Most new datacenters operate at optimal availability and with infrastructural energy efficiency close to theoretical design targets. As such, it might be argued that the two biggest challenges of datacenter technology in the past 30 years have been addressed.
But despite this progress, the pace of change in the datacenter industry will continue and is likely to accelerate over the next decade and beyond. This will be spurred by increasing demand for digital services, as well as the need to embrace new technologies and innovation while mitigating future disruption. At the same time, there will also be a requirement to meet increasingly stringent business parameters and service levels.
This combination of business and technology drivers is likely to result in the emergence of new classes of datacenter facili- ties. These emergent datacenter types will share some traits with existing facilities but will also be tailored to new use cases. For example, it is expected that new edge datacenter capacity will be required to aggregate, process, store and analyze data from Internet of Things (IoT) infrastructure.
This paper examines these forces of change and disruption over the next decade, makes predictions about new datacenter types and specific use cases, and finally suggests ways to future-proof existing datacenters against disruption and to capi- talize on innovation.
There are currently more than four million datacenters worldwide, according to 451 Research’s Datacenter Monitor. These range from small enterprise-owned server closets and rooms (the vast majority of sites today) to commercial multi-tenant datacenters (MTDCs) and large hyperscale sites. Some of the dynamic forces of change in the industry are already having an impact on the distribution and relative numbers of these different facility types. Approximately 80% of all datacenter space (by square footage) globally was owned by enterprises in 2015. However, that percentage is expected to drop below 75% by 2020. The main reason for this shift is the migration of certain workloads from enterprise-owned sites into typically more cost-efficient cloud and colocation facilities.
The expectation is that the shift toward hybrid IT models will continue, causing enterprises to consolidate sites and move into larger, yet fewer, facilities. Cloud and colocation space will be increasingly viewed by enterprises as an extension (or replacement) of their own on-premises capacity. Public cloud and other cloud service providers are also driving significant demand for wholesale colocation space. It is expected that this migration from generalist enterprise sites to highly efficient colocation and cloud facilities will continue, and accelerate. However, there will still be a requirement for dedicated, premi- um enterprise sites among some organizations. New edge micro-datacenters will also to some extent replace‘legacy edge’ enterprise server closets and rooms, as well as supporting new use cases.
UNDER LYING FORCES AND TECHNOLOGY TRENDSChanges in datacenter design and operation are to some extent shaped by a number of underlying forces. These forces are in turn driving specific emerging technology trends, some of which are already beginning to have an impact on the construction and operation of new facilities. So-called ‘wild cards’ – theoretically very disruptive and hard-to-predict tech- nologies such as post-silicon technologies and quantum computing – could also result in major changes in design and op- eration. Acting in conjunction, these forces and specific technologies are expected to result in new classes of facilities over the next decade that exploit innovative technologies but also increasingly meet specific business requirements.
FORCES OF CHANGEThe broader forces of change in the datacenter industry are closely coupled with a number of specific tech- nologies and trends that are reshaping the design and operation of new datacenter capacity.
Accurately predicting the evolution of physical datacenter design and operation over the long term is obviously challeng- ing. The number and types of datacenter form factors have changed significantly since the early days of the mainframe. We expect that pace of change to continue into the foreseeable future, with a move away from generalist, inefficient (usually) enterprise-owned sites.
It is possible to identify a number of future datacenter types (some of which exist today) that are likely to dominate in the next decade and beyond. Future datacenter types could include, but not be limited to, the following:
These various datacenter types will be defined by some of the following criteria and attributes:
Business Models
Business models will vary between datacenter types based on ownership. For example, colocation and service-provider sites will be required to offer high availability and, very often, low-latency services – frequently achieved through proximity and connectivity. Similarly, some enterprises will have specific workloads, data requirements or governance issues that dictate that they must continue to design and operate their own premium datacenters. Given this, these organizations then have the opportunity to innovate and customize in ways that may not be open to commercial (colo, hosting) operators.
Scale
There is every indication that the efficiencies of scale that existing hyperscale sites enjoy mean that despite advances in compute capacity and more workloads moving to the edge, there will be a continued and sustained need for hyperscales in the future. At the other end of the spectrum, small-scale micro-sites are expected to become more widespread to support IoT and other applications. However, there will continue to be a consolidation of server rooms and closets (legacy edge) into colocation, cloud and, in some cases, micro-datacenters.
Resiliency
Resiliency requirement will be even more closely related to business case and function. For example, hyperscales may have lighter physical infrastructure (reduced UPS, lower-tier designs) because of lower service levels and ability to manage avail- ability by load balancing. Some MTDCs may also build to different resiliency levels within the same facility, depending on customer requirements. There will also be greater adoption of software-based, distributed resiliency with less reliance on physical infrastructure (generators, UPS).
