Steps for Protecting Your Data Center After Virtualization
Shannon Sbar, VP of Channel at APC by Schneider Electric, has 14 years of channel experience and in her current role is responsible for driving revenue, strategy and profitability. She holds a Bachelors of Science degree from Florida State University.SHANNON SBAR
APC by Schneider Electric
Congratulations, you’re virtualized. You may be asking, “Now what?” Let’s examine the steps for protecting your data center after virtualization. As the consumers of around 2 percent of our nation’s electrical supply – and growing – there has been an increasing focus on data centers to become more energy efficient. Simultaneously, however, companies and IT providers are facing an unprecedented surge in demand for services as the nation’s dependence on technology grows. This has led many to look at solutions such as hybrid cloud deployments that reduce electrical load and significantly drive energy savings, without increasing the complexity of a data center’s infrastructure management (DCIM) strategy. But while the benefits of such deployments are substantial, many managers focusing on off-site deployments tend to meet an unexpected challenge post-virtualization: protecting their remaining physical equipment.
Threats to Physical Function & Business Continuity
Consolidation and protection strategies for remaining on-site equipment tend to be overlooked after an off-site deployment takes place, but power outages, physical and environmental threats can be detrimental to not only the physical function of the data center, but also to business continuity. However, sometimes it can often be difficult and confusing determining where to start. The following checklist outlines the top three considerations for business managers when developing a strategy for protecting their physical IT equipment post virtualization.
Top Considerations for Protecting Physical IT Gear
1. Power Maintenance
Uninterruptible Power Supply (UPS) systems provide run-time when utility power fails, keeping critical computing and connection equipment operational. To maintain optimal power availability and greatly decrease the risk of unnecessary downtime, redundant UPS systems are recommended for critical dual-corded gear, such as servers and domain controllers. IT managers should ensure that redundant power cords are plugged into a separate UPS or rack PDU (power distribution unit). Additionally, UPS systems with network management cards can also be crucial to maintaining power, as they allow remote monitoring of critical power conditions.
PDUs protect critical loads by removing points of power failure and allowing several devices to be powered by a single source. There are two basic power distribution methods: plugging IT gear into the receptacles on the back of the UPS, and plugging IT gear into a rack PDU that is plugged into the UPS (this method requires that IT gear be mounted in a rack). Working in tandem, PDUs and UPSs provide uninterrupted power for the longevity of the equipment, providing maximum availability.
2. Temperature Regulation
To determine the appropriate cooling solution, IT administrators and data center managers must first determine ideal temperature settings for the environment. The final design must consider all variables that affect cooling. ASHRAE TC 9.9 recommends operating temperatures fall within the range of 68-77 degrees Fahrenheit, with the allowable range being between 59-90 degrees. Outside of building HVAC systems, solutions such as in row cooling units provide supplemental air conditioning and consume minimal floor space.
Additional careful consideration must be given to closet environments where a UPS is deployed. Increases in temperature have a much more pronounced effect on battery longevity than other types of IT equipment, with higher than normal operating temperatures cutting battery life nearly in half (from three to five years, to just a year and a half). If optimal temperatures cannot be reached, IT administrators should consider placing UPSs in air conditioned spaces outside of the central IT environment.
Organizing IT equipment in a rack can also ensure reliable cooling, increased airflow, prevent thermal shutdown and reduce the need to “over-cool” the space.
3. Equipment Organization
Overall, the more organized the IT environment, the easier it is to cool the equipment by separating hot and cold air streams. A good practice for reliable cooling and maintenance is to consolidate equipment into one easy-to-manage rack enclosure. This will guarantee ultimate availability by ensuring that critical servers not only receive clean power, but are also operating at the right environmental conditions. Additionally, racks help prevent thermal shutdown events and reduce the need to over-cool the space with over-sized air conditioners. Rack enclosures also help reduce human error associated with troubleshooting via cable management while racks with environmental sensors provide increased physical security, allowing data center managers to respond to environmental and physical threats when they occur.
Going a step further, it is important for data center managers to consider the speed of building and implementing the physical equipment. Solutions which arrive onsite fully configured, or that can be assembled easily, can aid in cutting down the time before full deployment, while also providing additional protection from environmental and physical threats. Products such as rack fan trays, which can install easily, eliminate hotspots and help prevent infrastructure from overheating while a step down transformer avoids inrush current that may cause damage to fuses and circuit breakers.
As virtualization becomes more prevalent in the marketplace, businesses are moving from using one server for each of their applications to virtualizing these applications and migrating them to fewer machines. This results in a more efficient computing stack, but also more risk now that there are effectively more “eggs in one basket.” However, there are simple steps every data center manager can take to better ensure the physical safety of their systems.
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.
Alex PisaniPosted February 15th, 2013
Great insight; how do you recommend balancing the risk of the “more eggs in one basket” vs the efficiency of combining virtualizing applications?