Filters








2,761 Hits in 6.2 sec

Achieving Robust, Scalable Cluster I/O in Java [chapter]

Matt Welsh, David Culler
2000 Lecture Notes in Computer Science  
We present Tigris, a high-performance computation and I/O substrate for clusters of workstations that is implemented entirely in Java.  ...  Tigris automatically balances resource load across the cluster as a whole, shielding applications from asymmetries in CPU, I/O, and network performance.  ...  Tigris' design is based on River [5] , a robust cluster-based I/O system implemented in C++ on the Berkeley Network of Workstations [25] .  ... 
doi:10.1007/3-540-40889-4_2 fatcat:k2ewe36mkvc4fk2aawu7en62de

Designing SSI clusters with hierarchical checkpointing and single I/O space

Kai Hwang, Hai Jin, E. Chow, Cho-Li Wang, Zhiwei Xu
1999 IEEE Concurrency  
MPI-based I/O currently cannot achieve this transparency.  ...  SSI clusters or robust clusters will efficiently meet these requirements. Java-based intelligent agents are also suitable for distributed cluster computing.  ...  He leads a design group at NCIC building a series of cluster-based superservers and scalable multicomputers.  ... 
doi:10.1109/4434.749136 fatcat:tk7exsomkjfevjuypo7woihsiq

Device level communication libraries for high-performance computing in Java

Guillermo L. Taboada, Juan Touriño, Ramón Doallo, Aamir Shafi, Mark Baker, Bryan Carpenter
2011 Concurrency and Computation  
essential to obtain scalability in Java for HPC.  ...  o u t p u t S t r e a m . f l u s h ( ) r e l e a s e S t r e a m f o r D e s t i n a t i o n ( dstID , o u t p u t S t r e a m ) ; Listing 4.  ...  point-to-point performance on Scalable Coherent Interface (SCI) (x86_64 cluster).  ... 
doi:10.1002/cpe.1777 fatcat:hf77nzvkzzdftja475uaxrxmmu

Control Systems application in Java based Enterprise and Cloud Environments – A Survey

Ravi Kumar, Dr. Chelliah, Dr.A.Vinaya Babu
2011 International Journal of Advanced Computer Science and Applications  
Our paper presents this review with a specific emphasis on Java based Web, Application and Enterprise Server Bus environments.  ...  For more than a decade there is active research in exploring feedback control systems applications in computing and some of the results are applied to the commercial software products.  ...  CONCLUSIONS AND FUTURE RESEARCH The feedback control theory has inherent advantages as discussed in [2] such as Quantitative I/O models, Dynamics and Transients, Correlation between multiple metrics,  ... 
doi:10.14569/ijacsa.2011.020820 fatcat:udkt4x3rprgtllms5xmpd6b75i

Techniques for obtaining high performance in Java programs

Iffat H. Kazi, Howard H. Chen, Berdenia Stanley, David J. Lilja
2000 ACM Computing Surveys  
This survey describes research directions in techniques to improve the performance of programs written in the Java programming language.  ...  Another alternative for executing Java programs is a Java processor that implements the JVM directly in hardware.  ...  ACKNOWLEDGMENTS We thank Amit Verma and Shakti Davis for their help in gathering some of the information used in this paper.  ... 
doi:10.1145/367701.367714 fatcat:76fvbgkvnffodgpip2lhajvz4m

NativeTask: A Hadoop compatible framework for high performance

Dong Yang, Xiang Zhong, Dong Yan, Fangqin Dai, Xusen Yin, Cheng Lian, Zhongliang Zhu, Weihua Jiang, Gansha Wu
2013 2013 IEEE International Conference on Big Data  
In the meantime, MapReduce has already achieved a large install base, thus any performance improvement should keep the compatibility.  ...  The results show that the speedup NativeTask achieves ranges from 10% to 160%, and it paves the way for a better MapReduce that excels on both single node performance and scalability.  ...  First, it suggests that direct I/O and streaming I/O are helpful. However, due to the lightweight compression and decompression, CPU is the bottleneck in real world clusters.  ... 
doi:10.1109/bigdata.2013.6691703 dblp:conf/bigdataconf/YangZYDYLZJW13 fatcat:i6oblu6g6ffozaqc6supzl7zri

The Ninja architecture for robust Internet-scale systems and services

Steven D Gribble, Matt Welsh, Rob von Behren, Eric A Brewer, David Culler, N Borisov, S Czerwinski, R Gummadi, J Hill, A Joseph, R.H Katz, Z.M Mao (+2 others)
2001 Computer Networks  
The Ninja project seeks to enable the broad innovation of robust, scalable, distributed Internet services, and to permit the emerging class of extremely heterogeneous devices to seamlessly access these  ...  Our architecture consists of four basic elements: bases, which are powerful workstation cluster environments with a software platform that simpli®es scalable service construction; units, which are the  ...  Acknowledgements This work is supported, in part, by the Defense Advanced Research Project Agency (grant DABT 63-98-C-0038) and the National Science Foundation (grant RI EIA-9802069).  ... 
doi:10.1016/s1389-1286(00)00179-1 fatcat:vaghnmok75a5bfqpiapmnz6g6e

