Abstract

This paper presents Odyssey, a novel distributed data-series processing framework that efficiently addresses the critical challenges of exhibiting good speedup and ensuring high scalability in data series processing by taking advantage of the full computational capacity of modern clusters comprised of multi-core servers. Odyssey addresses a number of challenges in designing efficient and highlyscalable distributed data series index, including efficient scheduling, and load-balancing without paying the prohibitive cost of moving data around. It also supports a flexible partial replication scheme, which enables Odyssey to navigate through a fundamental trade-off between data scalability and good performance during query answering. Through a wide range of configurations and using several real and synthetic datasets, our experimental analysis demonstrates that Odyssey achieves its challenging goals.

Manos Chatzakis, Panagiota Fatourou, Eleftherios Kosmas, Themis Palpanas, Botao Peng. Odyssey: A Journey in the Land of Distributed Data Series Similarity Search. PVLDB 16(5), 2023

Data and Source Code

The datasets we have used in our paper are listed below:

• Seismic: contains seismic instrument recordings from stations around the world
• Astro: represents celestial objects
• Deep: contains deep embeddings originating from a convoltional neural network
• Sift: contains image (sift) descriptors
• Random: We produced a set of synthetic datasets with different sizes, composed by random walks of length 256. Each data point in the data series is produced as xi+1=N(xi,1), where N(0,1) is a standard normal distribution. The synthetic data generator code is included in the source code we are making available.

Here is the source code we used in the experimental evaluation of our paper.

Get Code