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 20 of 78 for “"Network Hardware"”.

  1. Inference neural network hardware acceleration techniques

    A number of recent researches focus on designing accelerators for popular deep learning algorithms. Most of these algorithms heavily involve matrix multiplication. As a result, building a neural processing unit (NPU) beside the CPU to accelerate matrix multiplication is a popular approach. The NPU …

    uiuc Repository record for Inference neural network hardware acceleration techniques (opens in a new tab)

  2. Co-Designing Distributed Systems with Programmable Network Hardware

    … distributed systems with the datacenter network. Specifically, we have taken advantage of new-generation programmable switches in datacenters to build several novel network-level primitives that offer strong guarantees. We then leveraged these primitives to enable more efficient protocol …

    washington Repository record for Co-Designing Distributed Systems with Programmable Network Hardware (opens in a new tab)

  3. TOWARD A RUNTIME PROGRAMMABLE SPIKING NEURAL NETWORK HARDWARE ACCELERATOR WITH ON-CHIP LEARNING

    … have led to the emergence of Spiking Neural Networks (SNNs), energy-efficient machine learning algorithms inspired by the activities of the biological brain. Due to their event-driven nature, SNNs are expected to consume less energy than Artificial Neural Networks (ANNs). However, …

    nus Repository record for TOWARD A RUNTIME PROGRAMMABLE SPIKING NEURAL NETWORK HARDWARE ACCELERATOR WITH ON-CHIP LEARNING (opens in a new tab) Download file for TOWARD A RUNTIME PROGRAMMABLE SPIKING NEURAL NETWORK HARDWARE ACCELERATOR WITH ON-CHIP LEARNING (opens in a new tab)

  4. Energy-Efficient Neural Network Hardware Design and Circuit Techniques to Enhance Hardware Security

    Artificial intelligence (AI) algorithms and hardware are being developed at a rapid pace for emerging applications such as self-driving cars, speech/image/video recognition, deep learning, etc. Today’s AI tasks are mostly performed at remote datacenters, while in the future, more AI workloads are …

    umn Repository record for Energy-Efficient Neural Network Hardware Design and Circuit Techniques to Enhance Hardware Security (opens in a new tab)

  5. Efficient Design of Neural Network Hardware Accelerator for Enhanced Performance in Deep Learning Systems

    Artificial intelligence (AI) hardware accelerator is an emerging research for several applications and domains. The hardware accelerator aims to provide high computational speed while retaining low cost and high learning performance. The main challenge is to design complex machine learning models …

    mississippi Repository record for Efficient Design of Neural Network Hardware Accelerator for Enhanced Performance in Deep Learning Systems (opens in a new tab)

  6. Energy-efficient recurrent neural network hardware architecture for heart rate estimation based on photoplethysmography

    … is an ever-growing interest in employing neural network-based applications for data processing on edge devices. Neural networks have complex structures in their pure software or hardware implementations, or in a combination of both approaches. They require millions of data fetches and arithmetic …

    brazil-ufrgs Repository record for Energy-efficient recurrent neural network hardware architecture for heart rate estimation based on photoplethysmography (opens in a new tab)

  7. BUILDING NETWORKED SYSTEMS FOR TERABIT ETHERNET

    Over the last two decades, hardware in datacenters has shown diverging trends. On the one hand, access link bandwidth has increased rapidly, with datacenters now commonly supporting Terabit Ethernet (i.e., Ethernet with speeds above 100Gbps). Modern network hardware is capable of supporting …

    cornell Repository record for BUILDING NETWORKED SYSTEMS FOR TERABIT ETHERNET (opens in a new tab)

  8. Graphical user interface network applications

    … and software applications on a local area network. Requirements for a graphical user interface on a local area network are discussed, specifically the Systems Management Department laboratories at Naval Postgraduate School. Compatibility and interoperability of network operating system, PC …

    nps Repository record for Graphical user interface network applications (opens in a new tab)

