973 Hits in 7.4 sec

Bulk file I/O extensions to Java

Dan Bonachea
2000 Proceedings of the ACM 2000 conference on Java Grande - JAVA '00  
The file I/O classes present in Java have proven too inefficient to meet the demands of high-performance applications that perform large amounts of I/O.  ...  We present two extensions to the Java I/O libraries which alleviate this problem.  ...  written for the Java I/O stream library.  ... 
doi:10.1145/337449.337459 dblp:conf/java/Bonachea00 fatcat:3a2uqcn7pjgmhpc7belbfabusm

Idempotent I/O for safe time travel [article]

Zoltan Somogyi
2003 arXiv   pre-print
If the part of the computation being jumped back over performs I/O, then the program will perform these I/O operations twice, which will result in unwanted effects ranging from the benign (e.g. output  ...  We present a simple mechanism for ensuring that every I/O action called for by the program is executed at most once, even if the programmer asks the debugger to travel back in time from after the action  ...  The Mercury standard library ultimately implements all I/O operations using the foreign language interface.  ... 
arXiv:cs/0311040v1 fatcat:4sdjdoxbybhi3m5qrqkdibt6pm

HPC the easy way:new technologies for high performance application deployment

M. Danelutto
2003 Eleventh Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2003. Proceedings.  
O>* p1 = new Farm<I,O>(worker,nw); Farm<I,O>* p2 = new Farm<I,O>(*p1,nw); return p2;} int main(int argc, char **argv){ try{ InitSkeletons(argc,argv); Initial<int> p1(init); Atomic<int,int> p2(square  ...  Danelutto Euromicro PDP 2003 36 Skeletons PIPE FARM FORALL skeleton (pattern) library (parametric, performance models) M.  ...  Components : Java beans Introspection, Properties, Customization, Events, Persistence, Methods  ... 
doi:10.1109/empdp.2003.1183585 dblp:conf/pdp/Danelutto03 fatcat:nob375kv5rdk7pfyggfvjlfyz4

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  
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.  ...  Although Hadoop MapReduce provides good programming abstractions and horizontal scalability, it's often blamed for its poor single node performance.  ...  Java and Hadoop I/O streams used the decorator pattern extensively which introduces complex class hierarchy.  ... 
doi:10.1109/bigdata.2013.6691703 dblp:conf/bigdataconf/YangZYDYLZJW13 fatcat:i6oblu6g6ffozaqc6supzl7zri

Hybrid Map Task Scheduling for GPU-Based Heterogeneous Clusters

Koichi Shirahata, Hitoshi Sato, Satoshi Matsuoka
2010 2010 IEEE Second International Conference on Cloud Computing Technology and Science  
The results also indicate that the performance of map tasks, including both CPU and GPU tasks, is significantly affected by the overhead of map task invocation in the Hadoop framework.  ...  Such large-scale computers employ GPUs to enjoy its good peak performance and high memory bandwidth.  ...  ACKNOWLEDGMENT This work is partially supported by the Ministry of Education, Culture, Sports Science, and Technology, the Grants-in-Aid for Scientific Research on Priority Areas (18049028) and the JST-CREST  ... 
doi:10.1109/cloudcom.2010.55 dblp:conf/cloudcom/ShirahataSM10 fatcat:46xq6lkpkbelfmhkggbddja7zu

spChains: A Declarative Framework for Data Stream Processing in Pervasive Applications

Dario Bonino, Fulvio Corno
2012 Procedia Computer Science  
While stream processing blocks are delivered as a standard, yet extensible, library of application-independent processing elements, chains can be defined by the pervasive application engineering team.  ...  Such data is crucial to enable applications to "understand" the current context and to infer the right actions to perform, be they fully automatic or involving some user decisions.  ...  Figure 4 :Figure 5 : 45 <spXML : from b l o c k I d ="Avg1 " p o r t =" o u t " / > <spXML : t o b l o c k I d ="Th1 " p o r t =" i n " / > </spXML : c o n n e c t i o n > </spXML : c o n n e c t i o n  ... 
doi:10.1016/j.procs.2012.06.042 fatcat:6vic24ph5vb5ffhntu3oyy4q64


Sahan Gamage, Cong Xu, Ramana Rao Kompella, Dongyan Xu
2014 Proceedings of the ACM Symposium on Cloud Computing - SOCC '14  
We observe that such overhead particularly affects many applications involving piped I/O data movements, such as web servers, streaming servers, big data analytics, and storage, because the data has to  ...  be transferred first into the application from the source I/O device and then back to the sink I/O device, incurring the virtualization overhead twice.  ...  We thank our shepherd, Ashvin Goel and the anonymous reviewers for their valuable comments and suggestions. This work was supported in part by NSF grants 0855141, 1054788, and 1219004.  ... 
doi:10.1145/2670979.2671006 dblp:conf/cloud/GamageXKX14 fatcat:kru7dnswf5hgpkrtnn4g7ek5mu

PipeGen: Data Pipe Generator for Hybrid Analytics [article]

Brandon Haynes, Alvin Cheung, Magdalena Balazinska
2016 arXiv   pre-print
To achieve high performance, PipeGen leverages the ability of all DBMSs to export, possibly in parallel, data into a common data format, such as CSV or JSON.  ...  We implement a prototype of PipeGen and evaluate it by automatically generating data pipes between five different DBMSs.  ...  Acknowledgements This work is supported in part by the National Science Foundation through NSF grants IIS-1247469 and IIS-1110370, and gifts from the Intel Science and Technology Center for Big Data, Amazon  ... 
arXiv:1605.01664v2 fatcat:3wqmro7rxzdedav66ra7oeegiq

iFlow (poster session)

