Computation |
Recent Submissions
-
Instruction-aware mapping (IAM) : a tool to mitigate functional unit contention in SMT processors
(2023) [Thesis]Modern computing architectures frequently rely on Simultaneous Multithreading (SMT) processors to boost computational throughput and handle parallel applications efficiently. However, the potential of SMT can be compromised ... -
Efficient and scalable load balancing through cross-plane cooperation
(2023) [Dissertation]Load balancing network traffic through multiple shortest-paths has become common practice to efficiently utilize the network infrastructure. Despite widespread adoption, EqualCost Multi-Path (ECMP) delivers performance far ... -
A reliability- and variation-aware methodology for improved processor designs for the edge computing domain
(2023) [Thesis]Technology scaling has been successfully improving the performance of current microprocessors primarily due to the reduced node size that enables increased transistor integration, allowing for the design and widespread ... -
A methodology for synthetic generation of failure data for data-driven prognostics and health management modeling for digital twins
(2023) [Dissertation]Prognostics and Health Management (PHM) is one of the main services encompassed by Industry 4.0. However, the scarcity of failure data due to the nature of machines’ opera tion is still a challenge to be transposed in this ... -
Improving network resilience through an environmental context-aware system
(2023) [Dissertation]Networks are a crucial resource in the digital age, enabling applications in domains such as agriculture, smart cities, and military systems. As technology advances, some applica tions require a more reliable and predictable ... -
Background subtraction based on robust subspace tracking
(2023) [Dissertation]Este trabalho propõe um método robusto de subtração de fundo online, que detecta o fundo e o primeiro plano em sequências de vídeo. Os métodos de subtração de fundo existentes tendem a ter um desempenho limitado em cenários ... -
Mitigando o impacto de dados non-IID em federated learning com entropia
(2023) [Dissertation]Algoritmos de Machine Learning (ML) possibilitam processar um conjunto de dados de entradas para gerar coeficientes que ajustem a saída a um resultado previamente conhecido, como menor erro possível, fazendo com que seja ... -
Uma abordagem integrada para posicionamento e composição de medições ativas de rede
(2023) [Dissertation]A evolução incessante da Internet originou uma vasta variedade de aplicações e serviços, elevando a importância de assegurar sua performance e segurança. Neste cenário, o monitoramento de redes é primordial para a detecção ... -
Client-transparent and self-managed MQTT broker federation at the application layer
(2023) [Dissertation]Scalability in messaging systems remains an open topic. Regular solutions present clus terized approaches, which can be very scalable for high-throughput systems but still rely on a unique orchestrator which is a single ... -
DWT in P4 : periodicity detection in the data plane
(2023) [Dissertation]This dissertation presents an extended implementation of the (1-D) Discrete Wavelet Transform (DWT) method in the P4 programming language, enabling efficient and real time analysis of periodic behavior in network traffic. ...
RSS 1.0
