An app-centric capacity planning approach to IoT
From the server rack to the storage array, the Internet of Things will stretch data center resources to the limit. Find out how to maximize and manage capacity more effectively.
I need a good cup of coffee in the morning to set me up for the day. And that means I need creamer in my fridge every morning. Unfortunately remembering to buy the creamer is not my strong point. With a smart refrigerator, I won’t have to remember.
This humble household appliance has become a somewhat clichéd poster child for the Internet of Things (IoT). Yet the idea for smart devices didn’t start in the family kitchen; it was born in our own data centers. We’ve been sending real-time performance data to management consoles for more than 20 years now.
The data center is full of ‘things’, with many of them playing a fundamental role in the world’s supply chains, financial markets, retail operations and, more recently, the Application Economy.
All these ‘things’ needs to be monitored, measured and managed every second of every day to maximize availability, efficiency and capacity.
A host of intelligent sensors and integrated tools provide this round-the-clock insight, capturing performance data, collating operational metrics and crunching business analytics. The result? A fully connected, fully optimized data center – the first iteration of the IoT.
The data center, however, has become much more than just the inspiration for the IoT; it’s become the instrument.
Every new fridge, heart monitor or app that is connected to the Internet means a stream of new data. With today’s 10 billion connected devices expected to top 26 billion by 2020, according to Gartner, this stream will soon become a tidal wave. And it’s heading in one direction: the data center.
From the server rack and the storage array to the network router and the CRAC unit, the IoT will stretch data center capacity to the limit.
To prevent an endless cycle of investment in new infrastructure resources, capacity planners must optimize every gigabyte of storage, every gigahertz of processing power, every bit of connectivity and every kilowatt of power unit in the operational IT supply chain.
This supply chain doesn’t just rely on continuous capacity but also continuous communication and collaboration between application delivery, infrastructure and facilities teams – many of which still operate within their own silos.
To enable a more holistic approach – and capture the competitive advantage that can come with it – organizations need end-to-end, real-time visibility of capacity at a business, infrastructure and data center layer.
As Gartner vice president and analyst Joe Skorupa said in the “The Impact of the Internet of Things on Data Centers” February 2014 report: “The enormous number of devices – coupled with the sheer volume, velocity and structure of IoT data – creates challenges, particularly in the areas of security, data, storage management, servers and the data center network, as real-time business processes are at stake.
“Data center managers will need to deploy more forward-looking capacity management in these areas to be able to proactively meet the business priorities associated with IoT.”
Getting capacity management right is not just about enabling the IoT; it’s about enabling the application economy.
The data captured by today’s connected devices – whether it’s a smart fridge or a heart monitor – is already shaping tomorrow’s apps. For example, smart heating thermostats can capture activity levels in consumers’ homes along with local weather conditions. This data is then fed through to a mobile app to help people make smarter choices about energy consumption.
To ensure the data center represents an opportunity and not an obstacle for both the Application Economy and the IoT, organizations need to ensure that their capacity planning processes – and the strategies they use – are more app-centric.
Today, capacity planning is executed effectively at an infrastructure and facility layer. We collect, analyze, plan and report on the data captured.
But – are we sharing this insight across the two groups and considering its impact on application delivery?
Probably not as quickly or as thoroughly as we need to.
Business and application demand (see graphic above) should drive and shape the capacity planning strategies of IT and the data center. These requirements should reach the IT supply chain in a timely and effective manner to enable the prompt provisioning of data center capacity and the right-sizing of application environment to meet business demand.
Financial organizations are leading this app-centric approach to capacity planning as part of their mobile banking initiatives. They plan for and model the number of concurrent downloads, logins and the millions of transactions occurring within their apps to remove infrastructure bottlenecks and guarantee the end user experience.
To make this happen requires a team effort across the IT supply chain, which means everyone involved needs to speak the same language and aim for the same goals.
I need my smart fridge to be equally in sync with my needs. At 6 a.m., my goal is simple: a hot cup of sweet coffee. Alerting me to depleted creamer supplies is essential. Without that first cup, I am not going to be speaking any language!