Results 31 to 40 of about 1,185 (184)

High‐Performance GPU Techniques for 2D Convolution Operators in Deep Neural Networks

Concurrency and Computation: Practice and Experience, Volume 38, Issue 14, July 2026.
ABSTRACT This paper presents an optimized implementation of Winograd non‐fused convolution. Our optimizations include both application‐independent and Winograd‐specific software techniques, such as a specialized interface‐kernel data format (tile‐united CNHW layout) to enhance memory access efficiency; warp specialization and double‐buffered ...
Hui Wei, Enjie Liu, Youbing Zhao, Baiyu Liu, Dayou Li, Jinquan Xiong, Chuan Zhang, Yongqiang Wei, Zhaoxing Xu   +8 more
wiley   +1 more source

Research and Development of an Algorithm for the Response Time Estimation in Multiprocessor Systems Under the Interval Uncertainty of the Tasks Execution Times

Моделирование и анализ информационных систем, 2020
The paper presents an algorithm for the worst case response time (WCRT) estimation for multiprocessor systems with fixed-priority preemptive schedulers and the interval uncertainty of tasks execution times.
Mark G. Gonopolskiy, Alevtina B. Glonina
doaj   +1 more source

Per Processor Spin-Based Protocols for Multiprocessor Real-Time Systems [PDF]

Leibniz Transactions on Embedded Systems, 2017
This paper investigates preemptive spin-based global resource sharing protocols for resource-constrained real-time embedded multi-core systems based on partitioned fixed-priority preemptive scheduling.
Afshar, Sara, Behnam, Moris, Bril, Reinder J., Nolte, Thomas   +3 more
doaj   +1 more source

Treeline Provides a Unified Strategy for Optimising Phylogenetic Trees Under Alternative Criteria

Molecular Ecology Resources, Volume 26, Issue 5, July 2026.
ABSTRACT The choice of optimality criterion is a key consideration in phylogenetic studies. Recent work challenges the notion that more computationally demanding optimisation objectives result in better phylogenetic trees. This finding underscores the importance of comparing trees across optimisation objectives in addition to different models of ...
Erik S. Wright
wiley   +1 more source

Developing a Platform Using Petri Nets and GPenSIM for Simulation of Multiprocessor Scheduling Algorithms

Applied Sciences
Efficient multiprocessor scheduling is pivotal in optimizing the performance of parallel computing systems. This paper leverages the power of Petri nets and the tool GPenSIM to model and simulate a variety of multiprocessor scheduling algorithms (the ...
Daniel Osmundsen Dirdal, Danny Vo, Yuming Feng, Reggie Davidrajuh   +3 more
doaj   +1 more source

Scheduling chained multiprocessor tasks onto large multiprocessor system [PDF]

Computing, 2017
In this paper, we proposed an effective approach for scheduling of multiprocessor unit time tasks with chain precedence on to large multiprocessor system. The proposed longest chain maximum processor scheduling algorithm is proved to be optimal for uniform chains and monotone (non-increasing/non-decreasing) chains for both splitable and non-splitable ...
Tarun K. Agrawal, Aryabartta Sahu, Manojit Ghose, R. Sharma   +3 more
openaire   +3 more sources

Rethinking Per‐Thread Computation for Machine Learning Design Exploration: A Work‐Efficient GPU Strategy for K‐Means and XGBoost

Concurrency and Computation: Practice and Experience, Volume 38, Issue 11, June 2026.
ABSTRACT This paper presents novel GPU implementation strategies that effectively exploit the available parallelism, based on the “more work per thread” approach, for three machine learning design exploration tasks: Multiple K‐means evaluation, dimensionality reduction through parallel K‐means encoding for XGBoost trees, and XGBoost tree pruning ...
Olavo Barros, Isabela Freitas, Pedro H. Pereira, Walter de B. Neto, Adrián Alcolea, María Villarroya‐Gaudó, Javier Resano, José Augusto M. Nacif, Ricardo Ferreira   +8 more
wiley   +1 more source

Don't be Tricked by Iterative Masking

Computational Intelligence, Volume 42, Issue 3, June 2026.
ABSTRACT A common approach to quantifying neural text classifier interpretability is to calculate faithfulness metrics based on iteratively masking salient input tokens and measuring changes in the model prediction. We propose that this property is better described as “sensitivity to iterative masking,” and highlight pitfalls in using this measure for ...
Evan Crothers, Herna Viktor, Nathalie Japkowicz   +2 more
wiley   +1 more source

Optimising Multiprocessor Image-Based Control Through Pipelining and Parallelism

IEEE Access, 2021
Image-based control (IBC) systems have a long sensing delay due to compute-intensive image processing. Modern multiprocessor IBC implementations consider either parallelisation of the sensing task or pipelining of the control loop to cope with this long ...
Sajid Mohamed, Dip Goswami, Sayandip De, Twan Basten   +3 more
doaj   +1 more source

Effect of Peanut Shell Powder as a Pore‐Forming Additive in the Production of Ceramic Membranes

International Journal of Ceramic Engineering &Science, Volume 8, Issue 3, May 2026.
This study investigates the sustainable use of an agricultural by‐product, Peanut shell powder (PSP), to create pores in clay‐alumina membranes sintered at 1150°C. With 30 wt.% PSP, porosity reaches 49%, achieving high water flux (∼2300 L·h−1·m−2) and usable strength (17 MPa). Ideal for low‐cost membrane and effluent treatment.
Mykaell Y. M. Souza, Lizandra K. S. Lima, Hélio L. Lira, Miguel A. Rodríguez, Lisiane N. L. Santana   +4 more
wiley   +1 more source