Filters








302 Hits in 3.4 sec

Nitro: A Framework for Adaptive Code Variant Tuning

Saurav Muralidharan, Manu Shantharam, Mary Hall, Michael Garland, Bryan Catanzaro
2014 2014 IEEE 28th International Parallel and Distributed Processing Symposium  
This paper describes Nitro, a programmer-directed autotuning framework that facilitates tuning of code variants, or alternative implementations of the same computation.  ...  Nitro provides a library interface that permits programmers to express code variants along with metainformation that aids the system in selecting among the set of variants at run time.  ...  ACKNOWLEDGEMENTS We would like to thank NVIDIA Corporation for generous equipment donations, and members of the NVIDIA research group including Duane Merrill and Sean Baxter for discussions on tuning the  ... 
doi:10.1109/ipdps.2014.59 dblp:conf/ipps/MuralidharanSHGC14 fatcat:km5kdegmkrfg3cofmudmm23bkm

Architecture-Adaptive Code Variant Tuning

Saurav Muralidharan, Amit Roy, Mary Hall, Michael Garland, Piyush Rai
2016 Proceedings of the Twenty-First International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS '16  
In this paper, we define a new approach called architectureadaptive code variant tuning, where the variant selection model is learned on a set of source architectures, and then used to predict variants  ...  Code variants represent alternative implementations of a computation, and are common in high-performance libraries and applications to facilitate selecting the most appropriate implementation for a specific  ...  Acknowledgments We would like to thank NVIDIA Corporation for generous equipment donations, and members of the NVIDIA research group including Duane Merrill and Albert Sidelnik for valuable discussions  ... 
doi:10.1145/2872362.2872411 dblp:conf/asplos/MuralidharanRHG16 fatcat:y3hhmp53avfivnuekajgwx2cwq

Architecture-Adaptive Code Variant Tuning

Saurav Muralidharan, Amit Roy, Mary Hall, Michael Garland, Piyush Rai
2016 SIGPLAN notices  
In this paper, we define a new approach called architectureadaptive code variant tuning, where the variant selection model is learned on a set of source architectures, and then used to predict variants  ...  Code variants represent alternative implementations of a computation, and are common in high-performance libraries and applications to facilitate selecting the most appropriate implementation for a specific  ...  Acknowledgments We would like to thank NVIDIA Corporation for generous equipment donations, and members of the NVIDIA research group including Duane Merrill and Albert Sidelnik for valuable discussions  ... 
doi:10.1145/2954679.2872411 fatcat:uu72lp6lkjdxxbcv4eh4tybq6y

Architecture-Adaptive Code Variant Tuning

Saurav Muralidharan, Amit Roy, Mary Hall, Michael Garland, Piyush Rai
2016 ACM SIGOPS Operating Systems Review  
In this paper, we define a new approach called architectureadaptive code variant tuning, where the variant selection model is learned on a set of source architectures, and then used to predict variants  ...  Code variants represent alternative implementations of a computation, and are common in high-performance libraries and applications to facilitate selecting the most appropriate implementation for a specific  ...  Acknowledgments We would like to thank NVIDIA Corporation for generous equipment donations, and members of the NVIDIA research group including Duane Merrill and Albert Sidelnik for valuable discussions  ... 
doi:10.1145/2954680.2872411 fatcat:ly4zo74uafgynpqpknfx32hjqe

Architecture-Adaptive Code Variant Tuning

Saurav Muralidharan, Amit Roy, Mary Hall, Michael Garland, Piyush Rai
2016 SIGARCH Computer Architecture News  
In this paper, we define a new approach called architectureadaptive code variant tuning, where the variant selection model is learned on a set of source architectures, and then used to predict variants  ...  Code variants represent alternative implementations of a computation, and are common in high-performance libraries and applications to facilitate selecting the most appropriate implementation for a specific  ...  Acknowledgments We would like to thank NVIDIA Corporation for generous equipment donations, and members of the NVIDIA research group including Duane Merrill and Albert Sidelnik for valuable discussions  ... 
doi:10.1145/2980024.2872411 fatcat:greecv2bcfemjcdr5ludajncvi

Reflection as the Basis for Developing a Dynamic SoC Persistence System

Benjamin Lopez, Francisco Ortin, Javier Noval
2004 Journal of Object Technology  
Although this is a profitable principle, existing tools lack two main features: runtime adaptability and language independence.  ...  This paper shows how computational reflection can be employed as a suitable technique to overcome the two previous limitations, offering dynamic adaptation of persistence features in a language independent  ...  We separate the interpreter from the language specification and the persistence framework has been designed without any dependency of a specific language. • Parameter tuning.  ... 
doi:10.5381/jot.2004.3.8.a2 fatcat:7bd76wkp6jg4zis3zx7pwn7y4i

A performance- and energy-oriented extended tuning process for time-step-based scientific applications

Natalia Kalinnik, Robert Kiesel, Thomas Rauber, Marcel Richter, Gudula Rünger
2020 Journal of Supercomputing  
Scientific application codes are often long-running time-and energy-consuming parallel codes, and the tuning of these methods towards the characteristics of a specific hardware is essential for a good  ...  The investigation comprises a broad range of tuning possibilities, starting from the choice of algorithms, the parallel programming model, static implementation variants, input characteristics as well  ...  Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long  ... 
doi:10.1007/s11227-020-03402-y fatcat:pk25kbpyzbdkpkkbex2mb5qh6m

A performance- and energy-oriented extended tuning process for time-step-based scientific applications

