sponsors
usenix conference policies
Autonomic Cloud Workload Optimization: Placement in OpenStack
Half Day Afternoon
(1:30 pm-5:00 pm)
Washington AB
Description:
This tutorial opens the door for the ICAC audience to apply some of the autonomic computing ideas to the optimized deployment of workloads in the cloud. We have designed the tutorial to have two parts: (I) Overview of cloud management, OpenStack, Heat, and HOT technologies; and (II) Optimization algorithms for solving the large-scale placement problem of workloads in the cloud, in a scaleable manner. Part I acts as an introduction to the area for those who may be experts in autonomic computing, but are not quite familiar with the state-of-the-art of cloud management. And, part II should appeal to the theoreticians and application-oriented in the audience alike.
- Overview of cloud management (1.5 hrs):
- Overview of OpenStack open source cloud software
- Heat template-driven orchestration engine
- HOT: The Heat orchestration template
- Cloud workload definition
- Architecture of a workload placement engine
- End-to-end flow
- Workload Optimization (1.5 hrs)
- Definition of workload placement optimization problem
- Problem complexity and scalability
- Algorithmic approaches to placement optimization
- Examples and case studies
connect with us