Making the most of DCIM

Many data centre owners and operators are missing a trick by not integrating their physical M&E infrastructure with the technology infrastructure to centralise monitoring, management and intelligent capacity planning of a data centre’s critical systems. DCIM can now help achieve complete visibility over all assets to support faster and more informed decision-making, enabling operating efficiencies to be increased and energy consumption reduced.

Data centre market and DCIM
There is no question that the demand for Data Centres is growing and will continue to grow in both established markets as well as emerging countries and regions. Moreover, with this expansion we are not only seeing the spread of the large US technology and cloud companies, but also an overall increase in uptake in ‘The Cloud’.

Whether you believe in the true cloud or just simply outsourcing to a service provider, the mix and range of services along with the competitive nature of the industry is becoming greater. This makes for a much different and much more cost effective and flexible way to deliver IT and data centre services than even say two or three years ago. That said the current data centre market is made up largely of legacy facilities with much fewer large-scale enterprise facilities and virtually no speculative builds from developers or operators. Over time this shift from ‘on premise’ will increase with SMEs and many enterprise customers outsourcing data centres to a service provider or into the cloud.

With these options combined with the current economic and business drivers to reduce capex and consider total cost of ownership (TCO) the need for real time information to enable quick informed decision making and to maximise deployed capacity and assets has never been so great. It is fair to say that many new facilities are adopting best practice when considering efficient technologies and are implementing the latest data centre tools such as DCIM, and there are some big advantages in having this approach from the design phase, but some still are not committing. Moreover, with such a large percentage of legacy data centres there is currently a huge opportunity to work with what many companies already have.

Do we agree what DCIM is?
The DCIM market to some has been understandably a bit confusing over the last few years. Many market analysts have tried to define exactly what it is and predict growth rates and trends, whilst many technology service and solution providers large and small have proclaimed to offer a DCIM solution. This has led to some overall misunderstanding, and even some skepticism, as to the real benefit and value of DCIM. Any solution should cater for all aspects of your critical infrastructure, not just one or two such as efficiency or power usage, and should enable integration with other building and IT systems. Some of the common functionality across full DCIM tools includes asset management, energy monitoring, network usage, change control, airflow, capacity management, predictive modeling, and colocation specific billing and customer segmentation.

Moreover, whilst DCIM can be deployed in a virtual scenario with manually entered data this will only provide limited information for planning that is useful at the design stage and can help with decision making but does not give an accurate reflection of the true state of the live facility. Therefore, to achieve the true value of the solution it has to be integrated into the actual monitored data giving a real time view.

Opting for DCIM
A vast amount of investment has been directed at DCIM, both by large corporate enterprises and smaller private equity backed companies. We are now seeing some level of M&A activity and with the more recent developments a clearer view of the providers and their offerings. One of the key fundamentals is to understand that this is not just another flash in the pan idea, as some had predicted, but instead is a fundament change in the way data centres will be managed and operated in the future.

Therefore, getting the selection right is critical to ensure you can deliver all the required functionality now and in the future, as well as offer a competitive advantage through innovative and new ways of doing things. Any solution has to cater to the end user with their own facilities but also work in colocation and outsourced facilities. This is equally true for professional data centre operators when selling their services to ensure they possess a solution that delivers the information customers will inevitably demand from them.

It is also important to consider when evaluating different tools that it is not just about the software itself. We have seen many organisations commit to a project and purchase software that due to a lack of proper integration or internal management has quickly become obsolete or out of date so is no longer fit for purpose. What provides the true value and delivers the overall success is the implementation and ongoing use of the solution in the day-to-day operation. In fact, understanding how your organisation is structured, what tools and systems are already in place and how DCIM will integrate or replace them is key to any successful implementation and ongoing operation.

The opportunity for DCIM
Through a well-managed DCIM solution organisations can effectively manage all of their critical systems and infrastructure to achieve a range of business and operational benefits. For example, the ability to identify at the deployment stage where equipment should go to best make use of resources such as space, network, cooling and power without limiting yourself in the future. Meanwhile, the decision to build a new facility or upgrade an existing one to achieve additional capacity or improve efficiency can be better taken using added insight, without the need for lengthy site audits and consultant’s fees.

With the right data and knowledge and expertise to understand what critical systems are doing DCIM can deliver much more value and become an enabler to your organisation. However, getting the basic metering and monitoring in place and implementing a solution is simply the first step, with the real opportunities coming from how you use the software and when you start to integrate it with other systems.

Virtualisation is already playing a major part in the way organisations view their compute and storage with workloads moved around between data centres. This combined with cloud computing offers the opportunity to maximise resource utilisation to meet business demand. The integration of DCIM with the IT and BMS gives operators the opportunity to automate the movement of applications and workload dependent upon a range of scenarios. Some examples of this include:
£ Security where an intruder accesses a rack or area of a data centre
without authorisation the software linked to the security system can
automatically move the workload to another physical location in the
same facility or another one.
£ Resilience, at a data centre one of the UPS suffers a failure and
so redundancy is lost. The system can automatically move the
workload to a more stable environment at another facility with
the appropriate levels of redundancy.
£ Efficiency can be achieved through moving workloads around
based on internal or external environmental conditions for more
efficient cooling or even to reduce cost through cheaper power tariffs.
£ Increase revenue through schemes such as demand response
where electricy utility providers offer substantial incentives to
reduce power consumption. This can be better accessed with
the ability to move to another location rather than start up standby
generators.

There are many examples such as these, which leads us to the scenario where believe it or not facilities can be designed with much lower levels of resilience. By removing the need for multiple UPS systems and generators this not only significantly lowers capex but also helps to improve efficiency, overall environmental impact and saves space. With resilience built into the application layer it can deliver a much more robust, secure and flexible platform for delivering services.

Meanwhile, the biggest focus for improving efficiency in the data centre over the past few years has been the cooling. From a design and build point of view this is absolutely the right approach to take, but its worth bearing in mind that the biggest waste of energy is actually at the server level itself with only about 6-12% of power delivered to a server being used to perform a computation. With power capping already part of the DCIM suite, the next step is to match the server capacity to a workload or cluster to better utilise server capacity. This can improve utilisation rates up to 80% and ultimately remove the need for so many servers.

DCIM is certainly here to stay, so organisations will have little choice but to adopt these types of tools if they want to continue to meet the demands of their business moving forward. There are a number of considerations and pitfalls, so choosing the right partner and solution and achieving the necessary stakeholder engagement is fundamental. The technology and therefore opportunities will continue to develop and being able to take advantage will provide clear competitive advantage now and in the future.