Filters








5,178 Hits in 5.7 sec

Minimizing stall time in single and parallel disk systems

Susanne Albers, Naveen Garg, Stefano Leonardi
2000 Journal of the ACM  
Cao et. al. and Kimbrel and Karlin gave approximation algorithms for minimizing the total elapsed time in single and parallel disk settings.  ...  For the parallel disk problem we give an approximation algorithm for minimizing stall time. Stall time is an important and harder to approximate measure for this problem.  ...  Our contribution: In this paper we present a new approach to the problem of minimizing stall time in single and parallel disk systems.  ... 
doi:10.1145/355541.355542 fatcat:a66mcjjxjfcq7ap5yer7eg7kp4

Minimizing stall time in single and parallel disk systems

Susanne Albers, Naveen Garg, Stefano Leonardi
1998 Proceedings of the thirtieth annual ACM symposium on Theory of computing - STOC '98  
Cao et. al. and Kimbrel and Karlin gave approximation algorithms for minimizing the total elapsed time in single and parallel disk settings.  ...  For the parallel disk problem we give an approximation algorithm for minimizing stall time. Stall time is an important and harder to approximate measure for this problem.  ...  Our contribution: In this paper we present a new approach to the problem of minimizing stall time in single and parallel disk systems.  ... 
doi:10.1145/276698.276858 dblp:conf/stoc/AlbersGL98 fatcat:u2gfn7p72rcgvanujx77rzoydu

Minimizing Stall Time in Single and Parallel Disk Systems Using Multicommodity Network Flows [chapter]

Susanne Albers, Carsten Witt
2001 Lecture Notes in Computer Science  
Cao et. al. and Kimbrel and Karlin gave approximation algorithms for minimizing the total elapsed time in single and parallel disk settings.  ...  For the parallel disk problem we give an approximation algorithm for minimizing stall time. Stall time is an important and harder to approximate measure for this problem.  ...  Our contribution: In this paper we present a new approach to the problem of minimizing stall time in single and parallel disk systems.  ... 
doi:10.1007/3-540-44666-4_6 fatcat:sb2g6roehbedfbntmwbbmr6nnu

Implementing cooperative prefetching and caching in a globally-managed memory system

Geoffrey M. Voelker, Eric J. Anderson, Tracy Kimbrel, Michael J. Feeley, Jeffrey S. Chase, Anna R. Karlin, Henry M. Levy
1998 Performance Evaluation Review  
As a result, recent research has focused on reducing disk stall time through several approaches.  ...  prefetching alone. introduction The past decade has seen a two-order-of-magnitude increase in processor speed, yet only a two-fold improvement in disk access time.  ...  Darrell Anderson, Andrew Gallatin and Ken Yocum contributed Trapeze-related code and patiently answered many questions about Digital Unix internals and the workings of Trapeze and Myrinet.  ... 
doi:10.1145/277858.277869 fatcat:hifezjbowjgvdj67nxgbfs34ta

Implementing cooperative prefetching and caching in a globally-managed memory system

Geoffrey M. Voelker, Eric J. Anderson, Tracy Kimbrel, Michael J. Feeley, Jeffrey S. Chase, Anna R. Karlin, Henry M. Levy
1998 Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems - SIGMETRICS '98/PERFORMANCE '98  
As a result, recent research has focused on reducing disk stall time through several approaches.  ...  prefetching alone. introduction The past decade has seen a two-order-of-magnitude increase in processor speed, yet only a two-fold improvement in disk access time.  ...  Darrell Anderson, Andrew Gallatin and Ken Yocum contributed Trapeze-related code and patiently answered many questions about Digital Unix internals and the workings of Trapeze and Myrinet.  ... 
doi:10.1145/277851.277869 dblp:conf/sigmetrics/VoelkerAKFCKL98 fatcat:h7zslhm6fjfr3apf3hld6jkkum

Parallel Prefetching and Caching Is Hard [chapter]

Christoph Ambühl, Birgitta Weber
2004 Lecture Notes in Computer Science  
In this paper we study integrated prefetching and caching in parallel disk systems.  ...  This paper provides the first negative result in this area by showing that optimizing the stall time is APXhard.  ...  Kimbrel and Karlin [10] extended the single disk model to multiple disks and analyzed Conservative and Aggressive in parallel disks systems.  ... 
doi:10.1007/978-3-540-24749-4_19 fatcat:zcxtirngsfet7ihyuzgwr2fghu

Page 8502 of Mathematical Reviews Vol. , Issue 2002K [page]

2002 Mathematical Reviews  
2002k:68068 1996; see MR 97k:68012], who gave approximation algorithms for minimizing the total elapsed time in single and in parallel disk set- tings.  ...  For the parallel disk problem, we give an approximation algorithm for minimizing stall time. The solution uses a few extra memory blocks in cache.  ... 

