21,859 Hits in 7.6 sec

Remote I/O

Ian Foster, David Kohr, Rakesh Krishnaiyer, Jace Mogill
1997 Proceedings of the fifth workshop on I/O in parallel and distributed systems - IOPADS '97  
As high-speed networks make it easier to use distributed resources, it becomes increasingly common that applications and their data are not colocated.  ...  In addition to simplifying remote execution, remote I/O can improve performance relative to staging by overlapping computation and data transfer or by reducing communication requirements.  ...  Department of Energy, under Contract W-31-109-Eng-38, in support of the multiagency Scalable I/O project.  ... 
doi:10.1145/266220.266222 dblp:conf/iopads/FosterKKM97 fatcat:4z4kja2yvraqfc6h2wlluv7yby

High-speed I/O

Matthew Burnside, Angelos D. Keromytis
2003 Proceedings of the ACM SIGCOMM workshop on Network-I/O convergence experience, lessons, implications - NICELI '03  
This design parallels the evolution of modern network routers, and has the potential to enable high-performance I/O for end-systems, as well as fully exploit recent trends in programmability of peripheral  ...  (I/O) devices.  ...  [21] describes a mechanism for optimizing high-bandwidth I/O called fbufs.  ... 
doi:10.1145/944747.944756 fatcat:ob4ykky45nd7lc5z3nobszkvk4

I/O Virtualization

Mendel Rosenblum, Carl Waldspurger
2011 Queue  
At the most basic level, decoupling enables time-and space-multiplexing of I/O devices, allowing multiple logical devices to be implemented by a smaller number of physical devices.  ...  The same VM image can be run on computers with different I/O devices and configurations, with the I/O virtualization layer providing the necessary conversion.  ...  In other cases, even making an extra in-memory copy of I/O data may be prohibitively expensive-for example, for fast networks with extremely high packet rates.  ... 
doi:10.1145/2063166.2071256 fatcat:enq3tcmzjfabvokwnjeajcvxhq

I/O virtualization

Carl Waldspurger, Mendel Rosenblum
2012 Communications of the ACM  
In other cases, even making an extra in-memory copy of I/O data may be prohibitively expensive-for example, for fast networks with extremely high packet rates.  ...  It enables an I/O device to read and write host RAM directly without involving the CPU, which is critical for achieving high-performance I/O rates.  ...  The level of indirection between virtual and physical also provides a convenient hook for interposing on I/O operations transparently, enabling new capabilities such as replication, load balancing, encryption  ... 
doi:10.1145/2063176.2063194 fatcat:k2dddjnzxjfsdfuk75sn2hdzo4

Providing QOS guarantees for disk I/O

Ravi Wijayaratne, A.L. Narasimha Reddy
2000 Multimedia Systems  
In this paper, we address the problem of providing di erent levels of performance guarantees or quality-of-service (QOS) for disk I/O.  ...  We propose and evaluate a mechanism for providing deterministic service for Variable Bit Rate (VBR) streams at the disk.  ...  Acknowledgements Reviewers comments have greatly contributed to the improved presentation of the paper.  ... 
doi:10.1007/s005300050005 fatcat:b76ns4zdj5depekzism5rbd3qe

Cluster I/O with River

Remzi H. Arpaci-Dusseau, Eric Anderson, Noah Treuhaft, David E. Culler, Joseph M. Hellerstein, David Patterson, Kathy Yelick
1999 Proceedings of the sixth workshop on I/O in parallel and distributed systems - IOPADS '99  
River is dcsioed to pmvide maximum performance in the c&on case -even in the-race a* nor,uniformities in hardware, software, and workload.  ...  River is based on two simple design features: a high-performanccdirniburedqueue, and a storage redundancy mechanism called gradunfed declusrering.  ...  We'd also like to thank Alan Mainwaring for his work on and support of Active Messages; it is the sine quo non of our work. Many thanks to the anonymous reviewen for all of their helpful commc~ts.  ... 
doi:10.1145/301816.301823 dblp:conf/iopads/Arpaci-DusseauATCHPY99 fatcat:agexl2kxdzgnxjbqchyep4rh6u

Energy-Efficient Mobile Network I/O [article]

Kemal Guner, Tevfik Kosar
2018 arXiv   pre-print
depend on network I/O.  ...  O.  ...  [19] takes this approach to another step, and puts the device into sleep even during data transfer by exploiting the high-bandwidth wireless interface.  ... 
arXiv:1805.03970v1 fatcat:6o5erbszhffnzocfcnbdd7krga

