B
Björn Þór Jónsson
Researcher at IT University of Copenhagen
Publications - 65
Citations - 1446
Björn Þór Jónsson is an academic researcher from IT University of Copenhagen. The author has contributed to research in topics: Interactive Learning & Search engine indexing. The author has an hindex of 13, co-authored 64 publications receiving 1279 citations. Previous affiliations of Björn Þór Jónsson include Centre national de la recherche scientifique & Reykjavík University.
Papers
More filters
Proceedings Article
Semantic Data Caching and Replacement
TL;DR: A semantic model for client-side caching and replacement in a client-server database system and compared to page caching and tuple caching strategies is proposed and validated with a detailed performance study.
Proceedings Article
uFLIP: Understanding Flash IO Patterns
TL;DR: In this article, the authors focus on flash IO patterns, that capture relevant distribution of IOs in time and space, and their goal is to quantify their performance, and define uFLIP, a benchmark for measuring the response time of flash IO pattern.
Journal ArticleDOI
Performance and overhead of semantic cache management
Björn Þór Jónsson,Maria Arinbjarnar,Bjarnsteinn Þórsson,Michael J. Franklin,Divesh Srivastava +4 more
TL;DR: This article revisits the performance and overhead of semantic caching using a modern database server and modern hardware and demonstrates that semantic caching works well in a range of applications, especially in network-constrained environments.
Proceedings ArticleDOI
NV-Tree: nearest neighbors at the billion scale
TL;DR: The NV-Tree (Nearest Vector Tree) addresses the specific, yet important, problem of efficiently and effectively finding the approximate k-nearest neighbors within a collection of a few billion high-dimensional data points.
Proceedings ArticleDOI
Introduction to the Fourth Annual Lifelog Search Challenge, LSC'21
Cathal Gurrin,Björn Þór Jónsson,Klaus Schöffmann,Duc-Tien Dang-Nguyen,Jakub Lokoč,Minh-Triet Tran,Wolfgang Hürst,Luca Rossetto,Graham Healy +8 more
TL;DR: An overview of the challenge motivates the challenge, the dataset and system configuration used in the challenge are presented, and the participating teams are presented.