Filters








14 Hits in 3.7 sec

Providing Source Code Level Portability Between CPU and GPU with MapCG

Chun-Tao Hong, De-Hao Chen, Yu-Bei Chen, Wen-Guang Chen, Wei-Min Zheng, Hai-Bo Lin
2012 Journal of Computer Science and Technology  
In this paper, we propose MapCG, a MapReduce framework to provide source code level portability between CPUs and GPUs.  ...  This results in high development and maintenance costs. We believe it is desirable to have a programming model which provides source code portability between CPUs and GPUs, as well as different GPUs.  ...  In this paper, we propose MapCG, a framework which offers source code level portability between CPU and GPU.  ... 
doi:10.1007/s11390-012-1205-4 fatcat:37y5ilv4kva5jcodoyzsusdnre

MapCG

Chuntao Hong, Dehao Chen, Wenguang Chen, Weimin Zheng, Haibo Lin
2010 Proceedings of the 19th international conference on Parallel architectures and compilation techniques - PACT '10  
In this paper, we propose MapCG, a MapReduce framework to provide source code level portability between CPU and GPU.  ...  We believe it is desired to have a programming model which provides source code portability between CPUs and GPUs, and different GPUs: Programmers only need to write one version of code and can be compiled  ...  CONCLUSION In this paper, we present the MapCG, a high level framework that offers source code portability between CPU and GPU.  ... 
doi:10.1145/1854273.1854303 dblp:conf/IEEEpact/HongCCZL10 fatcat:muwp3cri2jex7dvqul6gqe34je

Pattern matching of signature-based IDS using Myers algorithm under MapReduce framework

Monther Aldwairi, Ansam M. Abu-Dalo, Moath Jarrah
2017 EURASIP Journal on Information Security  
The results show 1.3 and 1.7 times improvement for Phoenix++ and MAPCG MapReduce implementations over MPI respectively.  ...  Additionally, we use two implementations of MapReduce to parallelize the Myers algorithm using Phoenix++ and MAPCG.  ...  Acknowledgements We would like to acknowledge the efforts of Yaser Jararweh and the valuable feedback of the reviewers. Availability of data and material Not applicable.  ... 
doi:10.1186/s13635-017-0062-7 fatcat:nmpetehk2jgyvdmkwcshhy4b2u

Lit: A high performance massive data computing framework based on CPU/GPU cluster

Yanlong Zhai, Emmanuel Mbarushimana, Wei Li, Jing Zhang, Ying Guo
2013 2013 IEEE International Conference on Cluster Computing (CLUSTER)  
Since the architecture and programming model of GPU is different from CPU, Lit provided an annotation based approach to automatically generate CUDA codes from Hadoop codes.  ...  Lit hided the complexity of programming on CPU/GPU cluster by providing extended compiler and opti mizer.  ...  This work is supported by National 863 Programme(No 20 13AAOI A212 ) "Kernel Software and Sys tem for Intelligent Cloud Service and Management Platform" and the Beijing Institute of Technology Young Outstanding  ... 
doi:10.1109/cluster.2013.6702614 dblp:conf/cluster/ZhaiMLZG13 fatcat:mxcqtu762jdzlaevisuueloswq

Optimized Realization of Software Components with Flexible OpenCL Functionality

Gabriel Campeanu, Jan Carlson, Séverine Sentilles
2018 Proceedings of the 13th International Conference on Evaluation of Novel Approaches to Software Engineering  
Today, the newly available embedded boards with GPUs provide a solution to satisfy the ever-increasing requirements of modern embedded systems.  ...  However, this paradigm lacks GPU support to address the specifics of these new boards. This leads to components that typically have reduced reusability, poor maintainability and portability.  ...  When developing applications for heterogeneous systems, it is desirable to have code portability between CPU and GPU.  ... 
doi:10.5220/0006691500770088 dblp:conf/enase/CampeanuCS18 fatcat:brdmf6zjevbejeb7vx6coyskui

Scaling MapReduce Vertically and Horizontally

Ismail El-Helw, Rutger Hofman, Henri E. Bal
2014 SC14: International Conference for High Performance Computing, Networking, Storage and Analysis  
We experimentally evaluated the performance of five MapReduce applications and show that Glasswing outperforms Hadoop on a 64-node multi-core CPU cluster by factors between 1.2 and 4, and factors from  ...  20 to 30 on a 23-node GPU cluster.  ...  ACKNOWLEDGMENT The authors would like to thank the anonymous reviewers, Thilo Kielmann for his critical feedback and Kees Verstoep for excellent administration of the VU/DAS4 cluster.  ... 
doi:10.1109/sc.2014.48 dblp:conf/sc/El-HelwHB14 fatcat:uujilehmdrcflhwitwqxn2byfy

Adaptive input-aware compilation for graphics engines

Mehrzad Samadi, Amir Hormati, Mojtaba Mehrara, Janghaeng Lee, Scott Mahlke
2012 Proceedings of the 33rd ACM SIGPLAN conference on Programming Language Design and Implementation - PLDI '12  
Using this system, programmers develop their applications in a high-level streaming language and let Adaptic undertake the difficult task of input portable optimizations and code generation.  ...  The results show that Adaptic is capable of generating codes that can perform on par with their hand-optimized counterparts over certain input ranges and outperform them when the input falls out of the  ...  Acknowledgement Much gratitude goes to the anonymous referees who provided excellent feedback on this work.  ... 
doi:10.1145/2254064.2254067 dblp:conf/pldi/SamadiHMLM12 fatcat:eyebz4yvtra7znmz3zeinlkg2m

