Is your infra big data optimized?


Here’s a simple quiz.
Which are the hottest three letters in the ICT world today?
Answer: ABC. Analytics, Big Data, & Cloud.
Which are the hottest four letters in the manufacturing world today?
Answer: IIOT. Industrial Internet of Things.
Which are the hottest five letters in the data center (DC) world today?
Answer: SCORA. Scalable, Capable, Open, Refreshable. Affordable.
ABC includes BI (business intelligence), BA (business analytics) and BD (big data). Cloud includes private, public and hybrid. IIOT refers to the integration of physical machinery with networked sensors, devices, gateways and software. And SCORA is an acronym for big data-optimized infrastructure.

What’s big data? It means different things to different people. The one definition I like is from TechTarget which defines big data as an evolving term that describes any voluminous amount of structured, semi-structured and unstructured data that has the potential to be mined for information.

Why is big data so hot? Because enterprises and governments are rushing to understand, deploy and wring out predictive and prescriptive analytics out of it. A new forecast from IDC estimates that the big data technology and services market will cross US$48 billion by 2019 worldwide. That’s a compounded growth of 23% every year between now and then.

So what’s SCORA got to do with it? A lot. SCORA is a well-kept industry secret. It is what large enterprises seek in their core DC infrastructure that will run ABC or IIOT. The three core components in DCs are servers, storage and switches (or networking). Here’s a snapshot of SCORA and why it matters.

Scalable refers to the server’s architecture being able to support scaling out (adding more servers) or scaling up (adding more compute, memory or I/O capabilities on the same 2U rack. “Many enterprise-grade servers are incorporating technologies designed to enhance reliability and avoid downtime,” notes Stephen Bigelow, Senior Tech Editor at TechTarget. “Techniques that were once in the domain of the most powerful and expensive systems are quickly filtering down to entry-level models.”

Capable refers to servers that can run even if the air-conditioning doesn’t. Most DCs are intended to operate from 18 deg C to 35 deg C (95 deg F). Can you run your DC at higher temperatures and humidity? Yes, says Dell. Dell Fresh Air 2.0 hardware includes specific configs of server, networking and storage products which can operate at up to 45 deg C/113 deg F at a 29 deg C max dew point. DCs can also use clean outside air (meeting ISA-71 G1) for air intake. (Disclaimer:#Iwork4Dell)

Open refers to open standards. In servers, the standard is x86. Since its launch in 1978 by Intel, it has gone through many additions/extensions, with Intel ensuring backward compatibility. The architecture has been implemented in processors from Intel, Cyrix, AMD, VIA and others. In operating systems, the two de facto standards are Windows and Linux. But applications don’t follow any standards. But as long as they work on Windows or Linux on an x86 platform, they would be ABC compatible.

Refreshable has two components. If you’re running OpenStack, this open-source cloud refreshes every six months. The DC hardware should be able to handle the refresh seamlessly. As for the processor, it must be backward compatible, offer in-mem computing, up to 99.999% uptime and address up to 96TB on a 2U rack. The Intel Xeon E7 v3 addresses all this.

Affordable refers to two factors. One, feature-for-feature which server vendor is the lowest priced and offers the most value for money. Two, on a three-year TCO (total cost of ownership), or for a production-grade deployment, which vendor offers the most bang for the buck.

The final point. Start your big data journey the PPP way. That’s PoC to Pilot to Production. PoC is Proof of Concept; you start with a small stack, get your hands wet, get small wins and scale up. Pilot is when you overload your stack to set the point of failure, so that just before you get there, you will deploy more resources to scale up/out. And finally, ramp up to Production to support all your use cases.
Use SCORA as your baseline and PPP as your roadmap as you move to IIOT

Is Big Data Making Us Digital Entitities?

Previous article

Beyond Wearables

Next article

You may also like


Leave a reply

Your email address will not be published. Required fields are marked *

More in Articles