As informa??es desta p¨¢gina n?o est?o completamente dispon¨ªveis no seu idioma de escolha. Esperamos disponibiliza-las integralmente em outros idiomas em breve. Para ter acesso ¨¤s informa??es no idioma de sua prefer¨ºncia, fa?a o download do PDF ²¹±ç³Ü¨ª.
Atualizado em : Jul 08, 2014
N?O ENTROU NA EDI??O ATUAL
Este blip n?o est¨¢ na edi??o atual do Radar. Se esteve em uma das ¨²ltimas edi??es, ¨¦ prov¨¢vel que ainda seja relevante. Se o blip for mais antigo, pode n?o ser mais relevante e nossa avalia??o pode ser diferente hoje. Infelizmente, n?o conseguimos revisar continuamente todos os blips de edi??es anteriores do Radar.
Saiba mais
Jul 2014
Adote
Hadoop's initial architecture was based on the paradigm of scaling data horizontally and metadata vertically. While data storage and processing were handled by the slave nodes reasonably well, the masters that managed metadata were a single point of failure and limiting for web scale usage. Hadoop 2.0 has significantly re-architected both HDFS and the Map Reduce framework to address these issues. The HDFS namespace can be federated now using multiple name nodes on the same cluster and deployed in a HA mode. MapReduce has been replaced with YARN, which decouples cluster resource management from job state management and eliminates the scale/performance issues with the JobTracker. Most importantly, this change encourages deploying new distributed programming paradigms in addition to MapReduce on Hadoop clusters.
Jan 2014
Experimente
May 2013
Experimente
Publicado : May 22, 2013
