CloudTree: A Library to Extend Cloud Services for Trees

TL;DR

CloudTree is a cloud-based library enabling creation and management of tree data structures with performance optimizations, demonstrated through implementations of BST and Prefix Trees on Amazon Cloud.

Contribution

It introduces a new cloud service, CloudTree, supporting various tree structures with caching and prefetching to improve big data management.

Findings

CloudTree improves performance for big data applications.

BST and Prefix Tree implementations are efficient on Amazon Cloud.

The approach is adaptable to other data structures.

Abstract

In this work, we propose a library that enables on a cloud the creation and management of tree data structures from a cloud client. As a proof of concept, we implement a new cloud service CloudTree. With CloudTree, users are able to organize big data into tree data structures of their choice that are physically stored in a cloud. We use caching, prefetching, and aggregation techniques in the design and implementation of CloudTree to enhance performance. We have implemented the services of Binary Search Trees (BST) and Prefix Trees as current members in CloudTree and have benchmarked their performance using the Amazon Cloud. The idea and techniques in the design and implementation of a BST and prefix tree is generic and thus can also be used for other types of trees such as B-tree, and other link-based data structures such as linked lists and graphs. Preliminary experimental results show that CloudTree is useful and efficient for various big data applications.