Interact hosting centers serve multiple service sites from a common hardware base. This paper presents the design and implementation of an architecture for resource management in a hosting center op-erating system, with an emphasis on energy as a driving resource management issue for large server clusters. The goals are to provi-sion server resources for co-hosted services in a way that automati-cally adapts to offered load, improve the energy efficiency of server dusters by dynamically resizing the active server set, and respond to power supply disruptions or thermal events by degrading service in accordance with negotiated Service Level Agreements (SLAs). Our system is based on an economic approach to managing shared server resources, in which services "bid " for resources as a func-tion of delivered performance. The system continuously moni-tors load and plans resource allotments by estimating the value of their effects on service performance. A greedy resource allocation algorithm adjusts resource prices to balance supply and demand, allocating resources to their most efficient use. A reconfigurable server switching infrastructure directs request traffic to the servers assigned to each service. Experimental results from a prototype confirm that the system adapts to offered load and resource avail-ability, and can reduce server energy usage by 29 % or more for a typical Web workload. 1.

. In this paper, we propose ICrafter, a framework for services

Modern networks are extremely complex, varying both statically and dynamically. This complexity and dynamism are greatly increased when the network contains mobile elements. A number of researchers have proposed solutions to these problems based on dynamic adaptation to changing network conditions and application requirements. This paper summarizes the results of several such projects and extracts several important general lessons learned about adapting data flows over difficult network conditions. These lessons are then formulated into a conceptual framework that demonstrates how a few simple and powerful ideas can describe a wide variety of different software adaptation systems. This paper describes an Adaptation Framework in the context of the several successful adaptation systems and suggests how the framework can help researchers think about the problems of adaptivity in networks.

This paper presents a distributed persistent object store designed to simplify scalable service in cluster environment. This distributed object store, called TODS (Tsinghua Object Data Store), presents a single-imaged, transparent persistent and object-oriented view of the storage devices of the whole cluster. TODS is designed to be incremental scalable and efficient, and also has the properties of the high concurrency, high throughput and availability which are necessary for scalable service. TODS supports distributed ACID transactions within the cluster, which qualifies its use in the building of complex transactional services. And the user interface of TODS is fitter for building service than that of file system, and significantly easier to use than that of RDBMS. TODS is a reusable platform for scalable service in cluster by forming many general data management functions into one independent layer. This paper gives the motivation, principle and architecture of TODS. Some technique details are also discussed. In our performance experiments, the system scales smoothly to a 36-node server cluster and achieves 11,160 In-memory reads/sec and 396 transactions/sec. 1.

Today’s information systems are based on designs that are a quarter-century old: centralized servers with local metadata, black-box user interactions, complex monolithic software, and underlying hardware assumptions frozen at the date of initial design. Industrial system developers are firmly mired (by necessity) in this legacy of assumptions and historical arcana. Database researchers need not be. We are beginning anew, designing the B-1 (pronounced "be one") universal system for information. In building the B-1, we intend to explore a number of new ideas, combining them with suggestive features of recent research, as well as including a considerate exploration of roads not taken years ago. Our design goal is nothing less than a truly universal system for information: we envision a global federation of information stores inter-operating seamlessly and autonomously, removing today’s artificial distinctions between databases, file systems, and the world-wide web, while simultaneously acknowledging and addressing the complexity of searching, browsing and analyzing the world’s information. This paper presents our vision for the B-1 system. As motivation, we point out scenarios where current information systems provide insufficiently powerful interaction, and we highlight mechanisms both to capture the attention of a broad community of users, and enhance their ability to make use of information. We also observe trends in hardware that suggest significantly different software solutions. The B-1 addresses both of these themes with an architecture of federated components. We