CloudBzz

A few days ago, NASA announced their enterprise computing cloud, NEBULA.  As importantly, they announced that the NEBULA  cloud framework was to be released as an open source project.  In NASA’s own words, NEBULA “provides high-capacity computing, storage and network connectivity, and uses a virtualized, scalable approach to achieve cost and energy efficiencies.”

Here is the NEBULA platform.

The core virtualization and cloud services layer are provided by Eucalyptus, the Amazon Web Services open source clone.  Storage is provided by the open source Lustre clustering file system, while the core application development framework is the Python-based Django project.  Note that Google took serious bashing for releasing their AppEngine framework initially with only Python support.  Their IDE is an integrated stack of Subversion (source code control), Trac (bug tracking) and Agilio “agile development” project management tool set.   Lastly, the content repository is searchable using the Solr framework on top of the Apache Lucene search engine.

When this is released to the general public as an open source project, will this be solid competition vs. commercial enterprise cloud frameworks such as 3tera’s AppLogic?  Is this even a valid question?  I’ll see if I can find out…

Christopher Crowhurst, Thomson Reuters’s vice president and chief architect for infrastructure was interviewed for an article in The Industry Standard regarding his plans to implement storage and server virtualization on a global level. With 20,000 servers and over 6.5 petabytes of storage today, Thomson Reuters is a great candidate for virtualization.  One of his coments in this article points to an opening for cloud computing in their enterprise architecture moving forward:

“I think what we’ll eventually start doing is extend this into a private cloud and move to do some self-provisioning for our business units as we get more confident with the management tools in these virtual environments.”

This is a great opportunity for cloud vendors and services providers.

Here is a link.  Worth a read if you are interested in what’s going one with cloud storage (aka Storage as a Service).  This is a market that IDC predicts will hit $3 billion by 2012.  The paper was written by Storage Switzerland, but mentions by name only Bycast – so I assume it’s a paid piece.

There were several more great talks, demos and panels during the day.  See Bitcurrent for good summaries of each one.  Here are some additional bits.

Peter Laird’s talk on Taxonomies is a copy of the one he gave at Interop in NY this past Fall.  Here is his blog post from September 2008.  The vendor taxonomy chart is included in the post. Also, I was pretty impressed with Mike Repass, product manager for Google AppEngine.  They still have a lot of work to do to get AppEngine to be a major player with the lock-in they require, but their “serverless” scalability model is very interesting.  

I then attended CloudCamp which had some interesting discussions. I attended two breakout discussions – one on Amazon’s new auto-scalability capabilities, and the other on enterprise issues. 

There was more substantive content at the CloudCamp than the Enterprise Cloud Summit.  Hmm….

At the Enterprise Cloud Summit today at Interop, there were several examples given of constraints imposed by governments on where data and processes can reside.  For example, Canadian government data cannot reside in the U.S. due to the Patriot Act.  Similarly, the French government will not use Blackberry devices because at some point all emails route through the U.S. and also become visible to the Department of Homeland Security.

And it’s not just other countries.  Even in the U.S. there are different constraints on NPPI (non-public personal information) at the state level.  How can enterprises use cloud services where they have no control on the physical location of their data and processes in this patchwork of conflicting laws and regulations.  Can a CIO risk regulatory or even criminal liability against their company in order to get the benefits of cloud computing?

It is possible that over time these constraints may seriously retard the growth of cloud computing on a global basis.  At that point, is it possible that we may see a global treaty effort on cross-border privacy and infrastructure computing?

I will post more updates later, but the connection here at the Enterprise Cloud Summit (#ecs) makes it hard for me to do full-length posts.

Seeing demo now of SOASTA CloudTest seening load test stats for the sample app that they have been working with at the conference.

Panel led by Greg Ness from InfoBlox coming next…  Panelists include Bill McGee from Third Brigade, Geva Perry of Thinking Out Cloud, and Randy Rowland of Terremark.  Starting with “What types of cloud architectures are there and are they all alike?”  Now people are recapping stuff from this morning.  

Lots of vaporware and “vision lock-in” trying to get customers to hold-off.  Geva commenting about billing options for RackSpace or Terremark vs. Amazon (AWS only bills via ccard vs. others who will invoice IT).

“What are critical cloud dependencies… where things could go wrong?”

Geva – lots already discussed by James Statten etc.    Think carefully about who will become hub of ecosystems.  Amazon is there now.  Think iPhone ecosytem effect.