Filters








193,792 Hits in 3.1 sec

Computational caches

Amos Waterland, Elaine Angelino, Ekin D. Cubuk, Efthimios Kaxiras, Ryan P. Adams, Jonathan Appavoo, Margo Seltzer
2013 Proceedings of the 6th International Systems and Storage Conference on - SYSTOR '13  
Caching is a well-known technique for speeding up computation. We cache data from file systems and databases; we cache dynamically generated code blocks; we cache page translations in TLBs.  ...  We propose to cache the act of computation, so that we can apply it later and in different contexts.  ...  cache, and (5) using the computational cache to speed up execution.  ... 
doi:10.1145/2485732.2485749 dblp:conf/systor/WaterlandACKAAS13 fatcat:jiyblu3rjfeh7ea4gbrecbuxfa

Cache oblivious stencil computations

Matteo Frigo, Volker Strumpen
2005 Proceedings of the 19th annual international conference on Supercomputing - ICS '05  
We present a cache oblivious algorithm for stencil computations, which arise for example in finite-difference methods. Our algorithm applies to arbitrary stencils in n-dimensional spaces.  ...  On an "ideal cache" of size Z, our algorithm saves a factor of Θ(Z 1/n ) cache misses compared to a naive algorithm, and it exploits temporal locality optimally throughout the entire memory hierarchy.  ...  Cache oblivious algorithms for special cases of stencil computations have been proposed before.  ... 
doi:10.1145/1088149.1088197 dblp:conf/ics/FrigoS05 fatcat:gckwnarnbvh7dis5osheiw22s4

Mobile Communications, Computing and Caching Resources Optimization for Coded Caching with Device Computing [article]

Yingjiao Li, Zhiyong Chen, Meixia Tao
2020 arXiv   pre-print
In this paper, we design a general coded caching with device computing strategy for content computation, e.g., virtual reality (VR) rendering, to minimize the average transmission bandwidth with the caching  ...  We thus formulate a joint coded caching and computing optimization problem to decide whether the mobile devices cache the input data or the output data, which tasks to be coded cached and which tasks to  ...  The paper [20] leverages the cache-aided and computing-aided edge server for proactive computing and caching the frames.  ... 
arXiv:2002.06090v1 fatcat:oopvxjgjkna2threyqo3cr7ira

Predictive Caching in Computer Grids

E. Rappos, S. Robert
2013 2013 13th IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing  
We present a model for predictive caching where a shared cache is used to improve performance across a grid.  ...  Unlike local caching mechanisms, shared, grid or cloudbased caches incur high costs or latency associated with the additional data transfer.  ...  INTRODUCTION Caching constitutes a fundamental mechanism for improving performance in a variety of networked computer systems and applications.  ... 
doi:10.1109/ccgrid.2013.33 dblp:conf/ccgrid/RapposR13 fatcat:zwlgzvaf4jb5hfmslkmuvezziy

Static caching for incremental computation

Yanhong A. Liu, Scott D. Stoller, Tim Teitelbaum
1998 ACM Transactions on Programming Languages and Systems  
Since every nontrivial program computes by iteration or recursion, the cache-and-prune method provides a systematic approach for general program optimization via caching by computing each iteration or  ...  This article describes a systematic method, called cache-and-prune, for addressing these problems in incremental computation.  ...  This section proposes two methods, selective caching and cache-and-prune, b u t w e f a vor the cache-and-prune method and thus explore it in Static Caching for Incremental Computation 551 the sections  ... 
doi:10.1145/291889.291895 fatcat:eq4ozplbqfh5no46wgntvc7bay

Incremental computation via function caching

W. Pugh, T. Teitelbaum
1989 Proceedings of the 16th ACM SIGPLAN-SIGACT symposium on Principles of programming languages - POPL '89  
Computing g(x') without function caching requires computing g(y) for each y in du(x').  ...  Incremental set opwatlons Assume that the computations involved in computing A op B are stored in the cache and consider computing A' op B', where A' is similar to A and B' is similar to B.  ...  Appendix Function caching and stable decompositions provide a new paradigm for incremental evafuation.  ... 
doi:10.1145/75277.75305 dblp:conf/popl/PughT89 fatcat:otmdlf6qxzghtnk6jl7z4lhnei

A tunable cache for approximate computing

Magnus Själander, Nina Shariati Nilsson, Stefanos Kaxiras
2014 Proceedings of the 2014 IEEE/ACM International Symposium on Nanoscale Architectures - NANOARCH '14  
In this work we investigate the use of multivalue devices to design a cache that can tune the amount of resources used to store application data.  ...  These devices have different features than CMOS, such as the possibility for multivalue logic, which present new opportunities when designing computer systems.  ...  Approximate computing is an approach where accuracy is traded against other factors like performance, power, and energy [3] .  ... 
doi:10.1145/2770287.2770309 fatcat:qwdajhwswrapdpvih32lb2ojou

A reconfigurable multifunction computing cache architecture

Huesung Kim, A.K. Somani, A. Tyagi
2001 IEEE Transactions on Very Large Scale Integration (vlsi) Systems  
In this paper, we present a cache architecture to convert a cache into a computing unit for either of the following two structured computations, FIR and DCT/IDCT.  ...  Therefore, the cache can perform computations when it is reconfigured as a function unit.  ...  Whenever one of these cache modules is converted into a computing unit, the associativity of the cache drops or vice versa.  ... 
doi:10.1109/92.931228 fatcat:ew3tcd3vqrgc7nvj2nlizphejm

