Filters








732 Hits in 5.3 sec

Effectiveness of Dynamic Prefetching in Multiple-Writer Distributed Virtual Shared-Memory Systems

Magnus Karlsson, Per Stenström
1997 Journal of Parallel and Distributed Computing  
This approach uses the access history of each page to guide which pages to prefetch.  ...  Based on detailed architectural simulations and seven scientific applications we find that our prefetch algorithm can remove a vast majority of the remote operations which improves the performance of all  ...  This research has been funded in part by the Swedish National Board for Industrial and Technical Development (NUTEK) under project numbers P2363-1, P855 and by the Swedish Research Council on Engineering  ... 
doi:10.1006/jpdc.1997.1333 fatcat:e736a5vfczdilhzjbbnjlrva3a

Region-Based Prefetch Techniques for Software Distributed Shared Memory Systems

Jie Cai, Peter E. Strazdins, Alistair P. Rendell
2010 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing  
A good prefetch strategy can reduce this cost. We develop two prefetch techniques, TReP and HReP, which are based on the execution history of each parallel region.  ...  On average, TReP achieves an efficiency (ratio of pages prefetched that were subsequently accessed) of 96% and a coverage (ratio of access faults avoided by prefetches) of 65%.  ...  In order to record the page fault history, only the faulting page is set to valid after the requested data arrives. After the group is fetched, it is freed.  ... 
doi:10.1109/ccgrid.2010.16 dblp:conf/ccgrid/CaiSR10 fatcat:fur25po2cjbd3dtvjb3vqv5sf4

Comparing latency-tolerance techniques for software dsm systems

R. Pinto, R. Bianchini, C.L. Amorim
2003 IEEE Transactions on Parallel and Distributed Systems  
More specifically, we focus on data prefetching, update-based coherence, and single-writer optimizations for page-based software DSMs.  ...  results with six parallel applications show that, when these techniques are carefully combined, they can provide running time and speedup improvements of up to 54 percent and 110 percent, respectively, on  ...  The fault behavior of these applications can be predicted based on their past history of faults. Adaptive++ focuses solely on repeated-phase and repeated-stride types of regular applications.  ... 
doi:10.1109/tpds.2003.1247677 fatcat:dbndp7do3bb6veaqukqcori3ia

Using aspectC to improve the modularity of path-specific customization in operating system code

Yvonne Coady, Gregor Kiczales, Mike Feeley, Greg Smolyn
2001 Proceedings of the 8th European software engineering conference held jointly with 9th ACM SIGSOFT international symposium on Foundations of software engineering - ESEC/FSE-9  
As an initial experiment, we used AspectC to modularize the implementation of prefetching within page fault handling in the FreeBSD OS kernel.  ...  Layered architecture in operating system code is often compromised by execution path-specific customizations such as prefetching, page replacement and scheduling strategies.  ...  Acknowledgments: Many thanks to Stephan Gudmundson, Jan Hannemann, Gail Murphy and Andrew Warfield for their insightful comments on drafts of this paper.  ... 
doi:10.1145/503222.503223 fatcat:s3hf67yi5baavb655khwqs6jxq

Using aspectC to improve the modularity of path-specific customization in operating system code

Yvonne Coady, Gregor Kiczales, Mike Feeley, Greg Smolyn
2001 Software engineering notes  
As an initial experiment, we used AspectC to modularize the implementation of prefetching within page fault handling in the FreeBSD OS kernel.  ...  Layered architecture in operating system code is often compromised by execution path-specific customizations such as prefetching, page replacement and scheduling strategies.  ...  Acknowledgments: Many thanks to Stephan Gudmundson, Jan Hannemann, Gail Murphy and Andrew Warfield for their insightful comments on drafts of this paper.  ... 
doi:10.1145/503271.503223 fatcat:hrkuy4uv7ngezhbye7t7jo7kx4

Using aspectC to improve the modularity of path-specific customization in operating system code

Yvonne Coady, Gregor Kiczales, Mike Feeley, Greg Smolyn
2001 Proceedings of the 8th European software engineering conference held jointly with 9th ACM SIGSOFT international symposium on Foundations of software engineering - ESEC/FSE-9  
As an initial experiment, we used AspectC to modularize the implementation of prefetching within page fault handling in the FreeBSD OS kernel.  ...  Layered architecture in operating system code is often compromised by execution path-specific customizations such as prefetching, page replacement and scheduling strategies.  ...  Acknowledgments: Many thanks to Stephan Gudmundson, Jan Hannemann, Gail Murphy and Andrew Warfield for their insightful comments on drafts of this paper.  ... 
doi:10.1145/503209.503223 dblp:conf/sigsoft/CoadyKFS01 fatcat:pab6n4p3enhfbmaowrkkjlrjoi

Effectively Prefetching Remote Memory with Leap [article]

Hasan Al Maruf, Mosharaf Chowdhury
2019 arXiv   pre-print
At its core, Leap employs an online, majority-based prefetching algorithm, which increases the page cache hit rate.  ...  In this paper, we propose Leap, a prefetching solution for remote memory accesses due to memory disaggregation.  ...  The aggressiveness of this prefetcher depends on the accuracy of the past prefetch. • Linux Read-Ahead prefetches an aligned block of pages containing the faulted page [72] .  ... 
arXiv:1911.09829v1 fatcat:4jhmagwqhfcbdfb37424id2lmi

