Grid resource allocation: allocation mechanisms and utilisation patterns (original) (raw)
Related papers
2000
The emergence and widespread adoption of Grid computing has been fueled by continued growth in both our understanding of application requirements and the sophistication of the technologies used to meet these requirements. We provide an introduction to Grid applications and technologies and discuss the important role that resource management will play in future developments.
MUAT: An Environment for Accounting and Characterization of the Use of Computational Grids1
The use of computational grids, which allow sharing distributed resources to achieve high processing power, has been spreading and acquiring importance lately (mostly in business environments). When the grid is employed in an inter-institutional, decentralized manner, it becomes necessary to characterize and account the use of the infrastructure to identify the most used resources, the number of services executed, who contributes with more resources, and so on. The management tools available are limited to monitor the status of environment resources, such as CPU load and memory usage, neglecting statistical and historical data about the execution of applications on the grid. To bridge this gap, the paper presents MUAT (MyGrid/OurGrid Usage Accounting Tool), an environment that aims to evaluate the use of computational grids infrastructures based on MyGrid/OurGrid solutions.
MUAT: An Environment for Accounting and Characterization of the Use of Computational Grids
2005
The use of computational grids, which allow sharing distributed re- sources to achieve high processing power, has been spreading and acquiring importance lately (mostly in business environments). When the grid is em- ployed in an inter-institutional, decentralized manner, it becomes necessary to characterize and account the use of the infrastructure to identify the most used resources, the number of services executed, who contributes with more re- sources, and so on. The management tools available are limited to monitor the status of environment resources, such as CPU load and memory usage, neglect- ing statistical and historical data about the execution of applications on the grid. To bridge this gap, the paper presents MUAT (MyGrid/OurGrid Usage Ac- counting Tool), an environment that aims to evaluate the use of computational grids infrastructures based on MyGrid/OurGrid solutions.
Grids: Harnessing Geographically-Separated Resources in a Multi-Organisational Context
High Performance Computing Systems and Applications, 2003
Grids are becoming ubiquitous platforms for high-performance computing and distributed collaboration. A grid benefits users by permitting them to access heterogeneous resources, such as machines, data, people and devices, that are distributed geographically and organisationally. It benefits organisations by permitting them to offer unused resources on existing hardware and thus reclaim otherwise lost costs. Although worldwide grids can be constructed today, issues regarding heterogeneity, security and failures must be resolved especially if the participating resources are controlled by different organisations. A grid infrastructure that harnesses the power of distributed resources for computing and collaboration must respect the autonomy of organisations to choose policies for using their resources. Legion is a grid infrastructure that presents users a view of a grid as a single virtual machine [GRIM97]. This view reduces the complexities a user encounters before running applications or collaborating on a grid. The functions performed by Legion on a grid are similar to the functions performed by a traditional operating system on underlying hardware. The design principles of object-basedness and integration have enabled Legion to be extended and configured in a number of ways, while ensuring that the cognitive burden on the grid community is small. 1.1 Grid History Grids are the next step in a logical progression beginning with the Internet and the World Wide Web. The internet enabled connecting previously-isolated islands of computing resources to one another. With internet tools, a user could connect to a machine remotely, without being physically present at the machine. After connecting to a remote machine, the user could utilise a small set of services, such as transferring data or issuing limited commands. The World Wide Web improved over the internet in two ways. First, it made the internet more accessible by making the tools more usable. Second, it enabled a richer form of sharing among users. Previous internet tools transferred raw, uninterpreted data. However, a web browser interprets data, thus giving users a better interface and enabling more abstract collaboration, such as sharing a picture rather than transferring kilobytes. The web showed that for computing infrastructure to be considered useful, it must enable collaboration. A grid extends the notions of collaboration while preserving the traditional role of computers as resources used for computing. In essence, computing is collaboration, where a resource provider and a consumer collaborate using a job or task as a unit of collaboration. A large number of applications are starved for computation resources (searches for extra-terrestrial intelligence, studies of protein folding, genomics, stock market models, etc.), whereas an overwhelming majority of computers are often idle. This disconnect can be bridged by permitting computationintensive applications to be run on otherwise idle resources, no matter where the resources are located. Running Java applets on the web is a form of computing-as-collaboration; however, it is still not a grid because the model for running applets merely extends the basic web model. The sophisticated collaboration enabled by a grid is desirable; scientific users expect to share more than images, financial users expect to share more than periodically-updated tables, and all users expect to control who accesses whatever they choose to share. Since a grid is a first-class step in the evolution of computational infrastructures, a design from first principles is indicated strongly to satisfy and anticipate current and future demands. 1.2 Legion History The Legion project evolved from the experience gained from an earlier project, Mentat, and the guidance of multiple professors of Computer Science at the University of Virginia. The domain expertise of each contributordistributed systems, networks, architecture, security, programming languages and information retrieval-led to an integrated infrastructure for managing grids [GRIM94]. This design process is a reflection, on a much smaller scale, of the design process that resulted in the sophisticated operating systems available today. Mentat, the precursor of Legion, was a data-parallel language that added a small number of keywords to the vocabulary of C++. Mentat programs were parsed by a compiler which determined data dependencies, placed data accordingly, and extracted as much parallelism as possible from the program [GRIM96]. Exploiting fine-grained parallelism is expensive; therefore, the designers of Mentat focussed on exploiting coarser-grained parallelism for grids.
Quantification of Grid Resource Heterogeneity Effects on Performance
2006
Abstract: Grid computing enables sharing, selection and aggregation of large collections of geographically and organizationally distributed heterogeneous resources to increase computational, and storage power, resource accessibility and utilization for solving large-scale data intensive problems in science, engineering and commerce. One of the distinct characteristics of grid system is resource heterogeneity. The effective use of a Grid requires the definition of an approach to manage the heterogeneity of the involved resources that ...
Division of Labor: Tools for Growing and Scaling Grids
Lecture Notes in Computer Science, 2006
To enable Grid scalability and growth, a usage model has evolved whereby resource providers make resources available not to individual users directly, but rather to larger units, called virtual organizations. In this paper, we describe abstractions that allow resource providers to delegate the usage of remote resources dynamically to virtual organizations in applicationindependent ways, and present and evaluate an implementation of this abstraction using the Xen virtual machine and Linux networking tools. We also describe how our implementation is being used in a specific context, namely the enforcement of resource allocations in the Edge Services Framework, currently deployed in the Open Science Grid.
Resource Management Services for a Grid Analysis Environment
Computing Research Repository, 2005
Selecting optimal resources for submitting jobs on a computational Grid or accessing data from a data grid is one of the most important tasks of any Grid middleware. Most modern Grid software today satisfies this responsibility and gives a best-effort performance to solve this problem. Almost all decisions regarding scheduling and data access are made by the software automatically, giving users little or no control over the entire process. To solve this problem, a more interactive set of services and middleware is desired that provides users more information about Grid weather, and gives them more control over the decision making process. This paper presents a set of services that have been developed to provide more interactive resource management capabilities within the Grid Analysis Environment (GAE) being developed collaboratively by Caltech, NUST and several other institutes. These include a steering service, a job monitoring service and an estimator service that have been designed and written using a common Grid-enabled Web Services framework named Clarens. The paper also presents a performance analysis of the developed services to show that they have indeed resulted in a more interactive and powerful system for user-centric Gridenabled physics analysis.