  9. An integrated architecture to support the Hastily Formed Network (HFN)

    This thesis analyzes the nature of Hastily Formed Networks (HFN) and develops a framework and architecture for effective collaboration needs. The analysis considers review of a spectrum of Commercial off-the-shelf (COTS) hardware (including Mobile Command and Control (C2) center and network …

    nps Repository record for An integrated architecture to support the Hastily Formed Network (HFN) (opens in a new tab)

  10. Design and implementation of next-generation accelerated networking systems

    … and rely heavily on ultra-high-performance networks for connectivity. As a result, the network has become a critical factor influencing application scalability and performance, with link speeds now approaching terabit levels. Keeping up with this speed has become increasingly difficult as …

    texas Repository record for Design and implementation of next-generation accelerated networking systems (opens in a new tab)

  11. Advanced threat hunting over software-defined networks in smart cities

    The emergence of Software-Defined Networking (SDN) has brought along a wave of new technologies and developments in the field of networking with hopes of dealing with network resources more efficiently and providing a foundation of programmability. SDN allows for both flexibility and adaptability …

    utc Repository record for Advanced threat hunting over software-defined networks in smart cities (opens in a new tab)

  12. Direct Photon Differential Cross Section in p̄p Collisions at the Square Root of s = 1.8 TeV

    … of the Central Preradiator Chambers, the neural network hardware trigger upgrades, and the six times increase in integrated luminosity. Two different methods, conversion method and profile method, were used to separate prompt photons from photons produced by decay of hadrons. The profile method …

    rockefeller Repository record for Direct Photon Differential Cross Section in p̄p Collisions at the Square Root of s = 1.8 TeV (opens in a new tab)

  13. Web-based network management configuration for the Indonesian Eastern Fleet Wide Area Network

    … presents a model of an integrated wide area network using web-based network management to support fleet operations of the Indonesian Eastern Fleet. It surveys possibilities for improving the Indonesian Eastern Fleet's computer communications network systems to provide a fast, reliable, and …

    nps Repository record for Web-based network management configuration for the Indonesian Eastern Fleet Wide Area Network (opens in a new tab)

  14. JITNIC: Just-in-Time Compilation for In-Network Processing

    … to design and effectively leverage specialized hardware components. This has taken shape in the domain of networking in the form of programmable switches and new domain-specific languages used to configure such devices. However, a large portion of the packet-processing workload is still handled …

    cornell Repository record for JITNIC: Just-in-Time Compilation for In-Network Processing (opens in a new tab)

  15. Congestion-Aware Path Selection for Load Balancing in AI Datacenters

    … influenced by the efficiency of the underlying network infrastructure, typically based on RDMA and Clos-like data center topologies. Effective load balancing across mul-tiple redundant network paths is essential to fully utilize available bandwidth. However, existing load balancing solutions are …

    calgary Repository record for Congestion-Aware Path Selection for Load Balancing in AI Datacenters (opens in a new tab)

  16. Spider : An agent-based message-passing architecture

    … hence the importance of controlling network resources. Current network designs offer a closed, rigid interface that emphasizes hardware solutions for routing, leaving little room for communication control. In this thesis we address interface issues between network hardware and …

    ubc Repository record for Spider : An agent-based message-passing architecture (opens in a new tab) Download PDF for Spider : An agent-based message-passing architecture (opens in a new tab)

  17. A Framework for the Self-Configuration of Wireless Mesh Networks

    … step towards providing high-bandwidth wireless network coverage, spectral efficiency, and economic advantage. However, the widespread adoption and use of Wireless Mesh Networks (WMN) as a backbone for large wireless access networks and for last-mile subscriber access is heavily dependent on the …

    waterloo Repository record for A Framework for the Self-Configuration of Wireless Mesh Networks (opens in a new tab)

  18. Automated cyber threat analysis and specified process using vector relational data modeling

    Computer network defense systems should be sufficiently integrated to pull data from any information source, model an expert cyber analyst’s decision process, continuously adapt to an evolving cyber threat environment, and amalgamate with industry standard network hardware. Unfortunately, cyber …

    nps Repository record for Automated cyber threat analysis and specified process using vector relational data modeling (opens in a new tab)

Page 1 of 4