There was much talk about cloud computing today, all of it hitting different aspects — from how IT organizations will adopt it to what makes a “niche” cloud to how AT&T’s spotty network helped drive the need for it. Some Big Data items also caught my eye, including the results of a test on Hadoop efficiency and new features added to Cassandra.
A Profile of Apache Hadoop MapReduce Computing Efficiency (continued) (From the Cloudera blog) I posted the first part of this yesterday, and this is the result. It looks like running jobs using Hadoop is not as efficient as one might desire, but there are adjustments that can be made to improve results.
Why We Have the iPhone and AT&T to Thank For Cloud… (From Rational Survivability) This is an interesting take on how At&T might have, inadvertently, helped drive cloud computing to where it is today. No awful network, no App Store.
In a World of Niche Clouds, How Do You Define a Useful Niche? (From CloudAve) This is a good analysis of what actually makes a cloud a “niche” cloud. I think they’ll continue to proliferate, but providers need to spell out the differentiation.
Cloud Computing Hits #1 at Gartner Data Center Conference (From Data Center Dialog) As usual, though, Gartner ignores the startups that are driving the cloud movement. Whatever the large vendors do, they’ll accomplish it by acquisition.
What’s New in Cassandra 0.7: Hadoop Output to Cassandra (From the Riptano blog) It says something about Hadoop that the big improvement in Cassandra 0.7 is better Hadoop interoperability. Bi-directional writing is actually a big deal, though.
For more cloud-related news analysis and research, visit GigaOM Pro.