Adaptive input-aware compilation for graphics engines

Mehrzad Samadi, Amir Hormati, Mojtaba Mehrara, Janghaeng Lee, Scott Mahlke
2012 SIGPLAN notices  
Using this system, programmers develop their applications in a high-level streaming language and let Adaptic undertake the difficult task of input portable optimizations and code generation.  ...  The results show that Adaptic is capable of generating codes that can perform on par with their hand-optimized counterparts over certain input ranges and outperform them when the input falls out of the  ...  Acknowledgement Much gratitude goes to the anonymous referees who provided excellent feedback on this work.  ... 
doi:10.1145/2345156.2254067 fatcat:dihubqsczvcs7dhqfxdqgi7i5y

COX: CUDA on X86 by Exposing Warp-Level Functions to CPUs [article]

Ruobing Han, Jaewon Lee, Jaewoong Sim, Hyesoon Kim
2021 arXiv   pre-print
The examples are DPC, Hipfy, where CUDA source code have to be translated to their native supporting language and then they are supported.  ...  COX can support the most recent CUDA features, and the application coverage is much higher (90 also show that the warp-level functions in CUDA can be efficiently executed by utilizing CPU SIMD (AVX) instructions  ...  Most modifications are in the runtime level, and users can directly use the original GPU source code. As reported in [17] , the AMD CPU OpenCL implementation is based on this technology.  ... 
arXiv:2112.10034v1 fatcat:l7pgnpsjgzfrviptgblu3f5r5u

Multi-GPU MapReduce on GPU Clusters

Jeff A. Stuart, John D. Owens
2011 2011 IEEE International Parallel & Distributed Processing Symposium  
We use persistent map and reduce tasks and stress aspects of GPMR with a set of standard MapReduce benchmarks.  ...  We conclude with an exposition on the types of MapReduce tasks well-suited to GPMR, and why some tasks need more modifications than others to work well with GPMR.  ...  , and to NCSA and Wen-Mei Hwu for allowing us access to their GPU cluster.  ... 
doi:10.1109/ipdps.2011.102 dblp:conf/ipps/StuartO11 fatcat:4wsvdwkncjfojifgmh6bbbvoeu

An automatic input-sensitive approach for heterogeneous task partitioning

Klaus Kofler, Ivan Grasso, Biagio Cosenza, Thomas Fahringer
2013 Proceedings of the 27th international ACM conference on International conference on supercomputing - ICS '13  
Our approach has been evaluated over a suite of 23 programs and respectively achieves a performance improvement of 22% and 25% compared to an execution of the benchmarks on a single CPU and a single GPU  ...  Unleashing the full potential of heterogeneous systems, consisting of multi-core CPUs and GPUs, is a challenging task due to the difference in processing capabilities, memory availability, and communication  ...  [19] proposed MapCG, a framework that supports source code level portability between CPU and GPU.  ... 
doi:10.1145/2464996.2465007 dblp:conf/ics/KoflerGCF13 fatcat:bgolcfusbjdc7m4xft7mnbg2ni

Programming and Runtime Support to Blaze FPGA Accelerator Deployment at Datacenter Scale

Muhuan Huang, Di Wu, Cody Hao Yu, Zhenman Fang, Matteo Interlandi, Tyson Condie, Jason Cong
2016 Proceedings of the Seventh ACM Symposium on Cloud Computing - SoCC '16  
Unlike conventional CPU and GPU targeted programs, compiling an FPGA program can take several hours, which makes existing runtime systems that use dynamic code generation for CPU-GPU datacenters, such  ...  With the end of CPU core scaling due to dark silicon limitations, customized accelerators on FPGAs have gained increased attention in modern datacenters due to their lower power, high performance and energy  ...  Mentor Graphics; C-FAR, one of the six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA; grants NSF IIS-1302698 and CNS-1351047; and U54EB020404 awarded by  ... 
doi:10.1145/2987550.2987569 pmid:28317049 pmcid:PMC5351886 dblp:conf/cloud/HuangWYFICC16 fatcat:5f6bnm6xxbfk3k5fv3sgqarftu

Input-aware auto-tuning for directive-based GPU programming

Alberto Magni, Dominik Grewe, Nick Johnson
2013 Proceedings of the 6th Workshop on General Purpose Processor Using Graphics Processing Units - GPGPU-6  
The difficulties posed by GPGPU programming and the need to increase productivity have guided research towards directive-based high-level programs for accelerators.  ...  It significantly simplifies writing code for graphics engines leaving the programmer the opportunity to tune the application for the target hardware and input.  ...  We would then be able to specialize the training data to the needs of the programmer since we would work with the input sizes supplied by the user.  ... 
doi:10.1145/2458523.2458530 dblp:conf/asplos/MagniGJ13 fatcat:24mdgivunbeohga6wpfreqiskm

A secure data privacy preservation for on-demand cloud service

Dhasarathan Chandramohan, Thirumal Vengattaraman, Ponnurangam Dhavachelvan
2017 Journal of King Saud University: Engineering Sciences  
optimization of the service environment etc, cloud users data and their identity, reliability, maintainability and privacy may vary for different CPs (cloud providers).  ...  More remarkable occurrence is even the cloud provider does not have suggestions regarding the information and the digital data stored and maintained globally anywhere in the cloud.  ...  Hong et al. (2012) , propose a new MapCG model as a map-reducing framework to provide source code level portability between CPUs (central processing units) and GPUs (graphics processing units).  ... 
doi:10.1016/j.jksues.2015.12.002 fatcat:pttup3bi6ngprfdncwiujcbymm