Invisible technology

(Image: Stockfresh)

Print

PrintPrint
Longform

Read More:

10 February 2016 | 0

ICT has been using the analogies with ‘commodities’ and ‘utilities’ and so on for years. Cloud computing certainly extended that trend, because in truth public cloud services are very like such commodity services. The idea is essentially that we can plug in to computing resources as easily as turning on our supply of electricity or gas or oil. The snag, of course, is that energy resources are omni-purpose — heat, light or run an engine or indeed a computer.

Computing spans a huge range but it is omni-purpose only in a differentiated sense. Massive data storage is not the same as the need for high performance compute capability while communications today range from the few bytes of an instant message or a text to HD full motion video-on-demand. The only true ‘commodity’ elements are the bits and bytes and the broadband channel. In fact, it could be validly argued that it is the connectivity that is the commodity or utility service, not the ICT activity that is delivered or invoked.

OLYMPUS DIGITAL CAMERA

In a future where most businesses will embrace a hybrid cloud strategy it is imperative that the infrastructure of the private data centre closely matches the public cloud, both in how it is built and in the experience it delivers. That certainly means the ability to cope with elastic workloads without intervention, which in turn demands automation and the use of smart predictive analytics tools, Howard Ting, Nutanix

Good terms
But ‘Invisible technology’ is a good term, no vaguer than Cloud, that we can understand: we want the complexity and the technical bits to disappear into the background as we concentrate on what we actually want to do with computing. Consumers, business and institutional users alike want to concentrate on the current daily tasks and the smart and attractive capabilities that are flowing down the pipeline. So do CIOs and IT teams and their employers.

Despite the fact that it is figuratively and actually the most massive set of ICT components, it is the infrastructure that we want to make invisible — or hide. Nutanix and its marketing VP Howard Ting have been using the term ‘Invisible Infrastructure’ recently in a message that is certainly resonating with corporate ICT: “Our vision is to make the entire physical and virtual infrastructure stack invisible. Too much of our budgets, time and energy is spent managing the servers, the storage, the networks — racks and racks of boxes that take a lot of manual administration to manage and keep running. But really those are all very low value activities. Yet analysts suggest regularly that 75 or 80% of an organisation’s ICT budget and staff time is taken up by just managing and maintaining the infrastructure.”

Invisible idea
Public cloud services, in spite of their many shortcomings, do give us some idea of what invisible infrastructure looks like, Ting says. “By and large, whatever the service required, it just works. You also do not have to plan two or three years ahead for your anticipated capacity requirements — and pay up front and probably over-provision. You can draw down what you need, when you need it. Amazon Web Services in particular has transformed the expectations in our industry. That huge change in the consumption model for infrastructure has also completely changed the offerings of infrastructure technology companies.”

Ian Massingham, Amazon_web

The idea is that developers do not have to think about, much less manage, any server infrastructure. AWS Lambda allows developers to create functions, in Java or Python, for example, to upload them to Lambda together with specified event triggers. When the service receives a trigger it dynamically provisions the computing resources required to execute the function. There is zero administration. Once again, it abstracts the infrastructure even further away. But the service also allows fine-grained pricing, literally in milliseconds, Ian Massingham, Amazon

The aim of Nutanix, Ting says, is to bring the same attributes of simplicity, scalability and agility to the enterprise data centre. “In a future where most businesses will embrace a hybrid cloud strategy it is imperative that the infrastructure of the private data centre closely matches the public cloud, both in how it is built and in the experience it delivers. That certainly means the ability to cope with elastic workloads without intervention, which in turn demands automation and the use of smart predictive analytics tools. But that is what is demanded of ‘invisible infrastructure’. For the users, it can be as simple as an infographic showing simply how many days of capacity you have left.”

A key part of the Nutanix pitch, Ting says, is that an organisation no longer has to employ infrastructure and storage specialists. “Your IT team will have generalists, focussed principally on your applications and new elements. But the Nutanix solutions have so much automated management capability built in that they will be perfectly well able to make decisions that the system presents to them. It’s the same reason that a typical large organisation might have an administrator looking after 500 VMs when in web scale operations like Facebook and others the ratio may be more like one administrator to 12,000 VMs. Nutanix brings hyper-converged web scale infrastructure to the enterprise data centre.”

Avoiding heavy lifting
Amazon Web Services is clearly a technology leader in public cloud services and Ian Massingham is its Technology Evangelist in the UK and Ireland. He happily suggests that AWS offers many services that help ICT professionals, and not just infrastructure specialists, avoid a lot of the heavy lifting traditionally involved in operating data centres and networks.

“Application developers previously had to apply for the resources to be allocated to allow them to work on something new. We give them, for example, Elastic Beanstalk to allow them to test and run their apps in a deployment environment, in a variety of programming languages, without worrying about the infrastructure. When they push their apps into it, AWS will create all of the resources to enable them to run, usually automatically, in response to the specifications.”

Read More:



Leave a Reply

Back to Top ↑