Flexible data centre power ‘as a service’ in the world of Cloud?
How ARP brings huge benefits for users and suppliers
“the fundamental benefits of cloud services are undermined by misalignment with the power system infrastructure”
Enterprise user adoption of cloud services is rapidly accelerating in a crowded and highly competitive supply side market.
SaaS, PaaS, IaaS are all about providing flexible on demand services that scale up and down and based on end-user SLA requirements. But is there a missing piece as to what can and should be delivered?
The data centre power infrastructure is static and only one SLA (service level agreement). Of course, this SLA varies from one data centre to another depending on the system topology used. But within the data centre, it’s usual to have a single power system topology. This means the power system does not closely align with end-user power use or their SLA requirements. Consequently, the fundamental benefits of cloud services are undermined by misalignment with the power system infrastructure.
As firms grow used to ‘as a Service’ models so questions for suppliers include how long customers will be willing to pay for power they don’t use. For suppliers trying to maintain margin it begs the question: “Can we continue to spend Capex and Opex on static power infrastructure and operation in the era of flexibility?”
‘As a Service’ defined
If we borrow some definitions from IBM we see that “IaaS is a form of cloud computing that delivers fundamental compute, network, and storage resources to consumers on-demand, over the internet, and on a pay-as-you-go basis. IaaS enables end-users to scale and shrink resources on an as-needed basis…
IaaS providers manage large data centres, typically around the world, that contain the physical machines required to power the various layers of abstraction… In most IaaS models, end users do not interact directly with the physical infrastructure, but it is provided as a service to them.”
Whichever service is chosen, to a user, all forms of IaaS abstract away the physical while allowing them total flexibility to access IT capacity. Cloud flexibility is a fundamental benefit that users now expect.
Cloud service customers want to be charged for what they use and expect that the supplier will seamlessly accommodate the need to scale up or scale down as required.
A reasonable question to put to colocation and cloud service providers is: “If aaS is pay-per-use, then why is the data centre power price fixed whether it is used up or not?”
ARP flexes services up and costs down
Adaptable Redundant Power (ARP) from i3 Solutions Group is a technology and service which brings service models to power infrastructure.
One of the advantages ARP offers is a staggered approach to infrastructure roll out which responds to end user load and not design criteria.
By treating power as a service right from the construction phase, ARP addresses many flexibility issues surfacing in modern data centre operators.
Data centres are and will continue to be designed as physical buildings which are broken down into halls. So, for example, a 12MW data centre may be arranged into six 2MW data halls.
Traditionally the manner in which infrastructure is deployed, if a contract for all halls exists, is to design, source, pay for and roll in the full 12MW of genset, switchgear, UPS, PDU and ancillary equipment up front.
And then wait for the demand to arrive.
As all data centres start with a low load requirement, rolling in traditional modules is expensive and inefficient.
ARP modules are flexible enough to serve power across multiple halls. So, for example, 2MW of infrastructure in Hall 1 can be used to power Halls 1, 2 and 3 at partial load.
And with 2MW in Hall 6 doing the same for Halls 4, 5and 6 as demand increases a third ARP module can be deployed in Hall 3, and then Hall 4 and so on until all halls are powered.
Typically, at the early stages where up to two halls are occupied, the number of ARP modules needed to meet demand matches that of traditional modular power systems. Once three or more data halls are occupied the number of ARP modules needed is fewer than traditional modules.
In this 12MW scenario, as the number of halls in use reaches three and above and utilisation rates across all halls in use rises from around 50%, so deferred equipment costs through using fewer modules reaches 35% of overall power cost. Even at utilisation rates above 50% and rising to 70%, deferred cost savings remain high – averaging around 20% of total power cost.
Any initial premium invested for ARP modules is rapidly offset by deferred cost savings as the gap between the number of modules required expands.
With all halls occupied, once utilisation gets above 40%, so 35% end state savings are seen as a percentage of overall power costs.
Power provision ramps up as data halls become occupied and IT gets to work. That much is obvious. But because ARP matches the entire power infrastructure capacity (including existing redundant power) to that actually used by compute, storage and network this brings huge savings and provides flexibility to customers.
In the real world, not everything is linear. No data centre operator, whether owner occupied, commercial, or cloud, opens one hall, fills it to maximum utilisation and then moves onto the next one. It is simply not feasible.
In fact, for a variety of reasons familiar to all data centre operators, utilisation levels in different data halls will vary. Yet traditional design and equipment has deemed that each hall can only access the power infrastructure with which it is associated. ARP’s different operating modes do different complementary things: Adaptable Redundancy, Inherent Redundancy, Adaptable Inherent Redundancy and IT Load Prioritisation.
Predetermined redundancy levels are provisioned to IT loads where the power is derived from unused power capacity. And for availability, ARP IT Load prioritisation can service different hierarchies of application needs in the event of an outage.
ARP makes power elastic
CIOs are desperate to tap into flexibility offered by cloud. As their experience grows, they will want flexibility in all things. With power measurement and cost control growing in importance they are less accepting that IaaS can only rest on inelastic power infrastructure.
They are more likely to engage with a provider who can answer the reasonable question: “How can I get the transparency needed to inform my ‘as a service’ business decisions for pay as you go for IaaS, PaaS and SaaS when power provision and costing is fixed and inflexible?”
In the next article we will discuss how ARP matches power system SLAs to align with dynamic provisioning applications and container orchestration systems.