PREDICTIVE SIMULATION AND THE ACE PERFORMANCE SCORE – PART 2
Why predictive simulation?
Doesn’t every data center owner/operator (or better yet CFO, CIO, CTO) want an optimized data center?
What is an optimized data center?
Is it one that keeps the lights on as in ensuring all IT equipment is available?
Availability certainly is important and I would venture to say that it is on the forefront of every data center owner-operator’s priorities. After all, one of the best ways to measure success in IT Support or Facilities Management organizations is by how unnoticed they go. Issues tend to bring those organizations into the spotlight. IT Support and Facilities Managers will tell you things are great when everything is quiet. However, one question to ask is what is the cost of assurance of availability? I can talk about the ever-increasing footprint of data centers and the fact that computing power is becoming more and more dense. We can discuss the fact that data centers account for over 1.5% of power consumption in North America and that trend is growing. However, I would rather talk about how we can take one of our most valuable assets – the data center – and maximize its Return-on-Investment.
One question I want everyone to consider is what capacity are you getting out of your data center? When the data center was planned and/or purchased there was a capacity goal in mind. Perhaps 1mW was the target. How do you ensure you can still achieve 1mW? I think it is safe to assume that over the course of data center operations the equipment deployed is different from the initial plan. This leads to fragmentation of the cooling and ultimately capacity. The distribution typically isn’t optimized to ensure it is getting to the right place. After all, it is invisible. So how do you measure what can truly be achieved in terms of capacity? More importantly how do you measure how much capacity can be achieved without sacrificing availability or efficiency (i.e. making sure the air getting to the right place)?
Our partners and friends over at Future Facilities have come up with a straight forward comprehensive way to measure overall data center performance focusing on Capacity goals all the while maintaining Availability and Efficiency. It’s called the ACE Performance Score. ACE is an acronym for Availability, Capacity, and Efficiency. The key to calculating ACE is the utilization of a calibrated Computational Fluid Dynamics (CFD) model of the data center. We’ll get to the ‘how’ later in this blog series. For now, I would just like to ensure the concept of ACE is clear. So what exactly is the definition of each component of ACE?
ACE PERFORMANCE SCORE
The definitions above are fairly straight forward but I encourage you to take a moment and think about them particularly in the context of your own data center. Again, most of the focus is on availability. However, real estate is at a premium. What would it mean to your company if you can get 20% more capacity out of your data center, keep availability at 100% and also ensure efficiency to control your operational spend? Almost every data center in operation today has some stranded capacity. Predictive Simulation can help you reclaim that stranded capacity and ensure you are getting the most out of your data center.So think about ACE-ing your data center. Use the ACE Score to steer the course of maximum capacity all the while ensuring you achieve your availability and efficiency goals. Use ACE to communicate performance with Senior Management and provide meaningful and attainable goals to the data center team.
Hopefully this gives you a fundamental understanding of the ACE Performance Score. We’ll dive in deeper in the next few posts. Topics to be covered include: Establishing ACE Goals, Creation of the Virtual Facility, How to Calculate your Current ACE Score, How to Identify Improvements and Reclaim Lost Capacity, and Continuous Predictive Simulation.
About the author:
Jeff Brickley is a certified Project Manager and has been in the Information Technology sector for over 17 years. He has a unique and diverse background including programming, infrastructure and project management. He has lead large, complex efforts and teams with the common theme throughout his career being that of optimizing operations for business. His passion for continuous improvement and a background in Mathematics made him an ideal candidate to study, utilize and promote the use of Computational Fluid Dynamics modeling for data centers.