January SF Hadoop Users Meetup

The January SF Hadoop User Group meetup will be held Wednesday, January 21 from 6:00pm to 8:00pm. This meetup will be hosted by BlueData and held at Pivotal Labs, 875 Howard Street, San Francisco. Food and drinks will be served.

Presentation: Hadoop Use Cases vs. Run Time Environments

Tom Phelan, Chief Architect and co-founder, BlueData

This presentation will investigate the advantages and disadvantages of running a Hadoop cluster in each of the following environments:

• bare metal

• public cloud (i.e. Amazon EMR, Azure, GCE, Rackspace etc)

• private cloud (i.e. on-premises cloud)

• in cloud environments

• cloud using a traditional virtual machine environment

• cloud using a containerized virtual environment

The presentation will include a discussion of real-world Hadoop uses cases and how each would, or would not, be suitable for running in each environment.

There will be two breakout sessions following the presentation:

1. Hadoop and compute virtualization

How do the unique compute requirements of a Hadoop cluster match up against the strengths and weaknesses of such virtualization environments as traditional VMs (KVM, VMW, etc) and containers (docker, etc)?

2. HDFS and storage virtualization

What are the benefits and drawbacks of "virtualizing" HDFS? How can technologies such as Tachyon, MemHDFS, and distributed file systems such as GlusterFS and CEPH aid this storage virtualization?