Free Professional Development in Hadoop and NoSql

(Excerpt from original post on the Taneja Group News Blog)

A wonderful trend we have not only noticed but taken advantage is free online education. In the academic world, MIT has been offering it for years with OpenCourseWare, and recently even Harvard announced it’s producing free classes at edX. But today it is not just heavy-duty term-length classes in stats, data mining, AI and multivariate calculus – although those are all valuable and interesting (well, I am a geek). What is really exciting is that professional development is following suit, especially in IT and Big Data.

Certainly there has long been IT vendor-sponsored education and sometimes its even “free”, but when it is focused on the vendor’s own proprietary commercial offerings there is some question of the actual agenda. Will you spend most of your time learning how to just use that vendor’s solutions? This can be valuable if you own them, although you still might suspect that this kind of education is really filling in for lack of great usability or built-in automation features. I’m reminded of some of the old Novell and Windows certifications that were more about memorizing the trivia of the product lines than learning how to capitalize on the capabilities they enabled. I personally considered that indoctrination more than professional development.

What you can look for today are courses in learning and applying new solutions and technologies, not just in vendor-specific products. A good example is the IBM sponsored BigDataUniversity offering free classes on Hadoop. Yes, there are also free classes on IBM’s DB2 and you can graciously use IBM’s distro of Hadoop to play with, but we are pleased with the vendor-neutral presentation.

If you are interested in NoSQL and the open source MongoDB in particular, you can get onto 10gen’s free education classes on MongoDB that start next week. Although there is a class listed for developers, IT folks whipping out scripts here and there to automate and integrate systems might really find some utility in learning how to leverage MongoDB’s document-centric database. And another course listed for DBA’s could be relevant to IT and Dev Ops folks – how do you actually shard and scale-out these kinds of NoSQL solutions?

While its clear that offering education can be a vendor marketing strategy, I think it can be a good one for both the vendor and the user community. Most companies encourage professional development overall, but your local HR might not promote much in the way of the cutting edge technical development that IT folks desire. I’d encourage some quick googling to see what’s out there – and maybe even request something specific from your favored vendors. Free online technical development courses look like a win-win all the way around.

…(read the full post)

Riverbed’s Edge Virtualization Redefines What Data “Center” Means

(Excerpt from original post on the Taneja Group News Blog)

We’ve seen one physical barrier after another fall in the IT datacenter under increasing advances in virtualization.  It seems every hard-wired IT resource we’ve ever had to manage physically is experiencing re-invention as new “software defined” capabilities separate the “control plane” from the “data plane”. We certainly have a lot to explore and learn about where IT can go with the increasing virtual decoupling of the logical from the physical under increasingly dynamic management control.

Riverbed, leveraging their world-class WAN optimization and storage “extension” solutions, has just rolled out an interesting new capability that they are calling edge-virtualization. Basically, Riverbed’s latest Steelhead EX coupled with Granite storage not only enables highly efficient datacenter consolidation of physical resources, like servers and storage brought in from remote network “edges” while preserving remote performance and productivity, but also supports scenarios in which datacenter managed VM’s themselves are pushed dynamically out to the remote edge locations to run locally.

Sounds a bit like application virtualization where the app is a VM, but that kind of recursive thinking can make your head spin. Better is to just focus on the fact that centralized storage and data is made available at the edge, and now so can live VM’s running datacenter applications. This ability to effectively deliver not just centralized data, but also the processing application to the edge could accelerate further data center consolidation efforts and ensuing benefits.  

Combine this edge-virtualization capability with emerging constructs for managing a “software defined datacenter” (and mix of course with some cloud/hybrid elastic capabilities) and you are really redefining the single physical nature of what a data center means. I’m sure we’ll soon see some very interesting use cases emerging based on Riverbed’s edge-virtualization, not least of which I’m sure will be superb VDI enablement with high ROI due to the underlying consolidation.  

…(read the full post)