It’s a bird, It’s a plane, it’s Supercluster


Whenever I give my kids a choice of breakfast, they will invariably choose what I reckon as the “wrong” choice. It’s always the one with more sugar, or non-natural ingredients. When they don’t have a choice, they always state their preferences. In technology such preferences also exist. Some of them very passionate. Specially after Oracle announced a new Engineered System called SPARC Supercluster (SSC), last October at Oracle Open World 2011 in San Francisco. Whilst the positioning and comparison of SSC seemed quite easy and spontaneous for some, the vast majority still wonders where does this new product sits in the midst of its closest cousins: Exalogic and Exadata. For the former I challenge you to step back a bit and try to think in a more agnostic fashion.

In order to do that let me go back to the concept of “preference”.

So what is a technological preference? It’s preferring a set or subset of technologies in detriment of others (not necessarily the “bad ones”). It’s just a preference. Why would you have a preference? Mainly because it gets the job done. For some people preference is attached to a career or knowledge area, so they are professionally defined by their knowledge. If you want to get the job done in the best way possible you must be technologically agnostic. Technology is a means to an end, not the end itself. So if you’re reading this and you are trusted by your customers, or want to be trusted as a strategic partner, you must concentrate on how technology products would get your customers from point A to point B. When you strip down a solution to its architectural components, you would want it to have building blocks that encapsulate the absolute best of breed in each area.

Let’s say you’re laying down an infrastructure for an enterprise application assuming the following granularity: storage, network, operating system, database, logic and interface.

Do you think that 5 years ago one could sum up all those layers into one single building block? Probably you would have to go even more back in time, but those days of closed systems are doomed to end (mainly because the knowledge to operate those is fading away). Assuming we are talking only about open systems the answer to the question is: no, such large building block didn’t exist. The technological purists might beg to differ, but this is no task for the light hearted. It’s something so big, so complicated, that only a concentration of enterprise class building blocks can achieve. In space research, or astrophysics, a large collection of groups of stars is called a Supercluster, and that’s exactly what Oracle has come up with. The good people of IT departments can now tackle the immensity of this task using a new building block. Your customers might “prefer” mixing other building blocks, it’s up to them to have the pains of integration. On the other hand, if they want to get the job done, they shouldn’t have the meaningless work of connecting an array of diffuse technologies just because of “preferences”. Getting the job done does not comply with these micro-questions.

So what comes inside a Supercluster? Storage, network, operating system, database, logic and interface! How cool is that? Exactly what we needed!

Some might say that they could build such a system but with different components, that would also get the job done. My question to them is: How many price tags would such a system had to have? Supercluster only has one.

LMC

If you want learn more about SSC:

http://www.oracle.com/technetwork/server-storage/hardware-solutions/sparc-supercluster-191641.html

About these ads

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s