Java and asynchronous iterative applications: large scale experiments

Jacques M. Bahi, Raphael Couturier, David Laiymani, Kamel Mazouzi
2007 2007 IEEE International Parallel and Distributed Processing Symposium  
The results we obtained, demonstrate both the scalability of the Jace environment and its ability to support wide-area deployments and the robustness of asynchronous iterative algorithms in a large scale  ...  In those applications, tasks are dependent and the use of distant clusters may be difficult, for example, because of latencies, heterogeneity, and synchronizations.  ...  NIO is a Java API (introduced in Java 1.4). It provides new features and improved performances in the areas of buffer management, scalable network and file I/O.  ... 
doi:10.1109/ipdps.2007.370420 dblp:conf/ipps/BahiCLM07 fatcat:scsvrjwbcnh23jxd7x6q3yxjm4

Towards benchmarking of AMQP

Stefan Appel, Kai Sachs, Alejandro Buchmann
2010 Proceedings of the Fourth ACM International Conference on Distributed Event-Based Systems - DEBS '10  
The researchers chose Qpid as AMQP middleware and did not enable persistence; by this network I/O becomes the bottleneck rather than disk I/O.  ...  Finally, scalability can be evaluated as well together with fail over capabilities in complex cluster setups.  ... 
doi:10.1145/1827418.1827438 dblp:conf/debs/AppelSB10 fatcat:f3rzirft6zaevkoapyk4tfljku

Review of Performance Analysis Tools for MPI Parallel Programs [chapter]

Shirley Moore, David Cronk, Kevin London, Jack Dongarra
2001 Lecture Notes in Computer Science  
Issues pertaining to hybrid and heterogeneous, distributed programming environments and to MPI I/O performance analysis are also discussed.  ...  This paper reviews the available cross-platform MPI performance analysis tools and evaluates them according to a set of criteria that includes robustness, usability, scalability, portability, and versatility  ...  The new hierarchical Vampir display is a promising approach for visualizing performance on large SMP clusters in a scalable manner.  ... 
doi:10.1007/3-540-45417-9_34 fatcat:j25mwtgwlfasdjkt4azqm4m4cq

Comparative Study of Big Data Computing and Storage Tools: A Review

Bakshi Rohit Prasad, Sonali Agarwal
2016 International Journal of Database Theory and Application  
As a result of tremendous rise in internet usage like social media and forums, mail systems, scholarly and research articles, daily online transactions from multiple sources like health care systems, meteorological  ...  The available technologies deal with two broad aspects related to Big Data that are Big Data Storage Management and Big Data Computing, focused to overcome various challenges such as scalability, faster  ...  In Cloudera, purely I/O bound queries achieve approximately 3-4 times, queries of join or multiple MapReduces achieves approximately 7-45 and simple aggregation queries achieve 20-90 times performance  ... 
doi:10.14257/ijdta.2016.9.1.05 fatcat:xrhakmtvefcdzpqvakjhiro7em

Enabling Loosely-Coupled Serial Job Execution on the IBM BlueGene/P Supercomputer and the SiCortex SC5832 [article]

Ioan Raicu, Zhao Zhang, Mike Wilde, Ian Foster
2008 arXiv   pre-print
We present here the challenges of I/O performance encountered in making this model practical, and show results using both micro-benchmarks and real applications on two large-scale systems, the BG/P and  ...  Our preliminary benchmarks show that we can scale to 4096 processors on the Blue Gene/P and 5832 processors on the SiCortex with high efficiency, and can achieve thousands of tasks/sec sustained execution  ...  In previous work, we have shown that in both micro-benchmarks and a large-scale astronomy application, that a modest small Linux cluster (128 CPUs) can achieve aggregate I/O data rates of tens of Gb/s  ... 
arXiv:0808.3536v1 fatcat:qkppdveoj5gkpk2yecuh22q3zi

Operating Systems

Steve Chapin, Joachim Worringen
2001 The international journal of high performance computing applications  
The usual way is to employ a number (one or more) of dedicated I/O nodes in the cluster. However, every shared resource represents a potential bottleneck in a system that has to be scalable.  ...  provide general I/O services in a distributed manner.  ... 
doi:10.1177/109434200101500204 fatcat:troeuk7dbfhytjc4pyi7pjh7za

Scalable Approaches for Supporting MPI-IO Atomicity

P.M. Aarestad, A. Ching, G.K. Thiruvathukal, A.N. Choudhary
2006 Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID'06)  
Scalable atomic and parallel access to noncontiguous regions of a file is essential to exploit high performance I/O as required by large-scale applications.  ...  Parallel I/O frameworks such as MPI I/O conceptually allow I/O to be defined on regions of a file using derived datatypes.  ...  Acknowledgments This work was funded in part by a grant from the National Science Foundation to Loyola University Chicago (CCF-0444197) and a Cabell Fellowship provided by the Northwestern University McCormick  ... 
doi:10.1109/ccgrid.2006.88 dblp:conf/ccgrid/AarestadCTC06 fatcat:rz6blcrnibh43jav52wsonjm4q

Strengthening Consistency in the Cassandra Distributed Key-Value Store [chapter]

Panagiotis Garefalakis, Panagiotis Papadopoulos, Ioannis Manousakis, Kostas Magoutis
2013 Lecture Notes in Computer Science  
In this short note we argue that the resulting system fits a wider range of applications, and is more robust and easier to reason about.  ...  In this work we strengthen the consistency of Cassandra through the use of appropriate components: the Oracle Berkeley DB Java Edition High Availability storage engine for data replication and a replicated  ...  Furthermore we improve performance by using client-coordinated I/O, avoiding a forwarding step in Cassandra's original I/O path.  ... 
doi:10.1007/978-3-642-38541-4_17 fatcat:pn4svinvubaaxn7fgujksbptvq
« Previous Showing results 1 — 15 out of 2,761 results