Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 19 of 19 for “"Unified Memory"”.
-
Efficient Computing for Three-Dimensional Quantitative Phase Imaging
… research, OpenMP Tasking and CUDA Streaming with Unified Memory (TSUM) is proposed to speed up the tomographic angle computations in 3D TDPM. TSUM leverages CPU multithreading and GPU computing on a System on a Chip (SoC) with unified memory. Unified memory eliminates data transfer between CPU and …
-
Parallelizing unit test execution on GPU
… test case inputs may increase the number of memory transactions.</p> <p>To address these challenges, three studies were conducted. The first study investigated three programming models: CUDA Unified Memory, CUDA Non-Unified Memory, and OpenMP GPU Offloading to parallelize the test execution. …
-
Holistic Performance Analysis and Optimization of Unified Virtual Memory
… codes has been greatly reduced by the advent of Unified Memory technologies that abstract the management of physical memory away from the developer. However, these systems incur substantial overhead that paradoxically grows for codes where these technologies are most useful. While these …
-
Fine-grain acceleration of graph algorithms on a heterogeneous chip
… CPU cores reside in the same chip and share a unified memory, it is possible to have better execution schemes for many graph algorithms. Graph algorithms can exhibit producer-consumer behavior, a varying amount of parallelism during execution, and irregularity which results in inefficiency. The …
-
In the Shadow of the Bomb: Nagasaki's Place in Atomic Memory
This thesis examines the unique memory of Nagasaki formed after the atomic attack by the United States. Nagasaki’s Peace and Atomic Bomb Museum and the city’s narrative of the attacks are overshadowed by the narrative presented in Hiroshima and the Hiroshima Peace Memorial. While Hiroshima has …
-
Neužgimusi "terra irredenta": Karaliaučiaus krašto atmintis Lietuvos visuomenėje /
This study seeks to analyse the memory of the Karaliaučius (Kaliningrad) region in Lithuanian society with the theory of mnemonic actors and memory regimes by Michael Bernhard and Jan Kubik. In spite of beneficial conditions, discourse towards Karaliaučius has not evolved towards irredentism, which …
-
A Comprehensive Study and Performance Optimization of Shared Virtual Memory
… barrier in fully utilizing GPU acceleration. Unified Memory (UM) is a technology designed to lower the barrier of entry to GPU programming by merging all memory domains of a system. While UM provides easier access to the capabilities of heterogeneous systems, the performance cost of UM greatly …
-
Evaluating Design Choices for GPU-Accelerated Finite-Difference Time-Domain Simulation on Resource-Constrained Devices
… categories: thread allocation strategy, memory management approach, and numerical precision format. The results show that the choice of precision format is the most important factor: Floating point 32 (FP32) is nearly as accurate as floating point 64 (FP64) while taking about half the …
-
Visual Analysis Algorithms for Embedded Systems
… classifiers demand huge computational power and memory storage. That is an issue on embedded devices due to limited onboard resources. The computational demand of neural networks mainly stems from the convolutional layers. A significant improvement in performance can be obtained by reducing the …
-
MURAC: A unified machine model for heterogeneous computers
… key concepts that make up the MURAC model are a unified machine model, a unified instruction stream and a unified memory space. A simple programming model built upon these abstractions provides a consistent interface for interacting with the underlying machine to the user application. This …
-
Robust structured multigrid at extreme scales
… on each target processing unit. OpenMP with unified memory is then used to offload solve phase operations in the open-source, structured variational multigrid solver Cedar. The performance expectation is then used to provide context for performance gains by targeting GPUs on Sierra---a …
-
Techniques to mitigate performance impact of off-chip data migrations in modern GPU computing
In response to growing compute and memory capacity requirements, modern systems are equipped to distribute the work over multiple GPUs and pool the memory from the host and other GPUs transparently. Compute capacity scales out with multiple GPUs, and the memory capacity afforded by the host is an …
-
Gpu Accelerated Imaging Flow Cytometry And Real Time Sorting
… and from GPU by using a modular GPU board with a unified memory. This brought the transfer time from 0.7 ms to 0.1 ms, and the overall system latency down to 8ms. Combined with piezoelectric actuators for sorting, the system was able to perform brightfield image-activated sorting of samples at 56 …
-
True shared memory architecture for next-generation multi-GPU systems
… data duplication, and inefficient use of GPU memory leading to memory capacity issues, and the complexity in programming MGPUs pose serious limitations in the execution of ever-scaling DNN workloads on MGPU systems. To overcome the limitations of existing MGPU systems, we propose to unify the …
-
Directive-Based Data Partitioning and Pipelining and Auto-Tuning for High-Performance GPU Computing
… improvement in both computing power and memory connection bandwidth for accelerators. However, we also observe that the computation power has grown significantly faster than the interconnection bandwidth between the central processing unit (CPU) and the accelerator. Given that …
-
Memory system design and optimizations for data intensive computing
… exhibit tremendous datasets and irregular memory access patterns. Research has shown that these memory-bound applications are unable to effectively leverage the principles of data locality and regular memory accesses within traditional cache-based memory systems to mitigate the …
-
Akcelerace ultrazvukových simulací pomocí multi-GPU systémů
V této práci je kladen důraz na multi - GPU systémy a využití CUDA unifikované paměti.Hlavním cílem je akcelerovat výpočet 3D FFT, který je hlavní součástí simulací knihovny k- Wave .K- Wave je C++/ Matlab knihovna určena pro simulaci šíření ultrazvukových vln v 1D , 2D nebo 3D prostoru.Akcelerace …