Filters








523 Hits in 4.1 sec

Energy-aware Self-Adaptation for Application Execution on Heterogeneous Parallel Architectures

Richard Kavanagh, Karim Djemame, Jorge Ejarque, Rosa Badia, David Garzia Perez
2019 IEEE Transactions on Sustainable Computing  
This therefore not only mitigates complexities associated with heterogeneous devices but aims to take advantage of the heterogeneity.  ...  Hardware in High Performance Computing environments in recent years have increasingly become more heterogeneous in order to improve computational performance.  ...  In this paper we address these issues with a energy-aware self-adaptive framework for heterogeneous parallel architectures.  ... 
doi:10.1109/tsusc.2019.2912000 fatcat:2fki22aspzeifbcyzwcfsa2uua

Accuracy Aware Power Management for Many-core Systems running Error Resilient Applications

Anil Kanduri, M. Hashem Haghbayan, Amir Rahmani, Pasi Liljeberg, Axel Jantsch, Hannu Tenhunen, Nikil Dutt
2017 Zenodo  
They suggest that combinatorial usage of different power knobs is effective for system level power capping decisions. Cochran et al.  ...  This system deals with heterogeneous data generated by different sensors and involves computations on unclassified, redundant and noisy data.  ...  APPEND+ improves performance and energy efficiency with the APPX knob and sensitivity aware actuation of the APPX knob, while power capping is maintained with combination of APPX, DVFS and PG knobs. .  ... 
doi:10.5281/zenodo.1488335 fatcat:rz3eih72xfgszn46fvxuqwv6rm

Don't race the memory bus: taming the GC leadfoot

Ahmed Hussein, Antony L. Hosking, Mathias Payer, Christopher A. Vick
2015 Proceedings of the 2015 ACM SIGPLAN International Symposium on Memory Management - ISMM 2015  
Running established multi-threaded benchmarks shows that total processor energy can be reduced up to 30 %, with end-to-end performance loss of at most 10 %.  ...  The GC-aware governor with 2.15 GHz cap has the best MMU curve on the DaCapo benchmarks lusearch and xalan (Fig. 7a) .  ...  Introduction Shipped as systems-on-a-chip (SoC), mobile platforms feature heterogeneous multi-core hardware with on-die hardware peripherals such as WiFi and GPS.  ... 
doi:10.1145/2754169.2754182 dblp:conf/iwmm/HusseinHPV15 fatcat:savrzyeheredfillb2qk7rvsgu

Don't race the memory bus: taming the GC leadfoot

Ahmed Hussein, Antony L. Hosking, Mathias Payer, Christopher A. Vick
2015 SIGPLAN notices  
Running established multi-threaded benchmarks shows that total processor energy can be reduced up to 30 %, with end-to-end performance loss of at most 10 %.  ...  The GC-aware governor with 2.15 GHz cap has the best MMU curve on the DaCapo benchmarks lusearch and xalan (Fig. 7a) .  ...  Introduction Shipped as systems-on-a-chip (SoC), mobile platforms feature heterogeneous multi-core hardware with on-die hardware peripherals such as WiFi and GPS.  ... 
doi:10.1145/2887746.2754182 fatcat:gs47lfe77zhv7jawaxk4asptry

NLSC: Unrestricted Natural Language-based Service Composition through Sentence Embeddings [article]

Oscar J. Romero, Ankit Dangi, Sushma A. Akoju
2019 arXiv   pre-print
With recent developments in Machine learning and Natural Language Processing, we motivate the use of Sentence Embeddings by leveraging richer semantic representations of sentences for service description  ...  Experimental results show that service composition development effort may be reduced by more than 44\% while keeping a high precision/recall when matching high-level user requests with low-level service  ...  , concepts and natural language sentences that account for word usage in context to user's sentence by applying a state-of-the-art pretrained semantic representation model of English language.  ... 
arXiv:1901.07910v2 fatcat:m5feqrvr3bfalabxi46ulgsq2u

Graph Neural Networks for Natural Language Processing: A Survey [article]

Lingfei Wu, Yu Chen, Kai Shen, Xiaojie Guo, Hanning Gao, Shucheng Li, Jian Pei, Bo Long
2021 arXiv   pre-print
Although text inputs are typically represented as a sequence of tokens, there isa rich variety of NLP problems that can be best expressed with a graph structure.  ...  Deep learning has become the dominant approach in coping with various tasks in Natural LanguageProcessing (NLP).  ...  Context-aware graph segmentation for graph-based translation.  ... 
arXiv:2106.06090v1 fatcat:zvkhinpcvzbmje4kjpwjs355qu

Toward Engineering a Secure Android Ecosystem

Meng Xu, Chenxiong Qian, Sangho Lee, Taesoo Kim, Chengyu Song, Yang Ji, Ming-Wei Shih, Kangjie Lu, Cong Zheng, Ruian Duan, Yeongjin Jang, Byoungyoung Lee
2016 ACM Computing Surveys  
With the increasing number and complexity of security problems and solutions, we believe this is the right moment to step back and systematically re-evaluate the Android security architecture and security  ...  Alternatively, a context-aware noise (e.g., a deliberate vibrating when user taps PIN) may disturb the performance of sensors during a user's interaction with sensitive information.  ...  For example, an app may send an unlimited number of text messages as long as it is granted the SEND_SMS permission. 3) There is no way to enforce runtime context-aware policies.  ... 
doi:10.1145/2963145 fatcat:d5vhxpdywrevvbh4as6vvt576q

