Provided by:
National University of Singapore
Topic:
Storage
Format:
PDF
Given the explosion of big data analytics, it is important to understand the performance costs and limitations of existing approaches for in-memory data management. Broadly, in-memory data management covers two main types of roles: supporting analytics operations such as iterative algorithms and supporting storage and retrieval operations on arbitrary key-value objects. This paper focuses on three such popular systems: Memcached, Redis and RDD and proposes a thorough performance analysis of both analytics and key-value object operations.
