VU-ISIS Team Wins Best Paper Award at IC2E 2019

The best paper award at the recently concluded 9th IEEE International Conference on Cloud Engineering (IC2E) Conference, held in Prague, Czech Republic, was awarded to researchers from ISIS/Vanderbilt University. The title of the paper is "FECBench: A Holistic Interference-aware Approach for Application Performance Modeling" and was authored by Yogesh Barve, Dr. Shashank Shekhar, Ajay Chhokra, Shweta Khare, Anirban Bhattacharjee, Zhuangwei Kang, Dr. Hongyang Sun, and Dr. Aniruddha Gokhale. The IC2E Best Paper Award recognizes the paper that exhibits key contributions in the field of cloud computing and cloud engineering. The best paper is nominated by the IEEE IC2E organization committee.

Software services hosted in multi-tenant cloud platforms often encounter performance interference issue due to contention for non-partitionable resources.This in turn causes unpredictable behavior and degradation in software service's performance. FECBench (Fog/Edge/Cloud Benchmarking), which is an open source framework comprises of 106 applications covering a wide range of application classes. FECBench guides providers and users in building performance interference prediction models for their services without incurring undue costs and efforts. The FECBench paper makes the following concrete research contributions. First, it presents a technique to build resource stressors that can stress multiple system resources all at once in a controlled manner. This helps to understand and gain insights into the impact of interference on an application’s performance. Second, to overcome the need for exhaustive application profiling, FECBench leverages the design of experiments (DoE) approach to enable users to construct surrogate performance interference models of their services. Third, FECBench builds an extensible knowledge base of different application combinations that generate resource stresses across the multidimensional resource design space.

Yogesh Barve is a final year PhD. candidate at the Institute for Software Integrated Systems (ISIS) in the Department of EECS at Vanderbilt. His research focuses on performance engineering for multi-tenant systems, model-driven engineering and distributed simulations. He is advised by Prof. Aniruddha Gokhale and Prof. Janos Sztipanovits. Prof. Aniruddha Gokhale, Professor of Computer Science and Computer Engineering, leads the Distributed Object Computing (DOC) group. The DOC group’s research areas comprise dynamic resource management for Cloud Computing, Internet of Things/Cyber Physical Systems, Software-defined Networking, and Model-driven Engineering. Prof. Sztipanovits is the Director of the Institute for Software Integrated Systems.