Los grandes centros de proceso de datos (CPDs) son hoy en d\u00eda infraestructuras fundamentales en la transformaci\u00f3n digital de nuestra Sociedad. Gracias a su capacidad para resolver problemas complejos mucho m\u00e1s all\u00e1 de la capacidad de las computadoras ordinarias, estos sistemas impulsan avances en campos tan diversos como medicina, climatolog\u00eda, f\u00edsica de part\u00edculas, transmisi\u00f3n de v\u00eddeo, Inteligencia Artificial (IA) generativa y, en general, aceleran significativamente el desarrollo de pr\u00e1cticamente cualquier sector con potencial innovador. En los CPDs, la red de interconexi\u00f3n es un subsistema clave que permite que cientos de miles de nodos de c\u00f3mputo y almacenamiento cooperen para ejecutar las aplicaciones que se utilizan en los campos mencionados anteriormente. Estas aplicaciones generan innumerables operaciones de comunicaci\u00f3n a las que la red debe dar soporte lo m\u00e1s r\u00e1pidamente posible, convirti\u00e9ndose de lo contrario en el cuello de botella de todo el sistema. Hasta hace unos pocos a\u00f1os, un CPD era un sistema independiente y aislado, donde el \u00fanico enlace que necesitaba con el mundo exterior era relativamente sencillo, habitualmente s\u00f3lo para mostrar alguna visualizaci\u00f3n a los usuarios. Sin embargo, el gran volumen de datos y carga de trabajo que manejan estos sistemas en los \u00faltimos a\u00f1os, hace que surja la necesidad de interconectar varios CPDs, que pueden estar distribuidos geogr\u00e1ficamente. Por otro lado, actualmente cada nodo de c\u00f3mputo de los CPDs incluye una gran cantidad de aceleradores que se interconectan por medio de una red intra-nodo. Bas\u00e1ndose en este modelo de red de interconexi\u00f3n con tres niveles (intra-nodo, inter-nodo e inter-CPD), desde la industria y el \u00e1mbito acad\u00e9mico se est\u00e1 trabajando en desarrollar est\u00e1ndares de fabricaci\u00f3n y nuevas tecnolog\u00edas de red que satisfagan los requisitos de comunicaci\u00f3n ya mencionados.<\/span><\/p>\n <\/p>\n Datacenters (DCs) are nowadays critical infrastructures in the digital transformation of our Society. Thanks to their ability to solve complex problems far beyond the capacity of ordinary computers, these systems drive advances in fields as diverse as medicine, climatology, particle physics, video transmission, generative Artificial Intelligence (AI) and, in general, significantly accelerate the development of any sector with innovative potential. In data centers, the interconnection network is a key subsystem that enables hundreds of thousands of compute and storage nodes to cooperate to run the applications used in the fields mentioned above. These applications generate countless communication operations that the network must support as quickly as possible, otherwise it becomes the bottleneck of the entire system. Until a few years ago, a DC was a stand-alone, isolated system, where the only link it needed with the outside world was simple, usually just to display some visualization to users. However, the large volume of data and workloads handled by these systems in recent years has created the need to interconnect several data centers, which may be geographically distributed. On the other hand, each compute node in the data center currently includes a variety of accelerators that are interconnected by means of an intra-node network. Based on this three-level interconnection network model (intra-node, inter-node, and inter-DC), industry and academia are working to develop manufacturing standards and new network technologies that meet the aforementioned communication requirements.\u00a0<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n","protected":false},"excerpt":{"rendered":" T\u00e9cnicas Eficientes para Tecnolog\u00edas de Red Avanzadas (TETRA 3) Los grandes centros de proceso de datos (CPDs) son hoy en d\u00eda infraestructuras fundamentales en la transformaci\u00f3n digital de nuestra Sociedad. …<\/p>\n","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"open","template":"","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"class_list":["post-2","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/hipineb.i3a.info\/tetra-project\/wp-json\/wp\/v2\/pages\/2","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hipineb.i3a.info\/tetra-project\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/hipineb.i3a.info\/tetra-project\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/hipineb.i3a.info\/tetra-project\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/hipineb.i3a.info\/tetra-project\/wp-json\/wp\/v2\/comments?post=2"}],"version-history":[{"count":7,"href":"https:\/\/hipineb.i3a.info\/tetra-project\/wp-json\/wp\/v2\/pages\/2\/revisions"}],"predecessor-version":[{"id":200,"href":"https:\/\/hipineb.i3a.info\/tetra-project\/wp-json\/wp\/v2\/pages\/2\/revisions\/200"}],"wp:attachment":[{"href":"https:\/\/hipineb.i3a.info\/tetra-project\/wp-json\/wp\/v2\/media?parent=2"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}\n\n
\n Efficient Techniques for Advanced Interconnect Technologies (TETRA 3)<\/strong><\/em><\/p>\n