Power-awareness and smart-resource management in embedded computing systems

M. D. Santambrogio, Jose L. Ayala, Simone Campanoni, R. Cattaneo, G. C. Durelli, M. Ferroni, A. Nacci, J. Pagan, M. Zapater, M. Vallejo
2015 2015 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS)  
This knowledge is then used to make accurate predictions of the Time-To-Live (TTL) of a generic mobile device in its real usage context.  ...  assignment technique, based on heterogeneity and application-awareness, that redistributes low-demand computational tasks from high-performance facilities to idle nodes with low and medium resources in  ... 
doi:10.1109/codesisss.2015.7331372 dblp:conf/codes/SantambrogioACC15 fatcat:k2pooqp6vbedngyces72jq6wga

Deliverable D3.7 Consolidated Projects Summary Report

Alasdair Reid, Matias Barberis Rami
2021 Zenodo  
CAP-A A Community-driven Approach to Privacy Awareness (CAP-A)'s objective was to deploy a socio-technical solution based on collective awareness and informed consent, whereby data collection and use by  ...  Secondarily, the project has led the team behind these apps to move the focus from the apps themselves to the very context of our different users and in particular to a general lack of awareness regarding  ...  Simply logging in with your fingerprint sounded like the ideal solution. But when the entire flow was discussed, there were too many downsides on the onboarding part.  ... 
doi:10.5281/zenodo.5767430 fatcat:w2lxzgxktretjcv7h4xotz7xaq

Leveraging Machine Learning and Big Data for Smart Buildings: A Comprehensive Survey [article]

Basheer Qolomany, Ala Al-Fuqaha, Ajay Gupta, Driss Benhaddou, Safaa Alwajidi, Junaid Qadir, Alvis C. Fong
2019 arXiv   pre-print
In this paper, we survey the area of smart building with a special focus on the role of techniques from machine learning and big data analytics.  ...  For example, readings from smart meters can be used to better predict and balancing the usage of power.  ...  The EasyLiving project was designed to provide context-aware computing services.  ... 
arXiv:1904.01460v2 fatcat:rz2gsyurkradlkqvfankw3zue4

The ANTAREX Domain Specific Language for High Performance Computing [article]

Cristina Silvano, Giovanni Agosta, Andrea Bartolini, Andrea R. Beccari, Luca Benini, Loïc Besnard, João Bispo, Radim Cmar, João M. P. Cardoso, Carlo Cavazzoni, Daniele Cesarini, Stefano Cherubin (+10 others)
2019 arXiv   pre-print
In this paper, we present an overview of the key outcome of the project, the ANTAREX DSL, and some of its capabilities through a number of examples, including how the DSL is applied in the context of the  ...  However, these are only viable if the power capping logic is aware of the application requirements.  ...  It does not support constraints on the navigation quality, providing the maximum quality based on its prediction of data usage.  ... 
arXiv:1901.06175v1 fatcat:qypv32vhsffbzlpnbppijeeah4

Genetic Programming for Smart Phone Personalisation [article]

Philip Valencia, Aiden Haak, Alban Cotillon, Raja Jurdak
2014 arXiv   pre-print
Personalisation in smart phones requires adaptability to dynamic context based on user mobility, application usage and sensor inputs.  ...  This paper proposes genetic programming (GP), which can evolve program logic in realtime, as an online learning method to deal with the highly dynamic context in smart phone personalisation.  ...  Moreover, such content might quickly deplete the user monthly capped data plan.Available Google Reader apps on the applications market reports news from all the feeds to which the user has subscribed.  ... 
arXiv:1408.2288v1 fatcat:3ibc24gpjjepdeldeharlk7shm

A Multi-Armed Bandit-based Approach to Mobile Network Provider Selection [article]

Thomas Sandholm, Sayandev Mukherjee
2021 arXiv   pre-print
We propose, prototype, and evaluate a spectrum market for mobile network access, where multiple network operators offer blocks of bandwidth at specified prices for short-term leases to users, with autonomous  ...  Notably, the History benchmark does poorly in comparison due to not taking the App demand into account, i.e. not being aware of which app is causing what QoE.  ...  eSIM Embedded Subscriber Identity Module (eSIM) or Embedded Universal Integrated Circuit Card (eUICC) is a programmable chip embedded in a device that allows it to connect to different mobile networks  ... 
arXiv:2012.04755v2 fatcat:pmoz6lrp3naabnkc5k73j52vx4

Balancing energy, latency and accuracy for mobile sensor data classification

David Chu, Nicholas D. Lane, Ted Tsung-Te Lai, Cong Pang, Xiangying Meng, Qing Guo, Fan Li, Feng Zhao
2011 Proceedings of the 9th ACM Conference on Embedded Networked Sensor Systems - SenSys '11  
From the experience of using Kobe to prototype two new applications, we observe that Kobe enables easier development of mobile sensing and classification apps.  ...  With the help of a SQL-like programming interface, Kobe performs profiling and optimization of classifiers to achieve an optimal energy-latency-accuracy tradeoff.  ...  Adapting to Changes We next evaluate Kobe's and existing solutions' abilities to adapt to device heterogeneity, networking and cloud availability, multi-programming and user usage behavior with Adaptation  ... 
doi:10.1145/2070942.2070949 dblp:conf/sensys/ChuLLPMGLZ11 fatcat:6tkg2es4yrcnhfufmffmbrcxl4

Ray: A Distributed Framework for Emerging AI Applications [article]

Philipp Moritz, Robert Nishihara, Stephanie Wang, Alexey Tumanov, Richard Liaw, Eric Liang, Melih Elibol, Zongheng Yang, William Paul, Michael I. Jordan, Ion Stoica
2018 arXiv   pre-print
The next generation of AI applications will continuously interact with the environment and learn from these interactions.  ...  in other contexts.  ...  simulation and embedded serving.  ... 
arXiv:1712.05889v2 fatcat:odyn47sxe5fstcknououtih4sm
« Previous Showing results 1 — 15 out of 523 results