Insight and analysis on the data center space from industry thought leaders.
How a Robot Can Simplify Data Center Management
With increased reliance on data centers, full software and hardware robotics automation is no longer a question of if, but a matter of when. Srinivasa Vivek of EMC offers a look at an inexpensive robot that automatically maps and navigates a data center, collects temperature, humidity and other data.
August 26, 2013
Srinivasa Vivek is a consultant at EMC India Center of Excellence.
vivek_tn
SRINIVASA VIVEKEMC
With all the latest advancements in the cloud and virtualization markets, who needs a power-draining, expensive data center? Unfortunately, more and more organizations are still continuing to run with this overhead. In new data center designs, capacity provisioning for ever-higher power requirements has always been an area of concern, with a lingering question on whether conventional room-conditioning systems can manage future information technology loads. Within existing data centers, computing capacity typically increases over time as IT requirements increase, resulting in increased power and cooling requirements. Data center operators are challenged to provide adequate support infrastructure that is provisioned, or adapt accordingly, to achieve future IT mission requirements while minimizing energy use.
With increased reliance on data centers, full software and hardware robotics automation is no longer a question of "if," but a matter of "when." Simple RFID tags, laser and barcode identifiers can create true data center automation. Data center, automation and robotics technologies have come a very long way over the past decade. From the warehousing or inventory perspective, robots are equipped to sense location, put the assets or inventory in order, and can directly interact with the human-created automation scenarios.
Using Robots to Improve Efficiency
As the demand for IT computing and network devices continues to increase, there is a growing need to remove heat generated by the equipment, effectively and efficiently. To improve efficiency, it makes sense to locate cooling (such as computer room air conditioners or computer room air handlers) close to the heat source, or another approach would be to have a removal device, such as a hot air chimney, close to the heat source.
Now, with techniques in place, are we really sure that the amount of cooling given per rack or equipment is just right? A typical data center should run anywhere between 18 degrees C to 27 degrees C as per ASHRAE (American Society of Heating, Refrigerating, and Air-Conditioning Engineers). The lower the temperature, higher will be the energy consumption and higher the temperature resulting in a possible problem of equipment availability. This raises the risk of temperature fluctuation inducing equipment failures. Thus, an accurate responsive data center temperature monitoring and control has become increasingly important. Running the data center at optimum temperature is thus a challenging technical effort.
Some of the earliest work on data center energy monitoring and management was based on sensors installed at fixed locations. Alternatively, another method used was a hand-pushed mobile sensing station that accepts human input to record data center layout while automatically recording a set of temperature and humidity readings.
The quest for an alternative to these approaches led to the development of an inexpensive robot that automatically maps and navigates a data center, collects temperature, humidity and other data, and feeds such data into a software tool for efficiency analysis and data center energy management. A set of engineers from EMC brainstormed and came up with the idea to build a low-cost robotic platform to monitor environmental parameters in a data center. The prototype or DC Robot as we call it (see below) consists of three sensors mounted to a vertical tube as per ASHRAE recommendations on measurement points, and is also equipped with remote navigation capabilities through cameras and remote control software. This DC Robot collects temperature data using three digital sensors and relays it through a Wi-Fi access point for post-processing. An algorithm converts the temperature data into a thermal map. The thermal map can then be used to easily identify the hot and cold spots of the data center aisles with location information.
DCrobot-EMC
Fixing a simple cooling leak saves a lot of energy. Small adjustments in the data center's cooling temperature can also provide large energy savings. Poor control and monitoring of the conditions in a data center may shorten the life of the equipment, overheating can cause intermittent faults, and in extreme cases cause equipment to fail catastrophically. The cost in time, money and lost business productivity can be considerable.
Having an independent system like the DC Robot check on the air conditioning unit and monitoring devices is a wise idea. It can also help give a clear picture of the conditions in different parts of the data center and coupled with alerts such a system can give ample warning before conditions become critical or out of control. Robots become the extension of data center staff and managers and assist in keeping the data center environment running smoothly.
Please note: Co-author of this article is Arun AT, Associate Consultant II at EMC India Center of Excellence.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.
About the Author
You May Also Like