The Truth About MapReduce Performance on SSDs
Offered By: USENIX via YouTube
Course Description
Overview
Explore the performance implications of using solid-state drives (SSDs) for MapReduce workloads in this 14-minute conference talk from USENIX LISA14. Discover a method for benchmarking MapReduce performance on SSDs and HDDs under constant-bandwidth constraints. Learn why cost-per-performance is a more relevant metric than cost-per-capacity when evaluating SSDs versus HDDs for performance-oriented tasks. Gain insights into the potential for SSDs to achieve up to 70% higher performance at 2.5x higher cost-per-performance compared to traditional HDDs. Presented by Karthik Kambatla from Cloudera Inc. and Purdue University, and Yanpei Chen from Cloudera Inc., this talk provides valuable information for those considering the adoption of SSDs in MapReduce environments.
Syllabus
LISA14 - The Truth About MapReduce Performance on SSDs
Taught by
USENIX
Related Courses
Intro to Data ScienceUdacity Mining Massive Datasets
Stanford University via edX Cloud Computing Concepts, Part 1
University of Illinois at Urbana-Champaign via Coursera Data Manipulation at Scale: Systems and Algorithms
University of Washington via Coursera Big Data Analytics in Healthcare
Georgia Institute of Technology via edX