James E. Hanson

Power and Performance Management of Virtualized Computing Environments Via Lookahead Control

There is growing incentive to reduce the power consumed by large-scale data centers that host online services such as banking, retail commerce, and gaming. Virtualization is a promising approach to consolidating multiple online services onto a smaller number of computing resources. A virtualized server environment allows computing resources to be shared among multiple performance-isolated platforms called virtual machines. By dynamically provisioning virtual machines, consolidating the workload, and turning servers on and off as needed, data center operators can maintain the desired quality-of-service (QoS) while achieving higher server utilization and energy efficiency. We implement and validate a dynamic resource provisioning framework for virtualized server environments wherein the provisioning problem is posed as one of sequential optimization under uncertainty and solved using a lookahead control scheme. The proposed approach accounts for the switching costs incurred while provisioning virtual machines and explicitly encodes the corresponding risk in the optimization problem. Experiments using the Trade6 enterprise application show that a server cluster managed by the controller conserves, on average, 26% of the power required by a system without dynamic control while still maintaining QoS goals.

An architectural approach to autonomic computing

We describe an architectural approach to achieving the goals of autonomic computing. The architecture that we outline describes interfaces and behavioral requirements for individual system components, describes how interactions among components are established, and recommends design patterns that engender the desired system-level properties of self-configuration, self-optimization, self-healing and self-protection. We have validated many of these ideas in two prototype autonomic computing systems.

Dynamic pricing by software agents

Abstract We envision a future in which the global economy and the Internet will merge, evolving into an information economy bustling with billions of economically motivated software agents that exchange information goods and services with humans and other agents. Economic software agents will differ in important ways from their human counterparts, and these differences may have significant beneficial or harmful effects upon the global economy. It is therefore important to consider the economic incentives and behaviors of economic software agents, and to use every available means to anticipate their collective interactions. We survey research conducted by the Information Economies group at IBM Research aimed at understanding collective interactions among agents that dynamically price information goods or services. In particular, we study the potential impact of widespread shopbot usage on prices, the price dynamics that may ensue from various mixtures of automated pricing agents (or “pricebots”), the potential use of machine-learning algorithms to improve profits, and more generally the interplay among learning, optimization, and dynamics in agent-based information economies. These studies illustrate both beneficial and harmful collective behaviors that can arise in such systems, suggest possible cures for some of the undesired phenomena, and raise fundamental theoretical issues, particularly in the realms of multi-agent learning and dynamic optimization.

Conversation support for business process integration

Business process integration and automation (BPIA) has emerged as an important aspect of the enterprise computing landscape. Intra-enterprise application integration (EAI) as well as the inter enterprise integration (B2B) are increasingly being performed in the context of business processes. The integration scenarios typically involve distributed systems that are autonomous to some degree. From the BPIA perspective, the autonomy refers to the fact that the systems being integrated have their own process choreography engines and execute internal business processes that are private to them. In the case of B2B integration, the systems being integrated are fully autonomous, while various levels of autonomy exist in systems partaking in EAI. We present a new paradigm for business process integration. Our approach is based on a conversation model that enables autonomous, distributed BPM (Business Process Management) modules to integrate and collaborate. Our conversation model supports the exchange of multiple correlated messages with arbitrary sequencing constraints and covers the formatting of messages that are to be sent as well as the parsing of the messages that have been received. The crux of our conversation model is the notion of a conversation policy, which is a machine-readable specification of a pattern of message exchange in a conversation. Our model supports nesting and composition of conversation policies to provide a dynamic, adaptable, incremental, open-ended, and extensible mechanism for business process integration. We discuss the current implementation of this conversation model and early experience in applying the model to solve customer problems. The implementation utilizes distributed object technology.

Dynamics of an Information-Filtering Economy

Our overall goal is to characterize and understand the dynamic behavior of information economies: very large open economies of automated information agents that are likely to come into existence on the Internet. Here we model a simple information-filtering economy in which broker agents sell selected articles to a subscribed set of consumers. Analysis and simulation of this model reveal the existence of both desirable and undesirable phenomena, and give some insight into their nature and the conditions under which they occur. In particular, efficient self-organization of the broker population into specialized niches can occur when communication and processing costs are neither too high nor too low, but endless price wars can undermine this desirable state of affairs.

