management Systems ◽  
Complex Problems ◽  
Prediction Algorithms ◽  
The World ◽  
Data Explosion

Machine Learning (ML) is transforming the world with research breakthroughs that are leading to the progress of every field. We are living in an era of data explosion. This further improves the output as data that can be fed to the models is more than it has ever been. Therefore, prediction algorithms are now capable of solving many of the complex problems that we face by leveraging the power of data. The models are capable of correlating a dataset and its features with an accuracy that humans fail to achieve. Bearing this in mind, this research takes an in-depth look into the of the problem- solving potential of ML in the area of Database Management Systems (DBMS). Although ML hallmarks significant scientific milestones, the field is still in its infancy. The limitations of ML models are also studied in this paper.

Sign in / Sign up

Export Citation Format

Share Document