Filters








30,019 Hits in 4.7 sec

Approximate computing with unreliable dynamic memories

Shrikanth Ganapathy, Adam Teman, Robert Giterman, Andreas Burg, Georgios Karakonstantis
2015 2015 IEEE 13th International New Circuits and Systems Conference (NEWCAS)  
CONCLUSIONS In this paper, we considered a paradigm shift towards approximate computing with unreliable memories.  ...  The approximate computing paradigm includes the development of processors and software that may not always produce 100% precise results, but their output fidelity is acceptable for human consumption with  ... 
doi:10.1109/newcas.2015.7182027 dblp:conf/newcas/GanapathyTGBK15 fatcat:njnkvw47avdufmahjalyszkv3m

Verifying quantitative reliability for programs that execute on unreliable hardware

Michael Carbin, Sasa Misailovic, Martin C. Rinard
2013 Proceedings of the 2013 ACM SIGPLAN international conference on Object oriented programming systems languages & applications - OOPSLA '13  
For example, approximate computing applications (such as multimedia processing, machine learning, and big data analytics) can often naturally tolerate soft errors.  ...  Emerging high-performance architectures are anticipated to contain unreliable components that may exhibit soft errors, which silently corrupt the results of computations.  ...  Unreliable Inputs. Some computations (for example, computations that read unreliable sensors or work with noisy data) may process unreliable inputs.  ... 
doi:10.1145/2509136.2509546 dblp:conf/oopsla/CarbinMR13 fatcat:2wqtxohpkzg6bjgb4ofyqjtq3y

Verifying quantitative reliability for programs that execute on unreliable hardware

Michael Carbin, Sasa Misailovic, Martin C. Rinard
2013 SIGPLAN notices  
For example, approximate computing applications (such as multimedia processing, machine learning, and big data analytics) can often naturally tolerate soft errors.  ...  Abstract Emerging high-performance architectures are anticipated to contain unreliable components that may exhibit soft errors, which silently corrupt the results of computations.  ...  Unreliable Inputs. Some computations (for example, computations that read unreliable sensors or work with noisy data) may process unreliable inputs.  ... 
doi:10.1145/2544173.2509546 fatcat:6pujz4n4jnhhbhnl6gzrraitb4

Verifying quantitative reliability for programs that execute on unreliable hardware

Michael Carbin, Sasa Misailovic, Martin C. Rinard
2016 Communications of the ACM  
For example, approximate computing applications (such as multimedia processing, machine learning, and big data analytics) can often naturally tolerate soft errors.  ...  Emerging high-performance architectures are anticipated to contain unreliable components that may exhibit soft errors, which silently corrupt the results of computations.  ...  Unreliable Inputs. Some computations (for example, computations that read unreliable sensors or work with noisy data) may process unreliable inputs.  ... 
doi:10.1145/2958738 fatcat:i47h6rkwjff4fknhcxisq4hzfy

Synergistic Architecture and Programming Model Support for Approximate Micropower Computing

Giuseppe Tagliavini, Davide Rossi, Luca Benini, Andrea Marongiu
2015 2015 IEEE Computer Society Annual Symposium on VLSI  
Using aggressive voltage scaling can reduce power consumption, but memory operations become unreliable.  ...  The voltage supply level is dynamically adjusted according to tolerance policies, with the overall goal of minimizing energy in full compliance with precision constraints.  ...  Approximate computing has recently emerged as a promising approach to design energy-efficient digital systems working at unreliable voltage levels.  ... 
doi:10.1109/isvlsi.2015.64 dblp:conf/isvlsi/TagliaviniRBM15 fatcat:do3naxkiibh33lxwddhvyhwiwq

Chisel

Sasa Misailovic, Michael Carbin, Sara Achour, Zichao Qi, Martin C. Rinard
2014 SIGPLAN notices  
The accuracy of an approximate computation is the distance between the result that the computation produces and the corresponding fully accurate result.  ...  We present Chisel, a system for reliability-and accuracyaware optimization of approximate computational kernels that run on approximate hardware platforms.  ...  with the approximate computation.  ... 
doi:10.1145/2714064.2660231 fatcat:a27g5ske3jcdxg7lc4bhxfew24

Chisel

Sasa Misailovic, Michael Carbin, Sara Achour, Zichao Qi, Martin C. Rinard
2014 Proceedings of the 2014 ACM International Conference on Object Oriented Programming Systems Languages & Applications - OOPSLA '14  
The accuracy of an approximate computation is the distance between the result that the computation produces and the corresponding fully accurate result.  ...  We present Chisel, a system for reliability-and accuracyaware optimization of approximate computational kernels that run on approximate hardware platforms.  ...  with the approximate computation.  ... 
doi:10.1145/2660193.2660231 dblp:conf/oopsla/MisailovicCAQR14 fatcat:sznsnmiupbh7vlrdnpqz2reofa

BDDs Strike Back: Efficient Analysis of Static and Dynamic Fault Trees [article]

Daniel Basgöze, Matthias Volk, Joost-Pieter Katoen, Shahid Khan, Marielle Stoelinga
2022 arXiv   pre-print
State-based techniques are favorable for the more expressive dynamic fault trees (DFT).  ...  This paper combines the best of both worlds by following Dugan's approach: dynamic sub-trees are analysed via model checking Markov models and replaced by basic events capturing the obtained failure probabilities  ...  The artifact contains the three tools, all SFT and DFT benchmark files, scripts to run the experiments as well as tables with detailed results. Random SFTs.  ... 
arXiv:2202.02829v2 fatcat:7hu3ydpwzncpfklp7i23quq7xi

