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 9 of 9 for “"Compute Accelerators"”.

  1. Methods for mitigating non-idealities in analog-domain compute accelerators

    … have responded by giving increasing attention to Compute in Memory (CiM) and reduced precision computing methods to meet the new levels of parallelism required and overcome the challenges presented by the Memory Wall Bottleneck problem. Of these, methods based on promising emerging materials and …

    texas Repository record for Methods for mitigating non-idealities in analog-domain compute accelerators (opens in a new tab)

  2. High Performance Compute Accelerators for Radio Frequency Signal Processing Applications

    … on the design of scalable, high-performance compute accelerators for RF signal processing applications, with a specific focus on high-bandwidth RF emulation and Multiple Input digital RF beamforming. This is achieved by combining In-memory/Near-memory architectures with high throughput …

    gatech Repository record for High Performance Compute Accelerators for Radio Frequency Signal Processing Applications (opens in a new tab)

  3. Towards scalar synchronization in SIMT architectures

    An important class of compute accelerators are graphics processing units (GPUs). Popular programming models for non-graphics computation on GPUs, such as CUDA and OpenCL, provide an abstraction of many parallel scalar threads. Contemporary GPU hardware groups 32 to 64 scalar threads as a single …

    ubc Repository record for Towards scalar synchronization in SIMT architectures (opens in a new tab) Download PDF for Towards scalar synchronization in SIMT architectures (opens in a new tab)

  4. Design Methodology And Electronic Design Automation Techniques For Heterogeneous 3D Machine Learning Accelerators

    … with scaling near-memory and in-memory compute accelerators in the X/Y dimensions and proposes optimized scaling solutions by leveraging heterogeneous 3D integration. While numerous 3D physical design EDA methodologies exist, they often involve a multitude of parameters that require …

    gatech Repository record for Design Methodology And Electronic Design Automation Techniques For Heterogeneous 3D Machine Learning Accelerators (opens in a new tab)

  5. Proposta de implementação em FPGA de máquina de vetores de suporte (SVM) utilizando otimização sequencial mínima (SMO)

    … importance of Field-Programmable Gate Arrays as compute accelerators has dramatically increased during the last couple of yers. Many companies such as Amazon, IBM and Microsoft included FPGAs in their data centers aiming to accelerate their search engines. In the center of those applications are …

    brazil-ufrn Repository record for Proposta de implementação em FPGA de máquina de vetores de suporte (SVM) utilizando otimização sequencial mínima (SMO) (opens in a new tab)

  6. Matrix factorization and Deep Learning in Scientific Domains: Understanding When and Why It Works

    Propelled by large datasets and parallel compute accelerators, deep neural networks have recently demonstrated human-like performance in many domains previously beyond the reach of machines. Computers can now recognize objects in images, transcribe speech and exhibit reading comprehension at the …

    cornell Repository record for Matrix factorization and Deep Learning in Scientific Domains: Understanding When and Why It Works (opens in a new tab)

  7. Architecture of block-RAM-based massively parallel memory structures : multi-ported memories and content-addressable memories

    … merely used as glue-logic to implementing entire compute accelerators. These massively parallel systems demand highly parallel memory structures to keep pace with their concurrent nature since memories are usually the bottleneck of computation performance. However, the vast majority of FPGA …

    ubc Repository record for Architecture of block-RAM-based massively parallel memory structures : multi-ported memories and content-addressable memories (opens in a new tab) Download PDF for Architecture of block-RAM-based massively parallel memory structures : multi-ported memories and content-addressable memories (opens in a new tab)

  8. A GPU performance estimation model based on micro-benchmarks and black-box kernel profiling

    … in the High Performance Computing sector as compute accelerators. The primary characteristics that justify this modern trend are the exceptionally high compute throughput and the remarkable power efficiency of GPUs. However, GPU performance is highly sensitive to many factors, e.g. the type …

    greece Repository record for A GPU performance estimation model based on micro-benchmarks and black-box kernel profiling (opens in a new tab)

  9. Techniques to maximize memory bandwidth on the Rigel compute accelerator

    The Rigel compute accelerator has been developed to explore alternative architectures for massively parallel processor chips. Currently GPUs that use wide SIMD are the primary implementations in this space. Many applications targeted to this space are performance limited by the memory all, so …

    uiuc Repository record for Techniques to maximize memory bandwidth on the Rigel compute accelerator (opens in a new tab)