Prefetching over a network

David Rochberg, Garth Gibson
1997 Performance Evaluation Review  
Consider a single-threaded application that makes a series of nonsequential reads one at a time in a large file striped over several disks: without knowledge of the upcoming reads, the storage system cannot  ...  This prefetching hides disk latency and exposes storage parallelism.  ...  Given information about the state of the system and good hints describing the future I/O plans of running applications, TIP attempts to allocate buffers in a way that minimizes the I/O stall time, or the  ... 
doi:10.1145/270900.270906 fatcat:yspeeei22bgq3a6ww63egpox3y

A trace-driven comparison of algorithms for parallel prefetching and caching

Tracy Kimbrel, Andrew Tomkins, R. Hugo Patterson, Brian Bershad, Pei Cao, Edward W. Felten, Garth A. Gibson, Anna R. Karlin, Kai Li
1996 ACM SIGOPS Operating Systems Review  
In this paper we study the effects of several combined prefetching and caching strategies for systems with multiple disks.  ...  High-performance I O systems depend on prefetching and caching in order to deliver good performance to applications.  ...  This research is supported in part by NSF grant n umbers ECD-8907068, CCR-9301186, GER-9450075, CCR-9632769, in part by DARPA Contract numbers DABT63-94-C-0049, DABT63-93-C-0054 and in part by generous  ... 
doi:10.1145/248155.238737 fatcat:lwvxppozg5cwndy54sbbedg4em

A trace-driven comparison of algorithms for parallel prefetching and caching

Tracy Kimbrel, Andrew Tomkins, R. Hugo Patterson, Brian Bershad, Pei Cao, Edward W. Felten, Garth A. Gibson, Anna R. Karlin, Kai Li
1996 Proceedings of the second USENIX symposium on Operating systems design and implementation - OSDI '96  
In this paper we study the effects of several combined prefetching and caching strategies for systems with multiple disks.  ...  High-performance I O systems depend on prefetching and caching in order to deliver good performance to applications.  ...  This research is supported in part by NSF grant n umbers ECD-8907068, CCR-9301186, GER-9450075, CCR-9632769, in part by DARPA Contract numbers DABT63-94-C-0049, DABT63-93-C-0054 and in part by generous  ... 
doi:10.1145/238721.238737 dblp:conf/osdi/KimbrelTPBCFGKL96 fatcat:sd3ab7gyhfauthuejbwlwkq4dm

On Performance Stability in LSM-based Storage Systems (Extended Version) [article]

Chen Luo, Michael J. Carey
2019 arXiv   pre-print
We have conducted extensive experiments in the context of the Apache AsterixDB system and we present the results here.  ...  Despite the popularity of LSM-trees, they have been criticized for suffering from write stalls and large performance variances due to the inherent mismatch between their fast in-memory writes and slow  ...  We thank Neal Young, Dongxu Zhao, and anonymous reviewers for their helpful feedback on the theorems in this paper.  ... 
arXiv:1906.09667v4 fatcat:zvbosnuh5nd3rdpel3ppdivk64

Multi-threaded Output in CMS using ROOT [article]

Daniel Riley, Christopher Jones
2019 arXiv   pre-print
We examine the architecture of the new parallel output module, the specific accommodations and modifications that were made to ensure compatibility with the CMS framework scheduler, and the performance  ...  In this paper we explore the changes made to the CMS framework and our ROOT output module to overcome the previous scaling limits, using two new ROOT features: the TBufferMerger asynchronous file merger  ...  Stall graphs comparing the standard output module with IMT disabled and the parallel output module with IMT enabled showing a substantial reduction in stalls and total processing time (note the change  ... 
arXiv:1905.02113v1 fatcat:j3zyi3d2u5eg5ibnkzvumnijja

Integrated prefetching and caching in single and parallel disk systems

Susanne Albers, Markus Büttner
2003 Proceedings of the fifteenth annual ACM symposium on Parallel algorithms and architectures - SPAA '03  
In the second part of the paper, we investigate the problem of minimizing stall time in parallel disk systems.  ...  We study integrated prefetching and caching in single and parallel disk systems. In the first part of the paper, we investigate approximation algorithms for the single disk problem.  ...  Minimizing stall time in parallel disk systems In this section, we present a polynomial time algorithm for systems with D parallel disks that, given a request sequence , computes a prefetching/caching  ... 
doi:10.1145/777429.777431 fatcat:ycsuaifu7fdxpmxrex6c3gxani

Integrated prefetching and caching in single and parallel disk systems

Susanne Albers, Markus Büttner
2003 Proceedings of the fifteenth annual ACM symposium on Parallel algorithms and architectures - SPAA '03  
In the second part of the paper, we investigate the problem of minimizing stall time in parallel disk systems.  ...  We study integrated prefetching and caching in single and parallel disk systems. In the first part of the paper, we investigate approximation algorithms for the single disk problem.  ...  Minimizing stall time in parallel disk systems In this section, we present a polynomial time algorithm for systems with D parallel disks that, given a request sequence , computes a prefetching/caching  ... 
doi:10.1145/777412.777431 dblp:conf/spaa/AlbersB03 fatcat:64tblaholzbghm6ltmuvebp7ji

Multi-threaded Output in CMS using ROOT

Daniel Riley, Christopher Jones, A. Forti, L. Betev, M. Litmaath, O. Smirnova, P. Hristov
2019 EPJ Web of Conferences  
We examine the architecture of the new parallel output module, the specific accommodations and modifications that were made to ensure compatibility with the CMS framework scheduler, and the performance  ...  In this paper we explore the changes made to the CMS framework and our ROOT output module to overcome the previous scaling limits, using two new ROOT features: the TBufferMerger asynchronous file merger  ...  Stall graphs comparing the standard output module with IMT disabled and the parallel output module with IMT enabled showing a substantial reduction in stalls and total processing time (note the change  ... 
doi:10.1051/epjconf/201921402016 fatcat:p5bgyvfzkvdxnjcpk7zwrsuttm
« Previous Showing results 1 — 15 out of 5,178 results