2,956 Hits in 8.4 sec

Tailbench: a benchmark suite and evaluation methodology for latency-critical applications

Harshad Kasture, Daniel Sanchez
2016 2016 IEEE International Symposium on Workload Characterization (IISWC)  
TailBench includes eight applications that span a wide range of latency requirements and domains, and a harness that implements a robust and statistically sound load-testing methodology.  ...  We present TailBench, a benchmark suite and evaluation methodology that makes latency-critical workloads as easy to run and characterize as conventional, throughput-oriented ones.  ...  This work was supported in part by NSF grant CCF-1318384, a grant from the Qatar Computing Research Institute, and a Google Research Award.  ... 
doi:10.1109/iiswc.2016.7581261 dblp:conf/iiswc/KastureS16 fatcat:2o4yjw4qfrgbtimr3muzwycqne

A Review of Deep Learning-based Human Activity Recognition on Benchmark Video Datasets

Vijeta Sharma, Manjari Gupta, Anil Kumar Pandey, Deepti Mishra, Ajai Kumar
2022 Applied Artificial Intelligence  
We propose a new taxonomy for categorizing the literature as CNN and RNN-based approaches.  ...  A short comparison is also made with the handcrafted feature-based approach and its fusion with deep learning to show the evolution of HAR methods.  ...  (Reining et al. 2019 ) performed a systematic literature review of HAR for production and logistics.  ... 
doi:10.1080/08839514.2022.2093705 fatcat:6on4g3sp3vaktnyyrk72k4mqta

SFMapReduce: An optimized MapReduce framework for Small Files

Fang Zhou, Hai Pham, Jianhui Yue, Hao Zou, Weikuan Yu
2015 2015 IEEE International Conference on Networking, Architecture and Storage (NAS)  
The sizes of files stored and processed in Hadoop are not bound to very large files anymore.  ...  On average, SFMapReduce achieves an improvement on MapReduce processing by 14.5 times and 20.8 times, compared with the original Hadoop and HAR layout.  ...  For Wordcount benchmark, the performance of SFMapReduce overperforms the Hadoop by 14.4x and HAR Hadoop by 21.0x; for Sort, SFMapReduce is better than the original Hadoop by 12.1x and HAR Hadoop by 16.1x  ... 
doi:10.1109/nas.2015.7255218 dblp:conf/nas/ZhouPYZY15 fatcat:mefkkurypjg73priwybmolq2ci

Cost- and QoS-Efficient Serverless Cloud Computing [article]

Chavit Denninnart
2020 arXiv   pre-print
With the goal of providing a low cost and a high QoS, the serverless computing paradigm offers opportunities that can be harnessed to attain the goals.  ...  We determined the minimum chance of success thresholds for tasks to pass to get scheduled and executed.  ...  Accordingly, in this chapter, we first strategically benchmark a variety of To systematically analyze the evaluation results and eliminate the impact of different video formats that affect the execution-time  ... 
arXiv:2011.11711v1 fatcat:ra66iigninem7gqibmivsn3dba

Benchmarking Graph Data Management and Processing Systems: A Survey [article]

Miyuru Dayarathna, Toyotaro Suzumura
2021 arXiv   pre-print
The development of scalable, representative, and widely adopted benchmarks for graph data systems have been a question for which answers has been sought for decades.  ...  graph stream processing as well as graph based machine learning, iii) benchmarks tend to use conventional metrics despite new meaningful metrics have been around for years, iv) increasing number of big  ...  UniBench is a generic benchmark for evaluation of state-of-the-art MMDBs.  ... 
arXiv:2005.12873v4 fatcat:jh3367b4vjaqbgyvaccjnxqjfi


Jingwen Leng, Tayler Hetherington, Ahmed ElTantawy, Syed Gilani, Nam Sung Kim, Tor M. Aamodt, Vijay Janapa Reddi
2013 Proceedings of the 40th Annual International Symposium on Computer Architecture - ISCA '13  
As such, GPU architects require robust tools that will enable them to quickly explore new ways to optimize GPGPUs for energy efficiency.  ...  We propose a new GPGPU power model that is configurable, capable of cycle-level calculations, and carefully validated against real hardware measurements.  ...  We thank Steve Keckler and John Edmondson for helpful discussions on the challenges of power modeling of GPUs.  ... 
doi:10.1145/2485922.2485964 dblp:conf/isca/LengHEGKAR13 fatcat:bkfi476bf5ed5lalls522mmd64

Sword: Scalable and Flexible Workload Generator for Distributed Data Processing Systems

Kay Anderson, Joseph Bigus, Eric Bouillet, Parijat Dube, Nagui Halim, Zhen Liu, Dimitrios Pendarakis
2006 Proceedings of the 2006 Winter Simulation Conference  
We present a Scalable WORkloaD generator (SWORD) that we have developed for the testing and benchmarking of high-volume data processing systems.  ...  Workload generation is commonly employed for performance characterization, testing and benchmarking of computer systems and networks.  ...  Correlation and Synchronization of Contents: There is an increased need to test and benchmark stream processing systems that process a large number of continuous streams containing potentially correlated  ... 
doi:10.1109/wsc.2006.323010 dblp:conf/wsc/AndersonBBDHLP06 fatcat:7vznbycconhybmkisf3wpcc3wa

I-BiDaaS - D1.3: Positioning of I-BiDaaS

Iliada Eleftheriou, Evangelia Kavakli, Thomas Lambert, Rizos Sakellariou
2018 Zenodo  
It is driven by the business objectives and user requirements of the specific use cases of the project industrial partners, as well as by the state-of-the-art in Big Data as-a-Self- Service research and  ...  Following systematic work, this has led to the specification of the requirements, their analysis and mapping to the architecture of the I-BiDaaS platform.  ...  Due to internal policies, it is hard to evaluate which technological solutions are the best to be used, so the objective is to develop, in a systematic and consistent way, a benchmark of graph databases  ... 
doi:10.5281/zenodo.4088296 fatcat:fjnmid5jpjhq3cpkbjnqi7knxu