An investigation of the effects of error correcting code on GPU-accelerated molecular dynamics simulations

Ross C. Walker, Robin M. Betz
2013 Proceedings of the Conference on Extreme Science and Engineering Discovery Environment Gateway to Discovery - XSEDE '13  
An argument often put forward for not using GeForce cards is that the results are unreliable due to the lack of ECC.  ...  Molecular dynamics (MD) simulations rely on the accurate evaluation and integration of Newton's equations of motion to propagate the positions of atoms in proteins during a simulation.  ...  This research used resources of the Keeneland Computing Facility at the Georgia Institute of Technology, which is supported by the National Science Foundation under Contract OCI-0910735.  ... 
doi:10.1145/2484762.2484774 dblp:conf/xsede/WalkerB13 fatcat:yak54iljqrg2jobfmsu5twpz3m

Multi-Layer Memory Resiliency

Nikil Dutt, Puneet Gupta, Alex Nicolau, Abbas BanaiyanMofrad, Mark Gottscho, Majid Shoushtari
2014 Proceedings of the The 51st Annual Design Automation Conference on Design Automation Conference - DAC '14  
Two specific exemplars are used to illustrate multilayer memory resilience: first we describe static and dynamic policies to achieve energy savings in caches using aggressive voltage scaling combined with  ...  With memories continuing to dominate the area, power, cost and performance of a design, there is a critical need to provision reliable, high-performance memory bandwidth for emerging applications.  ...  The last column of Figure 1 describes opportunities for relaxed and approximate computing in the face of memory error manifestations at that level of abstraction.  ... 
doi:10.1145/2593069.2596684 dblp:conf/dac/DuttGNBGS14 fatcat:v6rb7n5m5bfj5gbbgloksy7ckq

Optimizing the Energy Efficiency of Unreliable Memories for Quantized Kalman Filtering

Jonathan Kern, Elsa Dupraz, Abdeldjalil Aïssa-El-Bey, Lav R. Varshney, François Leduc-Primeau
2022 Sensors  
We consider that both the quantization and the unreliable memories introduce errors in the computations, and we develop an error propagation model that takes into account these two sources of errors.  ...  This paper presents a quantized Kalman filter implemented using unreliable memories.  ...  Here, we aim to optimize the energy consumption of a Kalman filter implemented with fixed-point quantization [41] and with unreliable memories.  ... 
doi:10.3390/s22030853 pmid:35161599 pmcid:PMC8838220 fatcat:kymf7i6kvjbofbgkvji4zlbydm

Synergistic HW/SW Approximation Techniques for Ultra-Low-Power Parallel Computing

Giuseppe Tagliavini, Davide Rossi, Andrea Marongiu, Luca Benini
2016 IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems  
On the other hand, several applications have inherent tolerance to computation errors, and executing such workloads with approximation has already proven a viable way to reduce energy consumption.  ...  unreliable at low voltages.  ...  ACKNOWLEDGMENT This work is supported by the European FP7 ERC Advanced project MULTITHERMAN (g.a. 291125) and by IcySoC and YINS RTD projects, evaluated by the Swiss NSF and funded by Nano-Tera.ch with  ... 
doi:10.1109/tcad.2016.2633474 fatcat:drbmba5xvbfubh2bdtrawwcnre

Distributed QR Factorization Based on Randomized Algorithms [chapter]

Hana Straková, Wilfried N. Gansterer, Thomas Zemen
2012 Lecture Notes in Computer Science  
However, in situations where computer systems may change dynamically, in particular, when they have unreliable components, algorithms with randomized communication schedule may be an interesting alternative  ...  Moreover, we illustrate that the randomized approaches are well suited for distributed systems with arbitrary topology and potentially unreliable communication, where approaches with fixed communication  ...  system with unreliable communication.  ... 
doi:10.1007/978-3-642-31464-3_24 fatcat:k4ktjvyemfd6dctu2kogmfktqm

Fault-tolerant linear solvers via selective reliability [article]

Patrick G. Bridges, Kurt B. Ferreira, Michael A. Heroux, Mark Hoemmen
2012 arXiv   pre-print
Furthermore, they store most of their data unreliably, and spend most of their time in unreliable mode.  ...  Energy increasingly constrains modern computer hardware, yet protecting computations and data against errors costs energy.  ...  The projected problem requires little memory or solution time computed with the large vectors and operators.  ... 
arXiv:1206.1390v1 fatcat:ttowhmggzrbb5hsawtvli7pgue

Markov models of fault-tolerant memory systems under SEU

L. Schiano, M. Ottavi, F. Lombardi
2004 Records of the 2004 International Workshop on Memory Technology, Design and Testing, 2004.  
It is shown that a duplex memory system encoded with error control codes has a higher reliability than the triplex arrangement.  ...  These arrangements exploit redundancy (either duplex or triplex replication) for dynamic fault-tolerant operation as provided by arbitration (for error detection and output selection) as well as in the  ...  Fig.7 also shows that the unreliability saturates faster for the triplex memory system than for two duplex arrangements with coding.  ... 
doi:10.1109/mtdt.2004.1327982 fatcat:ivxhtxzoofeuvixctfpiied7tq
« Previous Showing results 1 — 15 out of 30,019 results