A Software-Managed Approach to Die-Stacked DRAM

Mark Oskin, Gabriel H. Loh
2015 2015 International Conference on Parallel Architecture and Compilation (PACT)  
prefetcher that extends classic hardware prefetching algorithms to the page level, leading to 39% speedup.  ...  While much recent effort has focused on hardware-based techniques for using die-stacked memory (e.g., caching), in this paper we explore what it takes for a softwaredriven approach to be effective.  ...  Depth refers to how far in advance (in unit strides) a page address should be prefetched. History is the number of past addresses to use to base prefetches from.  ... 
doi:10.1109/pact.2015.30 dblp:conf/IEEEpact/OskinL15 fatcat:fn4mpetyhfhp5fven24s3iort4

Adaptive DSM-runtime behavior via speculative data distribution [chapter]

Frank Mueller
1999 Lecture Notes in Computer Science  
Histories of DSM accesses are used to anticipate future data requests. DSM-Threads implement an adaptive runtime history that speculatively distributes data based on access histories.  ...  These histories are kept locally on a node and are distinguished by each thread on a node.  ...  [2] proposed prefetching schemes using histories of length two and sequential prefetching but still take page faults on already prefetched pages to ensure that prefetched data will be used.  ... 
doi:10.1007/bfb0097939 fatcat:t6ajbakllrgwbcaowt2pdk5fjy

Deep Learning based Data Prefetching in CPU-GPU Unified Virtual Memory [article]

Xinjian Long and Xiangyang Gong and Huiyang Zhou
2022 arXiv   pre-print
However, on-demand paging soon becomes a performance bottleneck of UVM due to the high latency caused by page table walks and data migration over interconnect.  ...  This paper proposes a novel approach for page prefetching for UVM through deep learning. We first show that a powerful Transformer learning model can provide high accuracy for UVM page prefetching.  ...  locality-based page prefetching scheme in CPU-GPU UVM.  ... 
arXiv:2203.12672v1 fatcat:b5mxxkhvxzbivhb4aw67hvdafi

Optimal prefetching via data compression

Jeffrey Scott Vitter, P. Krishnan
1996 Journal of the ACM  
Our prediction algorithms for prefetching are novel in that they are based on data compression techniques that are both theoretically optimal and good in practice.  ...  We show for powerful models such as Markov sources and mth order Markov sources that the page fault rates incurred by our prefetching algorithms are optimal in the limit for almost all sequences of page  ...  We say that an algorithm is online if it must make its decisions based only on the past history. An offline algorithm can use the knowledge of the future.  ... 
doi:10.1145/234752.234753 fatcat:uca2h5khtreqzp2fl2kp6w2cma

Optimal Prediction for Prefetching in the Worst Case

P. Krishnan, Jeffrey Scott Vitter
1998 SIAM journal on computing (Print)  
We prove analytically that the fault rate of our online prefetching algorithm converges almost surely for every page request sequence to the fault rate of the optimal finite state prefetcher for the sequence  ...  Previous analytical work on prefetching [J. Assoc. Comput. Mach., 143 (1996), pp. 771-793] consisted of modeling the user as a probabilistic Markov source.  ...  An algorithm is online if it must make its decisions based only on the past history. An offline algorithm can use knowledge of the future.  ... 
doi:10.1137/s0097539794261817 fatcat:d6enta3mereuxbqebllpkpohye

Access pattern based stream buffer management scheme for portable media players

Heejin Ahn, Seongjin Cho, Hyunik Na, Hwansoo Han
2009 IEEE transactions on consumer electronics  
We devise a profile-based memory allocation scheme for multimedia stream buffers with the adoption of page prefetching and pinning.  ...  Such page fault mechanism provided by the operating system often becomes the main hindrance to the uninterrupted playback of multimedia players on portable electronic devices.  ...  ., who first brought up the page fault problem of multimedia applications to us. Many comments and discussion with them guided us to better evaluate the issues and pursue our solution.  ... 
doi:10.1109/tce.2009.5278022 fatcat:j2tlcdj6jfanngximq7swwyhey

Compiler-based I/O prefetching for out-of-core applications

Angela Demke Brown, Todd C. Mowry, Orran Krieger
2001 ACM Transactions on Computer Systems  
In our schemethe compiler provides the crucial information on future access patterns without burdening the programmer; the operating system supports nonbinding prefetch and release hints for managing I  ...  /O; and the operating system cooperates with a run-time layer to accelerate performance by adapting to dynamic behavior and minimizing prefetch overhead.  ...  When a page fault occurs, the kernel checks if the faulting page was prefetched, and updates a set of counters based on the result.  ... 
doi:10.1145/377769.377774 fatcat:z3nrthbyxfhrffsfsx3u6zlugy

Synergistic coupling of SSD and hard disk for QoS-aware virtual memory

Ke Liu, Xuechen Zhang, Kei Davis, Song Jiang
2013 2013 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)  
Using a history of page access patterns HybridSwap dynamically creates an out-of-memory virtual memory page layout on the swap space spanning the SSD and hard disk such that random reads are served by  ...  capacity, power consumption, and price, solid state disk (SSD) has good potential to be employed as an extension of DRAM (memory), such that applications with large working sets could run efficiently on  ...  When HybridSwap is used 49% of page faults are eliminated by effective swap-page prefetching based on faults history.  ... 
doi:10.1109/ispass.2013.6557143 dblp:conf/ispass/LiuZDJ13 fatcat:qromydlkqfgc3dfmsd647xe2a4
« Previous Showing results 1 — 15 out of 732 results