Hasil Pencarian  ::  Simpan CSV :: Kembali

"Di era digital dengan pertumbuhan data yang sangat pesat, pengelolaan database menjadi semakin penting untuk mendukung kinerja sistem secara efektif dan efisien. Optimasi database melalui tuning parameter menjadi sangat penting dalam menangani volume data yang kian besar dari tahun ke tahun. Akan tetapi, proses tuning manual akan sangat kompleks dan memakan waktu karena melibatkan banyak parameter yang saling mempengaruhi dengan nilai yang bersifat diskrit dan kontinyu. Untuk mengatasi hal tersebut, penelitian ini mengusulkan implementasi auto tuning pada database MongoDB, yang merupakan database populer untuk NoSQL. Algoritma Hybrid Firefly dan Particle Swarm Optimization (FAPSO) akan digunakan untuk melakukan tuning parameter pada database. Penelitian ini akan dilakukan dengan dua skema database yaitu single node mode dan cluster mode. Untuk mengevaluasi hasil kinerja algoritma tersebut, akan dilihat rata-rata latensi dan throughput yang akan dibandingkan dengan nilai hasil dari algoritma firefly, PSO, dan parameter default dari database. Hasil dari penelitian menunjukkan bahwa algoritma FAPSO memberikan performa terbaik dalam pengujian kueri find() baik pada single node dan cluster mode dengan rataan latensi lebih rendah berturut-turut 7,43% dan 4,16% dari parameter default, serta throughput tertinggi hingga 42548,94 ops/detik pada skema single node dan 38245,91 pada skema cluster. Pada pengujian kueri lookup(), FAPSO unggul dalam beban kerja yang lebih rendah untuk kedua skema basis data dengan rataan latensi lebih rendah 0,23% dari parameter default dan throughput hingga 12338,05 ops/detik pada skema single node.

---

In this digital era with rapid data growth, database management has become increasingly important to support system performance effectively and efficiently. Database optimization through parameter tuning is important in handling the growing volume of data year by year. However, manual tuning processes are very complex and time consuming, because they involve numerous interdependent parameters with both discrete and continuous values. To address this issue, this research proposes the implementation of auto-tuning on MongoDB, which is a popular NoSQL database. Hybrid Firefly and Particle Swarm Optimization (FA-PSO) algorithm will be used to perform parameter tuning on the database. This research will be conducted on two database schemes, single-node mode and cluster mode. To evaluate the performance of the algorithm, the average latency and throughput will be measured and compared with the firefly algorithm, PSO, and default parameters of the database without tuning. The results of this research show that the FAPSO algorithm gives the best performance in find() query testing in both single node and cluster mode with an average of latency of 7,43% and 4,16% respectively from the default parameters, as well as the highest throughput of up to 42548,94 ops/sec on single node scheme and 38245,91 on the cluster scheme. On lookup() query tests, FAPSO excels in lower workloads for both database schemas with an average latency of 0.23% lower than default parameters and a throughput of up to 12338,05 ops/sec on the single node scheme. "

"This book constitutes the thoroughly refereed proceedings of the 16th East-European Conference on Advances in Databases and Information Systems (ADBIS 2012), held in Poznan, Poland, in September 2012. The 32 revised full papers presented were carefully selected and reviewed from 122 submissions. The papers cover a wide spectrum of issues concerning the area of database and information systems, including database theory, database architectures, query languages, query processing and optimization, design methods, data integration, view selection, nearest-neighbor searching, analytical query processing, indexing and caching, concurrency control, distributed systems, data mining, data streams, ontology engineering, social networks, multi-agent systems, business process modeling, knowledge management, and application-oriented topics like RFID, XML, and data on the Web."

"This book constitutes the thoroughly refereed post-conference proceedings of the 27th British National Conference on Databases, BNCOD 27, held in Dundee, UK, in June 2010. The 10 revised full papers and 6 short papers, presented together with 3 invited papers, 1 best paper of the associated event on Teaching, Learning and Assessment of Databases (TLAD), and 2 PhD forum best papers were carefully reviewed and selected from 42 submissions. Special focus of the conference has been "Data Security and Security Data" and so the papers cover a wide range of topics such as data security, privacy and trust, security data, data integration and interoperability, data management for ubiquitous and mobile computing, data mining and information extraction, data modelling and architectures, data provenance, dataspaces, data streaming, databases and the grid, distributed information systems, electronic commerce, enterprise systems, heterogeneous databases, industrial applications, infrastructures and systems, intermittently connected data, file access methods and index structures, managing legacy data, new applications and processes, parallel and distributed databases, peer-to-peer data management, performance modelling of ubiquitous data use, personal data management, query and manipulation languages, query processing and optimisation, scientific applications, semantic Web and ontologies, semi-structured data, metadata and xml, user interfaces and data visualisation, web data management and deep Web, Web services, and workflow support systems."

"The volume contains 6 research papers, which have been carefully reviewed and selected from 12 submissions, plus the 3 keynotes presented at the workshop. The topics cover all stages of the business intelligence cycle, including capturing of real-time data, handling of temporal or uncertain data, performance issues, event management, and the optimization of complex ETL workflows."