What Are Some Of The Pitfalls Of DCIM?
In the past data center infrastructure management (DCIM) had been seen as hardware issue. With more insight, it’s been determined that the real challenge is not with the hardware but with how all of the data is monitored, managed and analyzed. The main issue is the connectivity between all of the systems in the center.
Since the average data center might have up to 10,000 data points, which results in stranded data. Whether the operator is on the facility, infrastructure or IT side, the data points are not only between them but isolated in each category.
Disjointed Data in Data Center Infrastructure Management
Today’s modern data management centers include a multitude of systems like generator control systems, electrical power control systems, uninterruptable power systems and the building management system. However, their systems do not communicate with each other, and operators can’t view the data in a meaningful, connected way. Often, these systems are also isolated from the operator’s company network and cannot be accessed remotely.
The bottom line is that these systems collect a mass of data, yet the data is stranded. A good DCIM system need to provide the whole picture of data, servers, racks, mechanical, electrical and plumbing systems.
No Established Hierarchy
The solution is to design a comprehensive DCIM system that recognizes the breadth and diversity of the information and one that can arrange the data points in a hierarchy. There should also be multiple hierarchies across all systems. Once the structure has been built around a hierarchy, a repeatable system can be developed.
While a great deal of intelligence needs to be imposed on all of the systems in the data center to create a dynamic infrastructure that can fine tune itself to match data center resource supply with workload demand, success is dependent on having the data arranged hierarchically.
Inefficient Data Center Cooling with Data Center Infrastructure Management
As the densities of IT equipment increase, underfloor main distribution blocks or racks is not enough for efficient cooling and power costs. DCIM needs to be combined with building information modeling (BIM). A BIM system ensures that the computer room air conditioning is correctly positioned to provide the rights flows to equipment that has thermal limits. The integration of DCIM and BIM will allow companies to optimize energy usage and use IT more efficiently.
As the size and number of data centers continue to grow, they must be evaluated, re-imagined and innovated. There is no excuse for data centers to have stranded data and be energy draining. Data center operators need to educate staff about changes in the data center’s building blocks and the core. Software infrastructures and hardware infrastructures all need to work synergistically for efficiency and while surface-level fixes are good, there is much for needed.
Data centers need to be based on open architecture principles and more intuitive to manage. Taking all of these factors into consideration, solutions are than proactive and preventative.
Since data center houses are critical to the daily operations of today’s companies, it’s also important for companies to stop using outdated methods. Refusal to adopt innovation in data centers is shortsighted and leads to lost money and time.
Katrina in addition to her work with rack solutions Katrina has a video series featured on Youtube called ask Katrina that aids in answering your IT needs.
Comments are closed.