Filters








27 Hits in 7.8 sec

IO-Lite: a unified I/O buffering and caching system

Vivek S. Pai, Peter Druschel, Willy Zwaenepoel
2000 ACM Transactions on Computer Systems  
This paper presents the design, implementation, and evaluation of IO-Lite, a uni ed I O bu ering and caching system for general-purpose operating systems.  ...  IO-Lite eliminates all copying and multiple bu ering of I O data, and enables various cross-subsystem optimizations.  ...  Acknowledgments We are grateful to our OSDI shepherd Greg Minshall and the anonymous reviewers, whose comments have helped to improve this paper.  ... 
doi:10.1145/332799.332895 fatcat:mxhmhzjapjh4leyp6y65phrgde

A distributed shared buffer space for data-intensive applications

R. Lachaize, J.S. Hansen
2005 CCGrid 2005. IEEE International Symposium on Cluster Computing and the Grid, 2005.  
A prototype, implemented in the context of a distributed storage system, has been validated with several networking technologies, showing that such a framework can combine performance and flexibility.  ...  We propose separating control and data transfer traffic by accessing data through a DSM-like cluster-wide shared buffer space and only including buffer references in the control messages.  ...  Efficient buffer handling within a single networked system has been attacked by IO-Lite [17] -a unified buffering and caching infrastructure.  ... 
doi:10.1109/ccgrid.2005.1558659 dblp:conf/ccgrid/LachaizeH05 fatcat:6mugvfvmhrh4dnlnghevau6qmq

PipesFS

Willem de Bruijn, Herbert Bos
2008 ACM SIGOPS Operating Systems Review  
PipesFS extends the pipeline model to kernel I/O and communicates with applications through a Linux virtual filesystem (VFS), where directory nodes represent operations and pipe nodes export live kernel  ...  This paper presents PipesFS, an I/O architecture for Linux 2.6 that increases I/O throughput and adds support for heterogeneous parallel processors by (1) collapsing many I/O interfaces onto one: the Unix  ...  Unifies I/O processing by making all kernel IO accessible through a virtual filesystem and all data through Unix pipes.  ... 
doi:10.1145/1400097.1400104 fatcat:h2kag7s5wjdufinu6eyhb4nqtu

CARDS: cluster-aware remote disks

V. Olaru, W.F. Tichy
2003 CCGrid 2003. 3rd IEEE/ACM International Symposium on Cluster Computing and the Grid, 2003. Proceedings.  
This paper presents Cluster-Aware Remote Disks (CARDs), a Single System I/O architecture for cluster computing. CARDs virtualize accesses to remote cluster disks over a System Area Network.  ...  We designed and implemented a decentralized policy called Home-based Serverless Cooperative Caching (HSCC).  ...  We are grateful to Jürgen Reuter for thoroughly proofreading this paper as well as for his invaluable L A T E X assistance.  ... 
doi:10.1109/ccgrid.2003.1199359 dblp:conf/ccgrid/OlaruT03 fatcat:scjkyofq2nh4hkhx73jrbd3p7a

Beltway Buffers: Avoiding the OS Traffic Jam

W. de Bruijn, H. Bos
2008 2008 Proceedings IEEE INFOCOM - The 27th Conference on Computer Communications  
Beltway buffers are operating system I/O paths optimised for high-throughput network applications.  ...  INTRODUCTION We introduce a novel system-wide buffer management system based on extensive use of ring buffers.  ...  ACKNOWLEDGEMENTS We would like to thank Peter Druschel, Philip Homburg, Kees Verstoep and Tomas Hruby for commenting on earlier versions of this paper.  ... 
doi:10.1109/infocom.2007.36 fatcat:fssx3bla5baoplxgafx5suxpxi

Beltway Buffers: Avoiding the OS Traffic Jam

W. de Bruijn, H. Bos
2008 IEEE INFOCOM 2008 - The 27th Conference on Computer Communications  
Beltway buffers are operating system I/O paths optimised for high-throughput network applications.  ...  INTRODUCTION We introduce a novel system-wide buffer management system based on extensive use of ring buffers.  ...  ACKNOWLEDGEMENTS We would like to thank Peter Druschel, Philip Homburg, Kees Verstoep and Tomas Hruby for commenting on earlier versions of this paper.  ... 
doi:10.1109/infocom.2008.36 dblp:conf/infocom/BruijnB08 fatcat:jpaeeqsarvgenfy7wrxm5472ey

Operating system support for multimedia systems

T. Plagemann, V. Goebel, P. Halvorsen, O. Anshus
2000 Computer Communications  
of UFS and CMFS in [Ramakrishnan et al. 93]. © Integrated file systems multiplex all available resources (storage space, disk bandwidth, and buffer cache) among all multimedia data.  ...  of major performance bottlenecks, like copy operations in I/O subsystems; and (7) user-level control of resources including communication.  ...  ACKNOWLEDGEMENTS We would like to thank Frank Eliassen, Liviu Iftode, Martin Karsten, Ketil Lund, Chuanbao Wang, and Lars Wolf for reviewing earlier versions of this paper and their valuable comments and  ... 
doi:10.1016/s0140-3664(99)00180-2 fatcat:wmr4vqndazezjlocfa5kvtggxe

