6,311 Hits in 6.0 sec

Thermodynamic-RAM technology stack

M. Alexander Nugent, Timothy W. Molter
2017 International Journal of Parallel, Emergent and Distributed Systems  
The levels of the Thermodynamic-RAM technology stack include the memristor, synapse, AHaH node, kT-RAM, instruction set, sparse spike encoding, kT-RAM emulator, and SENSE server.  ...  We introduce a technology stack or specification describing the multiple levels of abstraction and specialization needed to implement a neuromorphic processor (NPU) based on the previously-described concept  ...  Campbell from Boise State University for graciously providing us with memristor device data.  ... 
doi:10.1080/17445760.2017.1314472 fatcat:e2mm7stwzzhldklms72vywmd6a

Stacking machine learning classifiers to identify Higgs bosons at the LHC

A. Alves
2017 Journal of Instrumentation  
that combining an ensemble of simpler and faster ML algorithms with MVA tools is a better approach than building a complex state-of-art algorithm for cut-and-count.  ...  Machine learning (ML) algorithms have been employed in the problem of classifying signal and background events with high accuracy in particle physics.  ...  Ankit Patel [42] and Kuver Sinha kindly read the manuscript and made valuable suggestions and comments.  ... 
doi:10.1088/1748-0221/12/05/t05005 fatcat:rhcs2nv4zzdkzbl7xc3a6qrv64

Dealing with Driver Failures in the Storage Stack

Jorrit N. Herder, David C. van Moolenbroek, Raja Appuswamy, Bingzheng Wu, Ben Gras, Andrew S. Tanenbaum
2009 2009 Fourth Latin-American Symposium on Dependable Computing  
The filter uses checksumming and mirroring in order to achieve end-toend integrity and provide hard guarantees for detection of silent data corruption and recovery of lost data.  ...  For example, transparent recovery is possible for network drivers because the TCP protocol can detect garbled and lost packets and retransmit the data.  ...  Since the file system is aware of the importance and on-disk layout of data structures, efficient checksumming algorithms are possible.  ... 
doi:10.1109/ladc.2009.12 dblp:conf/ladc/HerderMAWGT09 fatcat:g7ku5m662rfr7cpqlxacqxzdji

Designing a Storage Software Stack for Accelerators

Shinichi Awamoto, Erich Focht, Michio Honda
2020 USENIX Workshop on Hot Topics in Storage and File Systems  
In this paper, we explore the case for direct storage access inside the accelerator applications, and discuss the problem, design options and benefits of this architecture.  ...  Although modern accelerator devices, such as vector engines and SmartNICs, are equipped with general purpose CPUs, access to the storage needs the mediation of the host kernel and CPUs, resulting in latency  ...  Acknowledgments We are grateful to anonymous reviewers for their insightful comments and Hajime Tazaki at IIJ for his help with LKL experiments.  ... 
dblp:conf/hotstorage/AwamotoFH20 fatcat:xwv7sqa4ufdrni75ss36ru7pyq

Stacked Generalizations in Imbalanced Fraud Data Sets using Resampling Methods [article]

Kathleen Kerwin, Nathaniel D. Bastian
2020 arXiv   pre-print
This study uses stacked generalization, which is a two-step process of combining machine learning methods, called meta or super learners, for improving the performance of algorithms in step one (by minimizing  ...  the error rate of each individual algorithm to reduce its bias in the learning set) and then in step two inputting the results into the meta learner with its stacked blended output (demonstrating improved  ...  This is when more formal studies on the effectiveness and efficiency of the machine learning process are conducted.  ... 
arXiv:2004.01764v1 fatcat:mcy2rnk2dffn3awn3kfc67pdvm

Using Deduplicating Storage for Efficient Disk Image Deployment

Xing Lin, Mike Hibler, Eric Eide, Robert Ricci
2015 Proceedings of the 10th EAI International Conference on Testbeds and Research Infrastructures for the Development of Networks & Communities  
We also propose a new chunking algorithm, called AFC, which enables fixed-size chunking for deduplicating allocated disk sectors.  ...  Driven by our experience in operating the Emulab site at the University of Utah-a long-lived and heavily-used testbed-we have created a new service for efficiently storing and deploying disk images.  ...  ACKNOWLEDGMENTS We thank Fred Douglis, Eddie Kohler, Raghuveer Pullakandam, John Regehr, Gokul Soundararajan, and the anonymous TRIDENT-COM '15 reviewers for their comments on drafts of this paper.  ... 
doi:10.4108/icst.tridentcom.2015.259963 dblp:journals/sis/LinHER15 fatcat:5qstfmuidvg5to74hswuzylwcm

BLAS-on-flash: An Efficient Alternative for Large Scale ML Training and Inference?

Suhas Jayaram Subramanya, Harsha Vardhan Simhadri, Srajan Garg, Anil Kag, Venkatesh Balasubramanian
2019 Symposium on Networked Systems Design and Implementation  
This often forces an expensive redesign of the algorithm for distributed platforms such as parameter servers and Spark.  ...  We demonstrate that not only can such a library provide near in-memory performance for BLAS, but can also be used to write implementations of complex algorithms such as eigensolvers that outperform in-memory  ...  Acknowledgments The authors would like to thank Anirudh Badam, Ravi Kannan, Muthian Sivathanu, and Manik Varma for their useful comments and advice.  ... 
dblp:conf/nsdi/SubramanyaSGKB19 fatcat:bji26e4w4fahrivcab7myzfd5q

Study on the limitations of Stacking Technique for Bandwidth Improvement of Microstrip Patch Antennas

