Naganarasimha Garla

华为印研Hadoop工程团队架构师，Hadoop Committer

I am currently working on Apache Hadoop, an open source distributed computing system, with Huawei Technologies. I am an Architect in Hadoop Engineering Team. My primary focus in Hadoop has been for YARN and MapReduce. My major contributions has been in YARN Node Labels and YARN ATS feature and I am also a Apache Hadoop Committer. Prior to Hadoop, I worked in providing a Big Data Analaytics (OLAP) capability to Huawei's Data Platform over HBase.

Scheduling layer of Hadoop have significantly evolved from the MRv1 days. Current Scheduler supports variety of applications and heterogenous hardware. Lot of improvements have been done for Multitenant scenarios too. And lot more is coming up to enable YARN to be the defacto Resource Scheduler of any BigData cluster. In this presentation I would like to present evolution of YARN scheduler to present day scheduler and its road map for the future work and how we in Huawei are contributing to it.