Efficiency
Efficiency will continue to be a requirement across the board. Some facilities, such as hyperscale sites, will heavily prioritize efficiency – in some cases, above most other criteria. A percentage of hyperscale sites will also continue to focus on sustain- ability and carbon reduction by utilizing more renewable energy (through power purchase agreements, renewable tariffs or, in some cases, on-site generation).
IT Density
Average rack power density, currently less than 5kW, is likely to continue increasing over time, driven by applications such as AI/machine learning, high-performance computing and big data. However, some wild-card technologies (e.g., quantum computing) have the potential to increase compute capacity while significantly reducing power requirements. Density will increasingly be tied to business function and workload. High-density zones may be created to enable more efficient cooling and power distribution. HPC and other specialist sites are likely to have high-density IT equipment (>25kW per rack) and consume more energy per unit of space/rack, for example. This means the cooling is also likely to be close-coupled: i.e., targeted to the requirements of a relatively small number of high-density racks.
Geography and Distribution
A number of large cloud operators have built out in specific locations (e.g., Europe) or have leased from MTDC providers, partly in order to comply with data regulations. This trend is likely to continue for future datacenter types. Hyperscale sites will also continue to be built in areas with low energy costs, tax incentives and climates that allow for free-air cooling. Edge capacity will be added in centralized datacenters as well as in metro sites outside of core datacenter hubs (see Figure 3).
Of all the trends shaping future datacenter development, the demand for edge computing is expected to be one of the more significant and, as such, deserves specific attention. Edge computing can be described as the distribution of compute and storage capabilities to the very edge of the network near the point of data generation and data use. This could be an enterprise factory floor or a carrier point of presence, a cell tower or a smart building.
While use cases such as content distribution networks (CDNs) and local processing and storage are expected to drive edge compute demand in the short term, the Internet of Things (IoT) is expected to be one of the long-term drivers for new edge capacity. IoT’s uses cases are vast but even within similar use cases, data paths and datacenter types will vary. In our view, it does seem likely that a number of IoT deployments will end up having data residing in a combination of public cloud and non-public cloud facilities, with the need for both distributed micro-datacenters and very large centralized sites.
Key data, including data needed by other applications and people, will in some cases be made available at the ‘near edge’ in large datacenters where colocation and other metro datacenters are sited close to where the data is generated. Cloud heavyweights are rapidly building hyperscale datacenters with direct fiber links to leased colocation sites. This brings hy- perscale cloud capacity closer to the edge – effectively functioning as ‘near edge’ datacenter capacity. Cloud providers will also utilize ‘cloudlets’ – distributed edge capacity for data caching or low-latency compute.
Once consumed or integrated, data will then typically be moved or streamed into large or hyperscale remote datacenters to be aggregated, analyzed (including through integration with other data and applications) and archived. These large facilities represent the ‘core layer.’
The broad schema in Figure 3 illustrates the main layers of IoT and edge datacenters, with some of the different types of datacenters and data paths that IoT applications might require.
IT hardware in a datacenter is typically refreshed in a three- to four-year time-frame, but most facilities infrastructure is con- siderably harder, and more costly, to update or retrofit. It is standard practice to overprovision cooling and power overhead to allow for future IT capacity requirements (balanced against efficiency). However, this could be described as future-proof- ing for changes in load rather than for new physical technologies that might render aspects of the design obsolete. There are, however emerging and established strategies and technologies (some of the same that are shaping future datacenter types) that can help maximize efficiency and capacity management and also reduce the risk of obsolescence. These include:
Datacenter operators are facing a time of unprecedented change. In 10 years, it is likely that the datacenter landscape will look very different as it responds to the macro forces and technology trends previously described:
During the next decade of unprecedented change and new disruptive technologies, it is clear that the datacenter – in all its new forms, shapes and roles – will continue to enable innovation and drive technological and business transformation.
Raritan, a brand of Legrand, is a trusted provider of rack power distribution units, branch circuit monitors, transfer switches, environmental sensors, KVM-over-IP switches, serial console servers, and A/V solutions for data centers and IT professionals. Established in 1985 and based in Somerset, N.J., Raritan has offices worldwide serving customers in 76 countries. In more than 50,000 locations, Raritan’s award-winning hardware solutions help small, midsize, enterprise, and colocation data centers to increase efficiency, improve reliability, and raise productivity. And provide IT departments with secure, reliable remote access tools needed to manage mission-critical environments. For more information, visit us at Raritan.com.
Download the Full Illustrated Report Now