Use this link to cite:
http://hdl.handle.net/2183/33996 Desenvolvemento dun contorno virtualizado de GPUs con VMware para a optimización do uso de recursos
Loading...
Identifiers
Publication date
Authors
Lareu Rendo, Jorge
Advisors
Other responsabilities
Universidade da Coruña. Facultade de Informática
Journal Title
Bibliographic citation
Type of academic work
Academic degree
Abstract
[Resumo]: Actualmente vivimos nunha era na que a computación clásica estase a abrir a novas frontes
descoñecidas que non seguen na totalidade os paradigmas marcados ata agora. Nestes momentos,
tecnoloxías como a Intelixencia Artificial, o Machine Learning ou o Big Data
supoñen entidades estratéxicas nas tácticas de mercado de calquera empresa e son unha revolución
no novo mundo dos asistentes virtuais como pode ser Chat GPT. Debido a este
novo paradigma, o uso das GPU de propósito xeral (GPGPU) gañan un gran terreno nos CPD,
xa que, debido a súa arquitectura son capaces de executar algoritmos de, por exemplo, IA ou
ML, moito máis rápido que os chips CPU. Este proxecto pretende aproveitar as capacidades
das GPGPU e da virtualización para aportar unha solución que optimice o emprego de recursos
no centro de procesamento de datos (CPD) do Centro de Investigación en Tecnoloxías da
Información e as Comunicacións (CITIC) da Universidade da Coruña. Mediante o uso da virtualización
podemos acadar uns niveis de aproveitamento dos recursos hardware moito maior
en tres ámbitos principalmente: o uso de procesador, o uso de almacenamento e por último,
e máis importante para este proxecto, o uso de procesadores gráficos, optimización na que se
centra a idea na que se basea este traballo.
[Abstract]: Currently, we are living in an era in which classical computing is opening up new unknown frontiers that do not entirely follow the paradigms established until now. At this moment, technologies like Artificial Intelligence, Machine Learning, or Big Data represent strategic entities in the market tactics of any company and are a revolution in the new world of virtual assistants, such as Chat GPT. Due to this new paradigm, the use of General-Purpose Graphics Processing Units (GPGPU) is gaining significant ground in Data Processing Centers (CPDs) since, due to their architecture, they can execute algorithms, for example, AI or ML, much faster than CPU chips. This project aims to leverage the capabilities of General-Purpose Graphics Processing Unit (GPGPU) and virtualization to provide a solution that optimizes resource utilization in the Data Processing Center (CPD) of the Research Center for Information and Communications Technologies (CITIC) at the Universidade da Coruña. Through the use of virtualization, we can achieve significantly higher levels of hardware resource utilization in three main areas: processor usage, storage usage, and, lastly, and most importantly for this project, the use of graphics processors, which is the central focus of the idea upon which this work is based.
[Abstract]: Currently, we are living in an era in which classical computing is opening up new unknown frontiers that do not entirely follow the paradigms established until now. At this moment, technologies like Artificial Intelligence, Machine Learning, or Big Data represent strategic entities in the market tactics of any company and are a revolution in the new world of virtual assistants, such as Chat GPT. Due to this new paradigm, the use of General-Purpose Graphics Processing Units (GPGPU) is gaining significant ground in Data Processing Centers (CPDs) since, due to their architecture, they can execute algorithms, for example, AI or ML, much faster than CPU chips. This project aims to leverage the capabilities of General-Purpose Graphics Processing Unit (GPGPU) and virtualization to provide a solution that optimizes resource utilization in the Data Processing Center (CPD) of the Research Center for Information and Communications Technologies (CITIC) at the Universidade da Coruña. Through the use of virtualization, we can achieve significantly higher levels of hardware resource utilization in three main areas: processor usage, storage usage, and, lastly, and most importantly for this project, the use of graphics processors, which is the central focus of the idea upon which this work is based.
Description
Editor version
Rights
Atribución 3.0 España








