Performance Evaluation of a Data Lake Architecture via Modeling Techniques

Enrico Barbierato, Marco Gribaudo, Giuseppe Serazzi, Letizia Tanca

Risultato della ricerca: Contributo in libroContributo a convegno

Abstract

Data Lake is a term denoting a repository storing heterogeneous data, both structured and unstructured, resulting in a flexible organization that allows Data Lake users to reorganize and integrate dynamically the information they need according to the required query or analysis. The success of its implementation depends on many factors, notably the distributed storage, the kind of media deployed, the data access protocols and the network used. However, flaws in the design might become evident only in a later phase of the system development, causing significant delays in complex projects. This article presents an application of queuing networks modeling technique to detect significant issues, such as bottlenecks and performance degradation, for different workload scenarios.
Lingua originaleEnglish
Titolo della pubblicazione ospiteEuropean Workshop on Performance Engineering International Conference on Analytical and Stochastic Modeling Techniques and Applications
Pagine115-130
Numero di pagine16
Volume13104
DOI
Stato di pubblicazionePubblicato - 2021
Evento17th European Performance Engineering Workshop, EPEW 2021, and the 26th International Conference on Analytical and Stochastic Modelling Techniques and Applications, ASMTA 2021 - Tokio
Durata: 9 dic 202110 dic 2021

Workshop

Workshop17th European Performance Engineering Workshop, EPEW 2021, and the 26th International Conference on Analytical and Stochastic Modelling Techniques and Applications, ASMTA 2021
CittàTokio
Periodo9/12/2110/12/21

Keywords

  • Data lake
  • Queuing networks
  • JMT

Fingerprint

Entra nei temi di ricerca di 'Performance Evaluation of a Data Lake Architecture via Modeling Techniques'. Insieme formano una fingerprint unica.

Cita questo