Optimizing the Cloud: 4 Next-Gen Considerations You Need to Know10 min read
The current data center model has truly become the home of all modern technologies. The proliferation of cloud computing, IT consumerization, and new ways to deliver content all impact how the data center is designed and deployed. In fact, today’s organization is actively creating their business strategy around the capabilities of their IT and data center environment.
As more organizations move to some type of cloud model – there will be new challenges to overcome. As with any new type of infrastructure, requirements around existing components must evolve. Cloud computing brings a number of new types of benefits to the organization:
- Dynamic content distribution
- Ease-of-access for end-users
- True device agnosticism and mobility
- The ability to deliver next-gen workloads based on a data-on-demand generation
The bottom line is this – There’s no escaping cloud computing and everything that it can bring. Multi-tenant, high-density data center models are allowing for a greater number of users, workloads and data points to reside in an ever-distributed infrastructure. With that in mind, how are you enabling your data center to support these emerging trends? It’s one thing to incorporate converged infrastructure and intelligent logical controls – But what are you doing under the hood? How are you optimizing your existing platform to support these new powerful systems?
Data center environmental controls play a huge part in the optimization process. Cloud platforms and heavy workloads require additional considerations around cooling and power management. The latest Uptime Institute survey showed that the average PUE improved from 2.5 in 2007 to 1.89 in 2011 and to 1.65 in this year’s survey. Still, the majority of data centers are not operating near the upper limit of server inlet air temperatures recommended by the American Society of Heating, Refrigeration, and Air-Conditioning Engineers (ASHRAE) of 80.6 degrees. According to the survey, nearly half of all data centers reported operating at 71 to 75 degrees Fahrenheit.
Based on this – it’s clear that there is some serious room for better air and cooling controls. In fact, any infrastructure housing critical cloud components and data workloads must look at new types of cooling and control technologies to make their data center even more efficient. New data center platforms help create a powerful, redundant, and optimized cooling and control system. So – when you’re designing a data center ready for the modern cloud, consider the following:
- Design your cloud and data center with good management and monitoring. You’ve no doubt heard this before; but you can’t manage what you can’t see. Cloud computing and the modern data center are designed around greater levels of multi-tenancy and workload delivery. This means your servers are hosting more people and doing more operations. Throughout all of this it’s important to not only have DCIM systems running within your data center; you must be able to use these technologies to respond quickly to data center demands. Whether through direct alarm notifications or a platform which provides data trending information – monitoring and management are the keys to creating a healthy cloud and data center ecosystem.
- Deploy optimization where ever you can (cooling, power, space, airflow). Your data center must be able to breathe easy in the age of the cloud. This means deploying infrastructure optimizations to support new data center technologies like convergence, flash technologies, and a lot more virtualization. Make sure your battery systems are set up properly and that power consumption is designed appropriately. Furthermore, make sure you consider floor, rack, and even aisle airflow requirements. All of this translates to a healthier data center system which is capable of supporting more business requirements.
- Conduct regular evaluations and assessments. Have you checked your latest capacity numbers? Are you running out of space? Have you examined your power, cooling, and even space requirements? Too often we get into the daily duties of running a data center and forget to examine the big picture. Remember, just because your data center is “operating,” doesn’t mean it’s operating efficiently. To keep up with cloud – your data center must be able to run as efficiently as possible. Furthermore, data center operations are now even more tightly coupled with business strategies. This means that as requirements ramp up, the need for greater levels of efficiency and cost controls will be prevalent.
- Design around your users (and your business) plan for the future. Your users are the consumers of data center and cloud services. This means a sluggish data center will impact productivity and user experiences. All of this translates to a negative impact on the overall business. When creating data center efficiency – work with your apps, desktop, and infrastructure teams to align strategies. Do you need more focus on VDI cabinets? Maybe there is a big data farm which requires additional power. The point is that your workloads are directly tied to data center operations. IT and data center teams must work together to create plans around deployment, resource control, and how to better support the overall business.
As your data center model evolves – the control layer of your infrastructure must be reviewed as well. This may mean better cooling and environment control technologies. Remember, a big part of cloud is the distributed nature of the entire platform. So, in creating your new cloud-ready data center model – make sure to take management, visibility, and optimization into consideration. By improving your data center’s environment control layer – you’ll be able to deploy a better next-generation infrastructure.
CTO, MTM Technologies
Airflow Management Awareness Month
Free Informative webinars every Tuesday in June.