Andrew Kwok-Fai Lui, Mark W. Grigg, Michael J. Owen, T. Andrew Au
2000 Addendum to the 2000 proceedings of the conference on Object-oriented programming, systems, languages, and applications (Addendum) - OOPSLA '00  
The creation of an application framework is highly desirable for the development of stream based applications.  ...  The stream based programming model constrains applications to operate with smaller individual data blocks in a continuous manner.  ...  The pipes can therefore abstract away the complexities of data streaming such as machine boundaries, data processing methods, and stream I/O controls.  ... 
doi:10.1145/367845.367940 dblp:conf/oopsla/LuiGOA00 fatcat:uuc5xld44fhrrbzcmpgqvx7q2a

GPGPU Composition with OCaml

Mathias Bourgoin, Emmanuel Chailloux
2014 Proceedings of ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming - ARRAY'14  
The main ones are : write and use hyper-parallel kernels on GPU, manage memory transfers between CPU and GPU, and compose kernels, keeping individual performance of components while optimizing the global  ...  To achieve it, we use the SPOC library, developed in OCaml. SPOC offers abstractions over the Cuda and OpenCL frameworks.  ...  PROP being a part of We only translated the computation part to OCaml, keeping I/O in Fortran [11] .  ... 
doi:10.1145/2627373.2627379 dblp:conf/pldi/BourgoinC14 fatcat:ydk2wy4ew5er3ks3a52ollctyy

Slipstream: Automatic Interprocess Communication Optimization

Will Dietz, Joshua Cranmer, Nathan Dautenhahn, Vikram S. Adve
2015 USENIX Annual Technical Conference  
Our design in particular (1) requires no changes to the kernel or applications, (2) correctly identifies (and optimizes) pairs of communicating local endpoints, without knowledge of routing performed locally  ...  or by the network, and (3) imposes little or no overhead when optimization is not possible, including communication with parties not using our technology.  ...  Transparent Userspace Libraries: The FABLE library, which is only described in a position paper [27] , provides automatic transport selection and dynamic reconfiguration of the I/O channel.  ... 
dblp:conf/usenix/DietzCDA15 fatcat:3jbhsprlejbw5imuysicfisdbu

Composable proxy services to support collaboration on the mobile internet

P.K. McKinley, U.I. Padmanabhan, N. Ancha, S.M. Sadjadi
2003 IEEE transactions on computers  
Unlike conventional Java I/O streams, detachable streams can be stopped, disconnected, reconnected, and restarted.  ...  The approach is based on detachable Java I/O streams, which enable proxy filters and transcoders to be dynamically inserted, removed, and reordered on a given data stream.  ...  ACKNOWLEDGMENTS The authors would like to thank Robel Barrios, Aaron Malenfant, Jesus Arango, Ji Li, Arun Mani, Suraj Gaurav, Peng Ge, and Chiping Tang for their contributions to this work.  ... 
doi:10.1109/tc.2003.1204828 fatcat:5f6e5z5aifcgbdy2xexeii5bba

Accessible near-storage computing with FPGAs

Robert Schmid, Max Plauth, Lukas Wenzel, Felix Eberhardt, Andreas Polze
2020 Proceedings of the Fifteenth European Conference on Computer Systems  
compute kernels to operate on the granularity of file data streams.  ...  The increased diversity of compute resources in turn affects programming models and practicalities of software development for near-data compute kernels and raises the question of how those resources can  ...  Acknowledgements We thank the anonymous reviewers and our shepherd, Tim Harris, for their valuable feedback.  ... 
doi:10.1145/3342195.3387557 dblp:conf/eurosys/SchmidPWEP20 fatcat:ykijhsup6jdpnkpcer2picfbfu

Symbolic computation in Java

Laurent Bernardin, Bruce Char, Erich Kaltofen
1999 Proceedings of the 1999 international symposium on Symbolic and algebraic computation - ISSAC '99  
Acknowledgement The authors thank the referees for their helpful comments.  ...  We intend to alter our design for MathServer-Bean so that it depends on a separate engine interaction component to specify the engine I/O protocol.  ...  Engine configuration and invocation for either the "pipe" or "native method" modes of engine can be performed at run time, allowing an application to adapt efficiently to new needs as users develop them  ... 
doi:10.1145/309831.309946 dblp:conf/issac/BernardinCK99 fatcat:wd4kvix74rdftgiiq2pp7srvsa

Analysis of content copyright infringement in mobile application markets

Ryan Johnson, Nikolaos Kiourtis, Angelos Stavrou, Vincent Sritapan
2015 2015 APWG Symposium on Electronic Crime Research (eCrime)  
These applications are both paid and free and in the case of free applications, the source of funding appears to be advertisements that are displayed while the content is streamed to the device.  ...  In this paper, we assess the extent of content copyright infringement for mobile markets that span multiple platforms (iOS, Android, and Windows Mobile) and cover both official and unofficial mobile markets  ...  When a log entry is created for reading or writing a single byte from an I/O stream, the performance starts to degrade.  ... 
doi:10.1109/ecrime.2015.7120798 dblp:conf/ecrime/0002KSS15 fatcat:wqgnohivwngztkphyrljb2eph4
« Previous Showing results 1 — 15 out of 973 results