Revisiting network I/O APIs

Luigi Rizzo
2012 Communications of the ACM  
Thanks to a careful design and the engineering of a new packet I/O API, netmap eliminates much unnecessary overhead and moves traffic up to 40 times faster than existing operating systems.  ...  At the time, memory was a scarce resource; links operated at low (by today's standards) speeds; parallel processing was an advanced research topic; and the ability to work at line rate in all possible  ...  These applications spend the majority of their time doing raw packet I/O.  ... 
doi:10.1145/2093548.2093565 fatcat:zqcmfhmxpfagfohrmysvz3lbgu

I/O Deduplication

Ricardo Koller, Raju Rangaswami
2010 ACM Transactions on Storage  
Evaluation of a prototype implementation using these workloads showed an overall improvement in disk I/O performance of 28 to 47% across these workloads.  ...  We introduce I/O Deduplication, a storage optimization that utilizes content similarity for improving I/O performance by eliminating I/O operations and reducing the mechanical delays during I/O operations  ...  content, thus providing alternate locations to retrieve the data from.  ... 
doi:10.1145/1837915.1837921 fatcat:snyfxksbyzdgzorutruv2ox3p4

Learning to Exploit Multiple Vision Modalities by Using Grafted Networks [article]

Yuhuang Hu and Tobi Delbruck and Shih-Chii Liu
2020 arXiv   pre-print
labeled data.  ...  The self-supervised training uses only synchronously-recorded intensity frames and novel sensor data to maximize feature similarity between the pretrained network and the grafted network.  ...  Learning to Exploit Multiple Vision Modalities by Using Grafted Networks  ... 
arXiv:2003.10959v3 fatcat:znkxpkrqwrbrtliifeiwb6ut3e

Application-Tailored I/O with Streamline

Willem de Bruijn, Herbert Bos, Henri Bal
2011 ACM Transactions on Computer Systems  
It improves performance of I/O-bound applications (such as webservers and streaming media applications) by constructing tailor-made I/O paths through the operating system for each application at runtime  ...  Streamline is a stream-based OS communication subsystem that spans from peripheral hardware to userspace processes.  ...  Acknowledgments We are grateful to Kees Verstoep and the anonymous TOCS reviewers for commenting on earlier drafts of the paper, Hamid Bazoobandi for helping with the evaluation of splicing, and Tudor  ... 
doi:10.1145/1963559.1963562 fatcat:z4z5yvufjvfvlcugtkdtmy6fc4

Evolution of the ROOT Tree I/O [article]

Jakob Blomer and Philippe Canal and Axel Naumann and Danilo Piparo
2020 arXiv   pre-print
In this contribution, we present the status and plans of the future ROOT 7 event I/O.  ...  On the performance side, we show first benchmarks using ROOT's new experimental I/O subsystem that combines the best of TTrees with recent advances in columnar data formats.  ...  We would like to thank Dirk Düllmann and Michal Simon from CERN IT for providing us an XRootD test node.  ... 
arXiv:2003.07669v1 fatcat:gvqgylfmvbfkxhguinahcqnlk4

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

Matt Welsh, David Culler
2000 Lecture Notes in Computer Science  
and disk I/O.  ...  We present Tigris, a high-performance computation and I/O substrate for clusters of workstations that is implemented entirely in Java.  ...  However, obtaining high-performance I/O and efficient exploitation of low-level system resources remain as important performance problems.  ... 
doi:10.1007/3-540-40889-4_2 fatcat:k2ewe36mkvc4fk2aawu7en62de

I/O in the ATLAS multithreaded framework

Jack Cranshaw, David Malon, Marcin Nowak, Peter Van Gemmeren, A. Forti, L. Betev, M. Litmaath, O. Smirnova, P. Hristov
2019 EPJ Web of Conferences  
In a multithreaded framework, an I/O infrastructure must efficiently supply event data to and collect it from many threads processing multiple events in flight.  ...  This I/O infrastructure must also address how to read, make available, and propagate in-file metadata and other non-event data needed as context for event processing.  ...  For multithreaded input this required changes to the data provider service to provide a separate cache of event data for all configured slots.  ... 
doi:10.1051/epjconf/201921405017 fatcat:5cmjrysbzjeibkrjqywbhlszrm

Collective prefetching for parallel I/O systems

Yong Chen, Philip C. Roth
2010 2010 5th Petascale Data Storage Workshop (PDSW '10)  
The idea is to exploit the correlation among I/O accesses of multiple processes of a parallel application and carry out prefetches collectively, instead of the traditional strategy of carrying out prefetches  ...  Data prefetching can be beneficial for improving parallel I/O system performance, but the amount of benefit depends on how efficiently and swiftly prefetches can be done.  ...  The authors are grateful to Prof. Xian-He Sun of Illinois Institute of Technology and Dr.  ... 
doi:10.1109/pdsw.2010.5668089 fatcat:gthme2jfcrajlnrgvxpetwi4ve
« Previous Showing results 1 — 15 out of 21,859 results