Application Performance Management in the Cloud using Learning, Optimization, and Control
Abstract:
Many businesses and organizations are increasingly relying on cloud based infrastructures and platforms to deliver their business-critical applications. In the meantime, a recent study shows that 79% of companies are concerned about the hidden costs of cloud services for their applications, citing “poor end-user experience due to performance bottlenecks” as their top management concern in relationship to cloud services.Existing practices in application performance management rely heavily on white-box modeling or heuristics-based, manual diagnostic approaches to find potential bottlenecks and remediation steps. However, the scalability and adaptivity of such approaches remain severely constrained, especially in a highly-dynamic, consolidated cloud environment. These challenges present unique opportunities in applying statistical learning, control, and optimization based techniques to developing model-based, automated application performance management frameworks. There has been a large body of research in this area in the last several years, but many problems remain. In this talk, I'll highlight some of the performance and resource management techniques we have developed within VMware, along with related technical challenges, and discuss open research problems, in hope to attract more innovative ideas and solutions from a larger community of researchers and developers.
Biography:
Xiaoyun Zhu is a Staff Engineer in the VMware Cloud Resource Management Group, focusing on developing automated resource and performance management solutions for virtualized datacenters and applications. Her general interests are in applying optimization, algorithms, statistical learning and control theory to IT systems management and automation. Prior to VMware, she was a Senior Research Scientist at HP Labs for eight years. Techniques and algorithms she developed had been incorporated into HP’s management products including HP Global Workload Manager and HP Integrity Essentials Capacity Advisor. She has co-authored over 50 refereed papers in journals and conference proceedings, and holds 20 patents. She has been a program committee member for IM, NOMS, CNSM, CCGrid, Middleware, ICDCS, MASCOTS, and SIGMETRICS. She was the program co-chair for ICAC 2013 and is the general chair for ICAC 2014. She co-founded the International Workshop on Feedback Computing in 2006 (formerly known as FeBID). Xiaoyun received her dual B.S. in Automation and Applied Mathematics from Tsinghua University, and her Ph.D. in Electrical Engineering from California Institute of Technology.