2021 American journal of science & engineering  
The SHAP Value is a method for assigning This is why these types of algorithms are called payouts to players of coalition game depending on unsupervised learning algorithms  ...  Calculated SHAP value In this type of algorithm, a machine (actually its for each data instance shows whether data is high controller program) is trained to make a  ... 
doi:10.15864/ajse.2305 fatcat:mr7d7o3rsfe4npbo6apifsi4yy

Separated Control and Data Stacks to Mitigate Buffer Overflow Exploits

Christopher Kugler, Tilo Müller
2015 EAI Endorsed Transactions on Security and Safety  
To substantiate the practicability of our approach, we provide SCADS as an open source patch for the LLVM compiler infrastructure.  ...  In this article, we present a novel compiler-level protection called SCADS: Separated Control and Data Stacks that protects return addresses and saved frame pointers on a separate stack, called the control  ...  The SCADS compiler builds a chain of machine instructions which decrease the stack pointer by a page and perform a write access at its current position successively, until the needed stack frame size has  ... 
doi:10.4108/eai.5-10-2015.150477 fatcat:tp2u3h6mhbehtjtr7veigoui2a

Forecasting Electricity Load With Hybrid Scalable Model Based on Stacked Non Linear Residual Approach

Ayush Sinha, Raghav Tayal, Aamod Vyas, Pankaj Pandey, O. P. Vyas
2021 Frontiers in Energy Research  
Finally, the efficacy of the proposed model is established through comparative studies with state-of-the-art models on Household Power Consumption Dataset (UCI machine learning repository) and Ontario  ...  Since it should be generated when demanded, it is necessary to forecast its demand accurately and efficiently.  ...  Notwithstanding of the model simplicity, Elmann RNNs are difficult to prepare because of less efficiency of gradient (back) propagation.  ... 
doi:10.3389/fenrg.2021.720406 fatcat:axvxzi7wgzgu3fmneshdwehfwm


Sowmiyaa P .
2015 International Journal of Research in Engineering and Technology  
Our results show that enables highly scalable implementations of algorithms using Hadoop for the purpose of spatial data processing problems.  ...  Inspite of developing these implementations requires specialized knowledge and user friendly.  ...  (by removing single points of failure) and cut back latency.  ... 
doi:10.15623/ijret.2015.0403030 fatcat:2vu3n7wysvdy7p367wnuuj75qe

dsync: Efficient Block-wise Synchronization of Multi-Gigabyte Binary Data

Thomas Knauth, Christof Fetzer
2013 USENIX Large Installation Systems Administration Conference  
Backing up important data is an essential task for system administrators to protect against all kinds of failures.  ...  We address the problem of efficient, periodic, multi-gigabyte state synchronization.  ...  for valuable comments on drafts of this paper.  ... 
dblp:conf/lisa/KnauthF13 fatcat:bi37v6w4jffz3ggjea6jyimhvu

Finding Needle in the Case-Stack: Effective Remote Monitoring of Courts

Amn Rahman, Hassan Nawaz, Omer Naeem, Fareed Zaffar, Farooq Naseer, Adeel Zaffar
2014 2014 47th Hawaii International Conference on System Sciences  
Efficient supervision and monitoring of courts can help in resolving some of the key problems in the system.  ...  Most cases are trapped in the system due to a huge backlog of cases, excessive litigation in urban areas, corruption by the court staff and lack of proper monitoring of the functioning of courts.  ...  The writing was often scribbled and smudged and only a trained eye could make sense of most of the documents.  ... 
doi:10.1109/hicss.2014.241 dblp:conf/hicss/RahmanNNZNZ14 fatcat:pjnfjdq57zfxpiccf4am7t72l4

Architectural document of the ENES CDI software stack (D10.1) [article]

S. Fiore, P. Nassisi, F. Antonio, L. Barring, A. Ben Nasser, K. Berger, D. Hassell, M. Juckes, P. Kershaw, S. Kindermann, G. Levavasseur, A. Nuzzo (+6 others)
2020 Zenodo  
This work will lead the development of the ENES CDI software stack architecture during the project timeframe and it will be periodically revised to include emerging needs and ensure full compliance with  ...  Taking into account the requirements collected with the milestone M10.1, it proposes an overall architectural view of the ENES Climate Data Infrastructure software stack and it goes into the details of  ...  The back-end software stack comprises utility libraries to automate [NFR#4] creation and publication of standardised documents and to store and systemise semantics for applicable controlled vocabularies  ... 
doi:10.5281/zenodo.4309892 fatcat:4qgocrkt4bey7gc2hjvxhdyamu

Fast FCoE: An Efficient and Scale-Up Multi-core Framework for FCoE-Based SAN Storage Systems

Yunxiang Wu, Fang Wang, Yu Hua, Dan Feng, Yuchong Hu, Jingning Liu, Wei Tong
2015 2015 44th International Conference on Parallel Processing  
Our results demonstrate that FastFCoE achieves efficient and scalable I/O throughput, obtaining 1107.3K/831.3K IOPS (5.43/4.88 times as much as Open-FCoE stack) for read/write requests.  ...  For scalable performance, existing works focus on improving the efficiency of lock algorithm or reducing the number of synchronization points to decrease the synchronization overhead.  ...  This method cuts the extra inter-layer and intra-layer cost and significantly improves the I/O efficiency (detailed analysis in subsection IV-B). 2) Multi-core Scalability Analysis: For scalability purpose  ... 
doi:10.1109/icpp.2015.42 dblp:conf/icpp/WuWHFHLT15 fatcat:dv23exipxna77n4ot6zsv5tbva
« Previous Showing results 1 — 15 out of 6,311 results