Filters








9 Hits in 1.0 sec

ApproxHadoop: Bringing Approximations to MapReduce Frameworks

Inigo Goiri, Ricardo Bianchini, Nagarakatte Santosh, Thu Nguyen
2017
We conclude that our framework and system can make approximation easily accessible to many application domains using the MapReduce model.  ...  In this paper, we We propose and evaluate a framework for creating and running approximation-enabled MapReduce programs.  ...  Verma and his co-authors for the barrier-less extension to Hadoop [47] . We also thank David Carrera and the ASPLOS reviewers for comments that helped us improve this paper.  ... 
doi:10.7282/t3cj8j0p fatcat:f4yflbwl5bhpvhen3wbjebvnou

Uncertainty Propagation in Data Processing Systems

Ioannis Manousakis, Íñigo Goiri, Ricardo Bianchini, Sandro Rigo, Thu D. Nguyen
2018 Proceedings of the ACM Symposium on Cloud Computing - SoCC '18  
We implement this framework in a system called UP-MapReduce, and use it to modify ten applications, including AI/ML, image processing and trend analysis applications to process uncertain data.  ...  We present a framework for uncertainty propagation (UP) that allows developers to modify precise implementations of DAG nodes to process uncertain inputs with modest effort.  ...  We implement the proposed framework in UP-MapReduce, an extension of the Hadoop MapReduce, to handle uncertainty propagation (UP).  ... 
doi:10.1145/3267809.3267833 dblp:conf/cloud/ManousakisGBRN18 fatcat:ng32pkxpyfdqldatji4u6ku3yi

Approximation with Error Bounds in Spark [article]

Guangyan Hu, Desheng Zhang, Sandro Rigo, Thu D. Nguyen
2019 arXiv   pre-print
We introduce a sampling framework to support approximate computing with estimated error bounds in Spark.  ...  We have implemented a prototype of our framework called ApproxSpark, and used it to implement five approximate applications from different domains.  ...  ApproxHadoop [25] introduces approximation to the MapReduce [26] paradigm.  ... 
arXiv:1812.01823v3 fatcat:4jhtqgkrq5fodlzdpgzqkujooe

Online Adaptive Approximate Stream Processing with Customized Error Control

Xiaohui Wei, Yuanyuan Liu, Xingwang Wang, Shang Gao, Lei Chen
2019 IEEE Access  
To address these problems, we present an online adaptive approximate processing framework with a delicate combination of data learning, sampling, and quality control.  ...  In approximate processing on stream data, most works focus on how to approximate online arrival data. However, the efficiency of approximation needs to consider multiple aspects.  ...  Based on the MapReduce framework, Goiri et al. [19] designed a prototype system, ApproxHadoop, to implement approximation-enabled applications through input data sampling and task dropping.  ... 
doi:10.1109/access.2019.2899825 fatcat:6dcw5efe55ctjo2huhm5ehhfvy

Constant- Time Approximate Sliding Window Framework with Error Control

Alvaro Villalba, David Carrera
2019 2019 IEEE 22nd International Symposium on Real-Time Distributed Computing (ISORC)  
It is, to our knowledge, the first general purpose sliding window programable framework that combines constant-time aggregations with error bounded approximate computing techniques.  ...  In this context, Stream Processing frameworks need to combine efficient algorithms with low computational complexity to manage sliding windows, with the ability to adjust resource demands for different  ...  [20] proposed an approximate computing set of mechanisms for batch processing for Hadoop, called ApproxHadoop.  ... 
doi:10.1109/isorc.2019.00031 dblp:conf/isorc/VillalbaC19 fatcat:52lcxpoz6nh43lbrut4vgf2nb4

Input responsiveness: using canary inputs to dynamically steer approximation

Michael A. Laurenzano, Parker Hill, Mehrzad Samadi, Scott Mahlke, Jason Mars, Lingjia Tang
2016 SIGPLAN notices  
Based on these runtime tests, the approximation that best fits the desired accuracy constraints is selected and applied to the full input to produce an approximate result.  ...  to automatically control how program approximation is applied on an input-by-input basis.  ...  Acknowledgements We would like to thank our anonymous reviewers for their comments and suggestions. We also thank our shepherd, Cindy Rubio González, for her valuable feedback.  ... 
doi:10.1145/2980983.2908087 fatcat:irsndunbwrfxtbnmq673qnmqxy

Input responsiveness: using canary inputs to dynamically steer approximation

Michael A. Laurenzano, Parker Hill, Mehrzad Samadi, Scott Mahlke, Jason Mars, Lingjia Tang
2016 Proceedings of the 37th ACM SIGPLAN Conference on Programming Language Design and Implementation - PLDI 2016  
Based on these runtime tests, the approximation that best fits the desired accuracy constraints is selected and applied to the full input to produce an approximate result.  ...  to automatically control how program approximation is applied on an input-by-input basis.  ...  Acknowledgements We would like to thank our anonymous reviewers for their comments and suggestions. We also thank our shepherd, Cindy Rubio González, for her valuable feedback.  ... 
doi:10.1145/2908080.2908087 dblp:conf/pldi/LaurenzanoHSMMT16 fatcat:vnqayw7manffhd67ujannbtyyi

Approximating distributed graph algorithms [article]

Michael Schramm, Universität Stuttgart, Universität Stuttgart
2019
This leads to the question: do we always need to know the exact answer for a large graph?  ...  This has led to the development of many distributed graph processing systems. But these existing graph processing systems take several minutes or even hours to execute popular graph algorithms.  ...  GRASS adopted them for approximation analytics. ApproxHadoop [GBNN15] enables approximation in MapReduce Systems.  ... 
doi:10.18419/opus-10383 fatcat:fxnkryebsrcv3gwdr5silz7idq

Wavelet-based Algorithms for Approximate Processing in the Big Data Era [article]

Ioannis Mytilinis, National Technological University Of Athens
2021
functions of MapReduce programs.  ...  not be affected by the overhead of the three MapReduce jobs.  ...  MapReduce for L2-error Synopses Index of Algorithms BUDGreedyAbs Distributed heuristic algorithm, that given a budget constraint, approximates the L ∞ -optimal wavelet synopsis.  ... 
doi:10.26240/heal.ntua.21878 fatcat:xlllphljkjdfhhgx7m4itgauye