TPF: TCP Plugged File System for Efficient Data Delivery over TCP

Sang Seok Lim, Kyu Ho Park
2007 IEEE transactions on computers  
As a consequence, TPF provides minimal context switching overhead, high buffer utilization, and highly effective disk access utilization. We have implemented and tested the mechanisms in Linux 2.4.  ...  The second is the inefficient uniform data prefetching for TCP connections, irrespective of their characteristics such as bandwidth, latency, and the status of a send buffer.  ...  This work was supported by the NRL (National Research Laboratory) of the Ministry of Science and Technology of Korea.  ... 
doi:10.1109/tc.2007.1009 fatcat:m7fhp6f2crholifizwplavlb2y

Measurement, analysis and performance improvement of the Apache Web server

Yiming Hu, A. Nanda, Qing Yang
1999 1999 IEEE International Performance, Computing and Communications Conference (Cat. No.99CH36305)  
In this paper we measure and analyze the behavior of the popular Apache Web server on a uniprocessor system and a 4-CPU SMP (Symmetric Multi-Processor) system running the IBM AIX operating system.  ...  Finally, our results suggest that operating system support for directly sending data from the file system cache to the TCP/IP network can further improve the Web server performance dramatically. £ A preliminary  ...  Acknowledgments Erich Nahum at IBM Watson read an early draft of this paper and gave many constructive comments and suggestions that greatly improved this paper.  ... 
doi:10.1109/pccc.1999.749447 dblp:conf/ipccc/HuNY99 fatcat:sx3sjf2zqnbxnipilq2cflcqfu

Network stack specialization for performance

Ilias Marinos, Robert N.M. Watson, Mark Handley
2014 Proceedings of the 2014 ACM conference on SIGCOMM - SIGCOMM '14  
Generality has historically been required so that individual systems could perform many functions.  ...  We present Sandstorm and Namestorm, web and DNS servers that utilize a clean-slate userspace network stack that exploits knowledge of application-specific workloads.  ...  IO-Lite [27] unifies the data management between OS subsystems and userspace applications by providing page-based mechanisms to safely and concurrently share data.  ... 
doi:10.1145/2619239.2626311 dblp:conf/sigcomm/MarinosWH14 fatcat:beox2czdqzhezblmyapqurom5e

Network stack specialization for performance

Ilias Marinos, Robert N.M. Watson, Mark Handley
2014 Computer communication review  
Generality has historically been required so that individual systems could perform many functions.  ...  We present Sandstorm and Namestorm, web and DNS servers that utilize a clean-slate userspace network stack that exploits knowledge of application-specific workloads.  ...  IO-Lite [27] unifies the data management between OS subsystems and userspace applications by providing page-based mechanisms to safely and concurrently share data.  ... 
doi:10.1145/2740070.2626311 fatcat:4ov4jz7va5dcblyub6dij5zbu4

Network stack specialization for performance

Ilias Marinos, Robert N. M. Watson, Mark Handley
2013 Proceedings of the Twelfth ACM Workshop on Hot Topics in Networks - HotNets-XII  
Generality has historically been required so that individual systems could perform many functions.  ...  We present Sandstorm and Namestorm, web and DNS servers that utilize a clean-slate userspace network stack that exploits knowledge of application-specific workloads.  ...  IO-Lite [27] unifies the data management between OS subsystems and userspace applications by providing page-based mechanisms to safely and concurrently share data.  ... 
doi:10.1145/2535771.2535779 dblp:conf/hotnets/MarinosWH13 fatcat:hfx7vkmolbfl5fbs25xitxatq4

The Cyclone Server Architecture: streamlining delivery of popular content

Stanislav Rost, John Byers, Azer Bestavros
2002 Computer Communications  
In this paper, we describe the design and implementation of our approach as a Linux kernel subsystem.  ...  We propose a new technique for efficiently delivering popular content from information repositories with bounded file caches.  ...  The IO-Lite system [37] is a good example of a recent effort that aim to achieve the first of the above two goals in the context of Internet servers [36] .  ... 
doi:10.1016/s0140-3664(01)00412-1 fatcat:rex7bc5ldzcvfmww6uehx6lcmm

Techniques for fast and scalable time series traffic generation

Jilong Kuang, Daniel G. Waddington, Changhui Lin
2015 2015 IEEE International Conference on Big Data (Big Data)  
Compared with a stock Linux baseline and a traditional mmap-based file I/O approach, we observe that our traffic generator significantly outperforms other alternatives in terms of throughput (10X), scalability  ...  To design and test such systems, it is desirable to have a high-performance traffic generator specifically designed for time series data, preferably using archived data to create a truly realistic workload  ...  using multi-queue and double-buffer schemes. • We implement our system on two multicore servers and compare our system with a stock Linux baseline and a traditional mmap-based file I/O approach.  ... 
doi:10.1109/bigdata.2015.7363747 dblp:conf/bigdataconf/KuangWL15 fatcat:7mb4uwfsgnh3xdfuleh6iqtxgq

Unifier: Unifying cache management and communication buffer management for PVFS over intiniband

Jiesheng Wu, P. Wyckoff, D. Pandal, R. Ross
IEEE International Symposium on Cluster Computing and the Grid, 2004. CCGrid 2004.  
Unifier has three notable features. (1) Unifier integrates cache management and communication buffer management.  ...  Experimental results show performance improvements between 30% and 70% over other approaches. Better scalability is also achieved by the PVFS I/O servers.  ...  Acknowledgments We are grateful to Phil Carns from Clemson University for providing us many insights into PVFS2 and many helps during our design and implementation.  ... 
doi:10.1109/ccgrid.2004.1336646 dblp:conf/ccgrid/WuWPR04 fatcat:nf3oyqlimjhonclepadrvn25oy
« Previous Showing results 1 — 15 out of 27 results