File Placement on Distributed Computer Systems

Wah,

doi:10.1109/mc.1984.1658928

Cited by 75 publications

(16 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Papers [10,11] discuss various optimization models for the distribution of files in a computer network. The costs considered include communication and delay.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Traffic Conscious Distribution of Service Components

Anagnostou

Lambrou

2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract.A service is commonly realized by a set of components distributed over different nodes. For example, Internet based applications are orchestrated across a large scale distributed computing infrastructures and underlying resource elements; Similarly, in TINAlike approaches intelligence for control and management of services and resource, in particular, is distributed among network nodes and user/terminal nodes. In fact a service can be seen as a set of interacting components with a common purpose, being it application-oriented or of a support nature . The placement of components in different machines is more or less empirically determined at the design phase, by loosely (and occasionally subconsciously) taking into account "reasonable" predictions of the component usage. Code mobility has added complexity to the distribution problem: The aim of this paper is to present a methodology of dealing with component distribution, to explore its limitations, and to present its effect on service and network design. Introd uctionThe service-oriented computing paradigm considers services as the fundamental elements for constructing applications. Composite services as resulting by basic service components aggregation are utilized by service providers as commercial solutions to be affered to a diverse customer base. Thus, services may be viewed as open, self-contained software components that support effi.cient configuration as well quality of service composition of distributed applications. Services may be supplied by different business stakeholders and comprise a distributed computing infrastructure in support of intra-and cross-enterprise application integration and collaboration [1].Quality of service (QoS) considerations, are seen as an integral part of the service design lifecycle, taking into account important functional and nonfunctional service properties, such as performance, security, reliability, transactional integrity and services overall cost. Today's dominant manifestat ion of service-oriented computing implementation is realized in terms of ernerging web technologies [2] Current service design frameworks merely focus on service capabilities, interface and behavior models and notations, which are expressed in a N.

show abstract

“…Papers [10,11] discuss various optimization models for the distribution of files in a computer network. The costs considered include communication and delay.…”

Section: Related Workmentioning

confidence: 99%

“…Additional requirements refer to parallelism, availability and security. The problern of file migration, that is the reorganization of the file allocation scheme, is also discussed in [10). Reference [12) discusses and compares various file migration and dynamic file allocation problems.…”

Section: Related Workmentioning

confidence: 99%

Traffic Conscious Distribution of Service Components

Anagnostou

Lambrou

2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…There is a large body of literature on load balancing techniques for parallel and distributed systems (e.g., [17,8,23,11,34,10]). Our current system provides a natural mechanism for performing load balancing, but we have not yet determined effective load balancing policies for our setting.…”

Section: Related Workmentioning

confidence: 99%

Cache-and-query for wide area sensor databases

Deshpande

Nath

Gibbons

et al. 2003

Proceedings of the 2003 ACM SIGMOD International Conference on Management of Data

View full text Add to dashboard Cite

Webcams, microphones, pressure gauges and other sensors provide exciting new opportunities for querying and monitoring the physical world. In this paper we focus on querying wide area sensor databases, containing (XML) data derived from sensors spread over tens to thousands of miles. We present the first scalable system for executing XPATH queries on such databases. The system maintains the logical view of the data as a single XML document, while physically the data is fragmented across any number of host nodes. For scalability, sensor data is stored close to the sensors, but can be cached elsewhere as dictated by the queries (auto-tuning). Our design enables self-starting distributed queries that jump directly to the lowest common ancestor of the query result, dramatically reducing query response times. We present a novel query-evaluategather technique (using XSLT) for detecting (1) which data in a local database fragment is part of the query result, and (2) how to gather the missing parts. We define partitioning and cache invariants that ensure that even partial matches on cached data are exploited and that correct answers are returned, despite our dynamic query-driven caching. Experimental results demonstrate that our techniques dramatically increase query throughputs and decrease query response times in wide area sensor databases.

show abstract

“…In this paper we look at the problem of partitioning the distributed storage systems into fixed size arrays in such a way as to minimize the cost of remote accesses that have to be performed to update the parity information. This problem is somewhat related to the problem of file allocation and replica placement in a distributed system which has been studied extensively in the literature [7,8]. However the two problems are different in nature [] because, in the RADD case, there is one redundant item for N data items while in the file allocation problem each file is replicated several times.…”

Section: Stonebraker and Schloss Have Proposed The Redundant Arrays Omentioning

confidence: 99%

Assigning Sites to Redundant Clusters in a Distributed Storage System

Mourad

Fuchs

Saab

1993

1993 International Conference on Parallel Processing - ICPP'93 Vol1

View full text Add to dashboard Cite

Fax (217) Redundant Arrays of Distributed Disks (RADD) can be used in a distributed computing system or database system to provide recovery in the presence of disk crashes and temporary and permanent failures of single sites. In this paper, we look at the problem of partitioning the sites of a distributed storage system into redundant arrays in such a way that the communication costs for maintaining the parity information are minimized. We show that the partitioning problem is NP-hard. We then propose and evaluate several heuristic algorithms for finding approximate solutions. Simulation results show that significant reduction in remote parity update costs can be achieved by optimizing the site partitioning scheme.

show abstract

File Placement on Distributed Computer Systems

Cited by 75 publications

References 13 publications

Traffic Conscious Distribution of Service Components

Traffic Conscious Distribution of Service Components

Cache-and-query for wide area sensor databases

Assigning Sites to Redundant Clusters in a Distributed Storage System

Contact Info

Product

Resources

About