CAC Research Program
Autonomic computing engines and applications
Consolidated and virtualized datacenter centers and clouds have become the dominant computing platforms in industry and research for enabling complex and compute-intensive applications. As scales, operating costs and energy requirements increase, however, maximizing efficiency, cost-effectiveness, and utilization of these systems becomes paramount. Furthermore, the complexity, dynamism, and often time-critical nature of application workloads makes on-demand scalability, integration of geographically distributed resources and incorporation of utility computing services extremely critical. Finally, the heterogeneity and dynamics of the system, application, and computing environment require context-aware dynamic scheduling and runtime management. CAC research focuses on developing autonomic computing and applications frameworks that can address these challenges and support a wide range of applications.
As the computing and communication capabilities of devices rapidly increases, device networks and peer-to-peer systems provide opportunities for enabling self-organizing distributed systems that are scalable and resilient to failures, and that provide value-added services such as automation of business processes. Security, quality-of-service (QoS), reliability and minimal power consumption are key factors for success of such systems. CAC research is this area is directed toward developing self-organizing networks that respond to dynamic network configurations while ensuring high performance, fault tolerance and security and integrating these networks with virtualization techniques to provide seamless peer connectivity to a wealth of existing, unmodified applications.
Autonomic defense systems are critical to the survivability of the information infrastructure that covers all aspects of our life. These systems of systems and their services must detect and protect against all types of network attacks, known or unknown. They require precise models of both attacks and normal behaviors which are hard to obtain due to sophisticated attacks and the extremely dynamic nature of the Internet.
With the rapid growth of servers and applications spurred by the Internet, the power consumption of servers has become critically important and must be efficiently managed. The EPA has recently reported that data centers consumed about 61 billion kilowatt-hours (kWh) in 2006 (1.5 percent of total U.S. electricity consumption) for a total electricity cost of about $4.5 billion. This estimated level of electricity consumption is more than the electricity consumed by the nation's color televisions and similar to the amount of electricity consumed by approximately 5.8 million average U.S. households (or about five percent of the total U.S. housing stock). CAC research in this area targets the development of a theoretical framework and a general methodology for autonomic power and performance management of high-performance distributed systems.
The explosive growth in scale and functionality of enterprise software systems and underlying IT infrastructures has resulted in complex systems whose control and timely management is rapidly exceeding human ability. Significant management challenges are resulting from, among other factors, their size, the complexity and distributed nature of their architectures, and the complexity and diversity of services provided. CAC research in this area is dedicated to the development and integration of autonomic techniques for monitoring, modeling, configuring, controlling, and optimizing the behaviors of applications, services, and resources, to ensure their robust and resilient operation in the face of these challenges.
Virtualized data centers
IT management, power consumption, and cooling costs make up an increasingly significant percentage of the overall cost of operating large data centers. These centers are used by banks, investment firms, IT service providers, and other large enterprises. Virtual machines provide a layer that is ideally positioned to provide fine-grained resource monitoring and control capabilities that are well-suited for the monitoring and execution phases in an autonomic computing framework. CAC is working to develop self-monitoring and self-optimization techniques which, when applied to the management of virtualized containers, can lead to increased efficiencies in resource utilization and reduced costs of provisioning.