How do I get capacity as fast as possible–where I need it, when I need it, and at scale? This is the question that many tech giants are asking right now. Currently, there are four “demand” trends driving this capacity challenge: demand is huge and rising fast, demand is unpredictable, demand is emerging in new markets, and demand is variable.
Demand is huge and rising fast
Every day, 2.5 Quintillion bytes of data are created, and the digital universe grows by 40% every year. If we look at these numbers and project outwards using the best research and data possible, we can conjecture that by the year 2021:
- There will be so much video distributed across global IP networks each month that it would take 5 million years (!) to watch it all.
- There will have been a 20x increase in VR and AR traffic between now and 2021.
- 22.5 billion always-connected IoT devices will be online (up from 6.6 billion in 2016).
- We will need 400 million servers in 2021 available to support 2020 IoT and technology demands.
- And, we will need 4000 ~400,000 square foot 50-megawatt data centers to support those 400 million servers.
Clearly, there is a tsunami of data coming. But there are more challenges ahead.
Demand is unpredictable
The fast-growing demand trends we looked at makes planning for compute, storage and network capacity difficult for even the best capacity planners. For example, when Pokemon GO was released in 2016, it achieved 50 times greater than its originally anticipated growth, bringing servers to a grinding halt until the company could adapt. Companies and the data centers they use have to be ready for both the predictable and unpredictable.
Joe Cava, VP of Google Data Centers, says: “When you build data centers for your own products, you’re able to stay ahead of demand. You have historical data you can work with to make projections. But we’re no longer building for just our own products. With the public cloud, you’re building for everyone else’s products.”
Demand is emerging in new markets
Right now, we’re seeing rapid growth in emerging markets with no signs of slowing. The fastest growth right now is taking place in Europe, Middle East, and Africa, as nations in the Eastern part of Europe and large swathes of the latter two areas undergo rapid development. Video is a significant driver for traffic spikes, while IoT pushes for capacity at the “Edge”–the area where faster networking and processing can happen between the Cloud and server networks. By 2019, nearly half of IoT data will be processed at the edge of the Cloud, providing a serious challenge for developers. What’s more, content delivery networks will be carrying up to 71% of Internet traffic by 2021–up from 52% in 2016.
Demand is variable
Compute loads are becoming more and more dynamic as demand varies from month to month, day to day, hour to hour and project to project. Pertinent examples include how eCommerce spikes during the holidays, how data demand spikes “follow the sun” across the globe, and the ramp ups and downs that take place when DevOps are testing new ideas, products, and services.
Capacity Challenges Ahead
So, what are the some of the capacity challenges ahead for companies and data centers who want to not only meet the demands of the near future but stay ahead of the curve? First, these companies must focus on the near term to deliver and they must focus on the long-term to meet the demands of the future.
Near-term tactical challenges include mitigating the risk of not being able to support the business–delivering capacity as quickly as possible to ensure the business can meet customers’ needs.
A few of the long-term strategic challenges arising are:
- Future-proofing current data center deployments against changing IT requirements as they arise during the term of a lease
- Supporting standard and high power densities without stranding capacity or space or having to reconfigure the data center footprint
- Deploying capacity at any scale 500 kW to 50 MW — around the world without reliability issues or power, water, or human capital restraints.
Capacity demands are huge and unpredictable. They’re emerging in new markets, and they’re increasingly variable. Adaptive data centers–like Aligned’s–allow you to respond to both the near-term tactical and long-term strategic challenges so you can deliver for capacity quickly while mitigating risks today and future-proofing for tomorrow.
KW: data center, capacity challenge,
Meta: Aligned is working with its clients to solve the rising data center capacity challenges of today and tomorrow.