Foundations of Composite Event Recognition (Dagstuhl Seminar 20071)

Alexander Artikis, Thomas Eiter, Alessandro Margara, Stijn Vansummeren
2020 Dagstuhl Reports  
Moreover, we lack a common understanding of the trade-offs between expressiveness and complexity, and a theory for comparing the fundamental capabilities of CER systems.  ...  CER is key in Big Data applications that require the processing of such event streams to obtain timely insights and to implement reactive and proactive measures.  ...  Another problem related to monitoring is the evaluation: changes in the input stream make it hard to test and systematically compare alternative solutions.  ... 
doi:10.4230/dagrep.10.2.19 dblp:journals/dagstuhl-reports/ArtikisEMV20 fatcat:35cbhc4l3be3pc7ccu2n54dbbq

Simultaneous consideration of remanufactured and new products in optimal product line design

R. Aydin, C. K. Kwong, P. Ji
2014 2014 IEEE International Conference on Industrial Engineering and Engineering Management  
Existing, implicit knowledge will be explicated and systematically bundled along the value stream from different departments of a company in order to enhance work system design and improvement.  ...  This article describes the further developments of "Value Stream Mapping" (Value Stream oriented Process Management, Value Stream Mapping and MTM, evaluation of alternative value streams, cost development  ...  By concentrating and coordinating these efforts, new requirements arise for improving value streams, processes and work systems.  ... 
doi:10.1109/ieem.2014.7058588 dblp:conf/ieem/AydinKJ14 fatcat:iva2nbm245glfh3pdc25dhvr3y

SemLinker: automating big data integration for casual users

Hassan Alrehamy, Coral Walker
2018 Journal of Big Data  
Availability of data and materials In a GitHub repository associated with our paper, we include all datasets used in our experiments.  ...  Ethics approval and consent to participate Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.  ...  Furthermore, S ij is regarded as a benchmark and is used to run schema checks on any new data instances ingested from i.  ... 
doi:10.1186/s40537-018-0123-x fatcat:orwq2j53k5hfdaufvysqbrayni

SLAMBench2: Multi-Objective Head-to-Head Benchmarking for Visual SLAM [article]

Bruno Bodin, Harry Wagstaff, Sajad Saeedi, Luigi Nardi, Emanuele Vespa, John H Mayer, Andy Nisbet, Mikel Luján, Steve Furber, Andrew J Davison, Paul H.J. Kelly, Michael O'Boyle
2018 arXiv   pre-print
SLAMBench2 is a benchmarking framework to evaluate existing and future SLAM systems, both open and close source, over an extensible list of datasets, while using a comparable and clearly specified list  ...  ElasticFusion, InfiniTAM, ORB-SLAM2, OKVIS, and integrating new ones is straightforward and clearly specified by the framework.  ...  ACKNOWLEDGEMENT This research is supported by Engineering and Physical Sciences Research Council (EPSRC), grant reference EP/K008730/1.  ... 
arXiv:1808.06820v1 fatcat:7yalfihhijeolikhimmwpq2ywq

Beauty and the Beast

Floriment Klinaku, Martina Rapp, Jörg Henss, Stephan Rhode
2022 Companion of the 2022 ACM/SPEC International Conference on Performance Engineering  
Moreover, to generate reliable CPU workloads we make use of ProtoCom, a well-known library for the generation of resource demands, and report the performance under various quality requirements in a Kubernetes  ...  In this paper, we present a work in progress towards a reference use case and application for data-intensive containerized cloud applications having an industrial motivation.  ...  For computational resources we acknowledge the bwCloud (, funded by the Ministry of Science, Research and Arts Baden-Württemberg (Ministerium für Wissenschaft, Forschung und Kunst  ... 
doi:10.1145/3491204.3527482 fatcat:lczsdijhafhtjm64kslmb3s5ra

MLPerf Tiny Benchmark [article]

Colby Banbury, Vijay Janapa Reddi, Peter Torelli, Jeremy Holleman, Nat Jeffries, Csaba Kiraly, Pietro Montino, David Kanter, Sebastian Ahmed, Danilo Pau, Urmish Thakker, Antonio Torrini (+10 others)
2021 arXiv   pre-print
However, continued progress is limited by the lack of a widely accepted and easily reproducible benchmark for these systems.  ...  Additionally, MLPerf Tiny implements a modular design that enables benchmark submitters to show the benefits of their product, regardless of where it falls on the ML deployment stack, in a fair and reproducible  ...  To enable more systematic development while fostering innovation, we need a fair, replicable, and robust method of evaluating TinyML systems.  ... 
arXiv:2106.07597v4 fatcat:ps4y36uq4nevxfbe7p3tne4opu

Exploring System Performance of Continual Learning for Mobile and Embedded Sensing Applications [article]

Young D. Kwon, Jagmohan Chauhan, Abhishek Kumar, Pan Hui, Cecilia Mascolo
2022 arXiv   pre-print
We also demonstrate for the first time that it is feasible and practical to run continual learning on-device with a limited memory budget.  ...  ) on six datasets from three mobile and embedded sensing applications in a range of scenarios having different learning complexities.  ...  ACKNOWLEDGMENTS This work is supported by a Google Faculty Award 2019 and by Nokia Bell Labs through their donation for the Centre of Mobile, Wearable Systems and Augmented Intelligence to the University  ... 
arXiv:2110.13290v2 fatcat:rtzneg33ijdw3hq4bhk6gtqk6y
« Previous Showing results 1 — 15 out of 2,956 results