Distributed Reproducible Research Using Cached Computations

Roger D. Peng, Sandrah P. Eckel
2009 Computing in science & engineering (Print)  
We present a framework in which reproducible research can be conducted and distributed via cached computations and describe specific tools for both authors and readers.  ...  The cacheSweave and stashR packages together provide tools for caching computational results in a key-value style database which can be published to a public repository for readers to download.  ...  Caching computations The cacheSweave package is the first package in our toolkit and it allows an author to cache the results of a computation in a code chunk by setting the cache=true option in the code  ... 
doi:10.1109/mcse.2009.6 fatcat:uzgusrgyd5cexee7qewk2gb4yy

A reconfigurable multi-function computing cache architecture

Hue-Sung Kim, Arun K. Somani, Akhilesh Tyagi
2000 Proceedings of the 2000 ACM/SIGDA eighth international symposium on Field programmable gate arrays - FPGA '00  
In this paper, we present a cache architecture to convert a cache into a computing unit for either of the following two structured computations, FIR and DCT/IDCT.  ...  Therefore, the cache can perform computations when it is reconfigured as a function unit.  ...  Whenever one of these cache modules is converted into a computing unit, the associativity of the cache drops or vice versa.  ... 
doi:10.1145/329166.329185 dblp:conf/fpga/KimST00 fatcat:ametua7cbzezbjc4phdynizeoe

Radiance Caching for Efficient Global Illumination Computation

J. Krivanek, P. Gautron, S. Pattanaik, K. Bouatouch
2005 IEEE Transactions on Visualization and Computer Graphics  
We also propose a novel method for computing translational radiance gradient at a point.  ...  In particular we extend the irradiance caching scheme of [WRC88] to cache and interpolate directional incoming radiance instead of irradiance.  ...  -Irradiance caching computes the indirect diffuse term. else // low-frequency BRDF; use radiance caching if ( one or more radiance cache records exist near p ) then -Compute the HSH coefficients of the  ... 
doi:10.1109/tvcg.2005.83 pmid:16144252 fatcat:gi7fbpu3afh5diop6vh2sfk7hq

Radiance caching for efficient global illumination computation

Jaroslav Křivánek, Pascal Gautron, Sumanta Pattanaik, Kadi Bouatouch
2008 ACM SIGGRAPH 2008 classes on - SIGGRAPH '08  
We also propose a novel method for computing translational radiance gradient at a point.  ...  In particular we extend the irradiance caching scheme of [WRC88] to cache and interpolate directional incoming radiance instead of irradiance.  ...  -Irradiance caching computes the indirect diffuse term. else // low-frequency BRDF; use radiance caching if ( one or more radiance cache records exist near p ) then -Compute the HSH coefficients of the  ... 
doi:10.1145/1401132.1401228 dblp:conf/siggraph/KrivanekGPB08 fatcat:tl36jylu4fcv5ddn2oyjnfi4xq

Cache oblivious parallelograms in iterative stencil computations

Robert Strzodka, Mohammed Shaheen, Dawid Pajak, Hans-Peter Seidel
2010 Proceedings of the 24th ACM International Conference on Supercomputing - ICS '10  
We present a new cache oblivious scheme for iterative stencil computations that performs beyond system bandwidth limitations as though gigabytes of data could reside in an enormous on-chip cache.  ...  Conclusions We have presented CORALS, a cache oblivious scheme for iterative stencil computations that performs beyond system bandwidth limitations.  ...  This is an expected result from a more specialized cache oblivious algorithm, but has not been demonstrated before.  ... 
doi:10.1145/1810085.1810096 dblp:conf/ics/StrzodkaSPS10 fatcat:5lnq6rbplbdexflrmqifj5mzs4

Holistic design and caching in mobile computing

Mwaffaq Otoom, JoAnn M. Paul
2008 Proceedings of the 6th IEEE/ACM/IFIP international conference on Hardware/Software codesign and system synthesis - CODES/ISSS '08  
We utilize application trends analysis, focused on webpage content, in order to examine the design of mobile computers more holistically.  ...  We find that both Internet bandwidth and processing local to the computing device is being wasted by re-transmission of formatting data.  ...  Conclusions We illustrated a simple caching scheme, which resulted in a performance increase of 84%, a decrease in power consumption by 71%, and an order of magnitude savings in communications bandwidth  ... 
doi:10.1145/1450135.1450161 dblp:conf/codes/OtoomP08 fatcat:gyb6qoustvaphigxyacer3uidu

Information-Theoretic Caching: Sequential Coding for Computing

Chien-Yi Wang, Sung Hoon Lim, Michael Gastpar
2016 IEEE Transactions on Information Theory  
Although the two caching models have a similar behavior for the single-user case, it is shown through a two-user example that the two caching models behave differently in general.  ...  Under the paradigm of caching, partial data is delivered before the actual requests of users are known.  ...  Decoder 1 then computes the reconstruction sequenceŝ 1i = g(u i (ℓ 2 ,l 3 ), v i (ℓ v ,l 1 ), y i ) for i ∈ [1 : k].  ... 
doi:10.1109/tit.2016.2604851 fatcat:z2eqetjoxjgn7nyp7s2h7qprxe
« Previous Showing results 1 — 15 out of 193,792 results