Coordinated management of power usage and runtime performance

With the continued growth of computing power and reduction in physical size of enterprise servers, the need for actively managing electrical power usage in large datacenters is becoming ever more pressing. By far the greatest savings in electrical power can be effected by dynamically consolidating workload onto the minimum number of servers needed at a given time and powering off the remainder. However, simple schemes for achieving this goal fail to cope with the complexities of realistic usage scenarios. In this paper we present a combined power-and performance-management system that builds on a state-of-the-art performance manager to achieve significant power savings without unacceptable loss of performance. In our system, the degree to which performance may be traded off against power is itself adjustable using a small number of easily-understood parameters, permitting administrators in different facilities to select the optimal tradeoff for their needs. We characterize the power saved, the effects of the tradeoff between power and performance, and the changes in behavior as the tradeoff parameters are adjusted, both in simulation and in a sample deployment of the real system.

Runtime Demand Estimation for effective dynamic resource management

Systems management techniques that allocate resources to running entities, such as processes and virtual machines (VMs), often require estimates of the resources required by each of these resource consumers. For example, many proposed virtual machine placement algorithms attempt to allocate VMs to physical hosts in such a way as to minimize the number of physical hosts that are occupied, while ensuring that each VM receives the CPU required to do its task adequately. The common practice is to assume that the CPU requirement is equal to the current CPU utilization, or to use a prediction of it over an appropriate time horizon. In this paper, we demonstrate that, when multiple VMs or processes co-reside on a physical host, the measured CPU utilization may provide a poor estimate of the actual requirement. We derive a simple, much more accurate alternative estimate of CPU demand, implement it, and demonstrate its superiority experimentally. Furthermore, we demonstrate that using our demand estimation framework in conjunction with dynamic resource allocation in a virtualized environment greatly improves the effectiveness of dynamic placement, resulting in one-shot convergence to optimal placement and significant improvements in the overall performance of the individual VMs.

Agile, efficient virtualization power management with low-latency server power states

One of the main driving forces of the growing adoption of virtualization is its dramatic simplification of the provisioning and dynamic management of IT resources. By decoupling running entities from the underlying physical resources, and by providing easy-to-use controls to allocate, deallocate and migrate virtual machines (VMs) across physical boundaries, virtualization opens up new opportunities for improving overall system resource use and power efficiency. While a range of techniques for dynamic, distributed resource management of virtualized systems have been proposed and have seen their widespread adoption in enterprise systems, similar techniques for dynamic power management have seen limited acceptance. The main barrier to dynamic, power-aware virtualization management stems not from the limitations of virtualization, but rather from the underlying physical systems; and in particular, the high latency and energy cost of power state change actions suited for virtualization power management. In this work, we first explore the feasibility of low-latency power states for enterprise server systems and demonstrate, with real prototypes, their quantitative energy-performance trade offs compared to traditional server power states. Then, we demonstrate an end-to-end power-aware virtualization management solution leveraging these states, and evaluate the dramatically-favorable power-performance characteristics achievable with such systems. We present, via both real system implementations and scale-out simulations, that virtualization power management with low-latency server power states can achieve comparable overheads as base distributed resource management in virtualized systems, and thus can benefit from the same level of adoption, while delivering close to energy-proportional power efficiency.

Conversation-oriented protocols for contract negotiations

The expression of contracts in computer readable form, and the development of automated tests for completeness and well-formedness of contracts, has opened the door to significant advances in automating contract negotiation. To meet the needs of automation, such negotiations must follow explicitly specified message-exchange protocols. But to meet the needs of the negotiating parties, these protocols must be independent of the decision-making processes driving them as well as neutral to the outcome of the negotiations. In this paper we illustrate how both needs may be simultaneously met by a small set of conversation policies employed within a general purpose conversation support architecture.

Multi-aspect hardware management in enterprise server consolidation

An autonomic manager for enterprise server hardware management, called AMP, is described. AMP is designed to handle multiple aspects of hardware management and to work in conjunction with other management components, in particular application managers, in a way that reduces energy waste, protects server health, and preserves a high degree of autonomy both for itself and for the managers with which it works. AMP interacts with other managers in two ways: (1) exchange of nominal control over individual servers; and (2) provision of a synthetic cost function giving AMPs assessment of relative desirability of using different servers. The high-level architecture of AMP is discussed, with particular focus on the way it effects a natural decomposition of the combined hardware-and-application management problem, and on initial versions of the algorithms it uses to manage server power states and determine the cost function. AMPs viability in practice is demonstrated via prototype implementation in which it operates on real servers in collaboration with a state-of-the-art application manager. The overall system behavior is investigated via simulation.