Natalia Kalinnik, Robert Kiesel, Thomas Rauber, Marcel Richter, Gudula Rünger
2021 Journal of Supercomputing  
Scientific application codes are often long-running time- and energy-consuming parallel codes, and the tuning of these methods towards the characteristics of a specific hardware is essential for a good  ...  The investigation comprises a broad range of tuning possibilities, starting from the choice of algorithms, the parallel programming model, static implementation variants, input characteristics as well  ...  Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long  ... 
doi:10.15495/epub_ubt_00005336 fatcat:75d4nwvfbrhbnpbibpkpil5rmi

Writing a performance-portable matrix multiplication

Jorge F. Fabeiro, Diego Andrade, Basilio B. Fraguela
2016 Parallel Computing  
As a consequence, programmers have to hand-tune the kernel codes for each device.  ...  There are several frameworks that, while providing functional portability of code across different platforms, do not automatically provide performance portability.  ...  The authors are also members of the CAPAP-H5 network, in whose framework the paper has been developed.  ... 
doi:10.1016/j.parco.2015.12.005 fatcat:yrldkoxhojahzmyvejd7eq7eay

Abstractions and strategies for adaptive programming

Saurav Muralidharan
2018
T h e U n i v e r s i t y o f U t a h G r a d u a t e S c h o o l STATEMENT OF DISSERTATION APPROVAL The dissertation of Saurav Muralidharan has been approved by the following supervisory committee members  ...  A framework for code variant tuning: We describe Nitro, a programmer-directed code variant tuning framework targeted at expert users.  ...  Nitro, a framework for code variant tuning targeted at expert programmers [57] .  ... 
doi:10.26053/0h-tzat-f800 fatcat:7bi3byv3kncxznma4lwdh46wve

SUGAR: Subgraph Neural Network with Reinforcement Pooling and Self-Supervised Mutual Information Mechanism [article]

Qingyun Sun, Jianxin Li, Hao Peng, Jia Wu, Yuanxing Ning, Phillip S. Yu, Lifang He
2021 arXiv   pre-print
To adaptively select striking subgraphs without prior knowledge, we develop a reinforcement pooling mechanism, which improves the generalization ability of the model.  ...  This paper presents a novel hierarchical subgraph-level selection and embedding based graph neural network for graph classification, namely SUGAR, to learn more discriminative subgraph representations  ...  Thanks for computing infrastructure provided by Huawei MindSpore platform.  ... 
arXiv:2101.08170v3 fatcat:62nxzvfqfnb75ibuoo2b432x64

Beyond the outer limits of nature by directed evolution

Patricia Molina-Espeja, Javier Viña-Gonzalez, Bernardo J. Gomez-Fernandez, Javier Martin-Diaz, Eva Garcia-Ruiz, Miguel Alcalde
2016 Biotechnology Advances  
For more than thirty years, biotechnology has borne witness to the power of directed evolution in designing molecules of industrial relevance.  ...  In this review we aim to present the most recent advances in this fascinating research field that are allowing us to surpass the limits of nature and apply newly gained attributes to a range of applications  ...  Two variants, EC#1 and EC#2, arose from this engineering approach with adapted properties to produce nanotubes for high-power lithium-ion batteries.  ... 
doi:10.1016/j.biotechadv.2016.03.008 pmid:27064127 fatcat:ie4qytenpnbczfg3bl2wfbpxqi

Synthetic biology approaches to biological containment: pre-emptively tackling potential risks

Leticia Torres, Antje Krüger, Eszter Csibra, Edoardo Gianni, Vitor B. Pinheiro
2016 Essays in Biochemistry  
Here, we describe recent advances in synthetic biology that contribute to the ongoing efforts to develop new and improved genetic, semantic, metabolic and mechanistic plans for the containment of GEMs.  ...  Physical containment barriers have proven effective but a number of strategies have been developed to further strengthen biocontainment.  ...  a sequence with the six-letter code (dA, dT, dC, dG, dNaM and d5SICS) to work as a template for replication.  ... 
doi:10.1042/ebc20160013 pmid:27903826 pmcid:PMC5264511 fatcat:caggfwj6dzb7dhox4xpoklioye

Autotuning algorithmic choice for input sensitivity

Yufei Ding, Jason Ansel, Kalyan Veeramachaneni, Xipeng Shen, Una-May O'Reilly, Saman Amarasinghe
2015 SIGPLAN notices  
Experimental results show that the new solution yields up to a 3x speedup over using a single configuration for all inputs, and a 34x speedup over a traditional one-level method for addressing input sensitivity  ...  This paper presents a novel two-level input learning algorithm to tackle the challenge for an important class of autotuning problems, algorithmic autotuning.  ...  Muralidharan and others have recently built a programmer-directed autotuning framework named Nitro for selecting the appropriate code variants at runtime [34] .  ... 
doi:10.1145/2813885.2737969 fatcat:xvqn3qnqnfeihisdfjiqyxjfsu

Biochemical and Structural Studies of the Mycobacterium tuberculosis O6-Methylguanine Methyltransferase and Mutated Variants

R. Miggiano, V. Casazza, S. Garavaglia, M. Ciaramella, G. Perugino, M. Rizzi, F. Rossi
2013 Journal of Bacteriology  
In particular, the suicidal enzyme O 6 -methylguanine-DNA methyltransferase (OGT) is responsible for the direct repair of O 6 -alkylguanine in double-stranded DNA and is therefore supposed to play a central  ...  the reduced affinity for the natural substrate displayed by mutated MtOGT.  ...  We acknowledge the European Synchrotron Radiation Facility (Grenoble, France) for provision of synchrotron radiation at beam lines ID14 and ID23.  ... 
doi:10.1128/jb.02298-12 pmid:23564173 pmcid:PMC3697256 fatcat:7ylurh7jdzhnhkthmjlvz74ige
« Previous Showing results 1 — 15 out of 302 results