New Tips to Improve Energy Management
Ways software can be used to improve overall data center energy use.
Arthur Cole spoke with Henrik Leerberg, product line director of enterprise software for Schneider Electric, about data center power conservation and what enterprises can do to improve data and power synchronicity.
At most data centers, going "green" means deploying the latest virtual infrastructure and replacing old hardware with new, lower-power devices. However, some of the biggest energy savings come from more closely integrating the management of actual power and cooling systems with the data loads and infrastructure they support. APC by Schneider Electric has pushed this concept a little further with the new InfraStruxure suite the integrates with VMware vSphere. The company's Henrik Leerberg explains some of the key aspects of the system.
Cole: Discussions of data center power conservation usually revolve around virtualization, dynamic load balancing and other changes to data environments. What are some of the ways management of physical heating and cooling infrastructure can be folded into these new data paradigms?
Leerberg: Moving compute loads around alters the power and cooling picture significantly and may lead to lower utilization of power within the data center. However, by enabling managers of the physical layer to gain insight and transparency into where FX cooling is being consumed in real time and thereby distributing the capacity to where it is needed, power usage will be optimized significantly.
Cole: The new InfraStruxure suite offers "bi-directional" integration with vSphere. How does that differ from standard integration, and how does it help virtual machines manage power consumption?
Leerberg: Migrating virtual loads in and out of data centers or racks or just between host servers, the power and cooling keeping these servers in production are crucial but are often overlooked or altogether forgotten. In reality, if the power and cooling infrastructures are interrupted or otherwise compromised they can easily run out of capacity when most needed leading to broken service-level agreements. Power and cooling should follow the load, and because this is not practical in real time, the virtual environments must ensure that required capacity is available in the racks to which loads are being migrated. By connecting the physical layer management and virtualization software, we can inform the IT manager where to place virtual loads or where not to place them. Helping to migrate loads away from failure ensures loads are always placed in the proper location and provide the highest availability at all times.
Cole: What would you say are the biggest mistakes when it comes to managing energy use in the enterprise?
Leerberg: One of the biggest mistakes is over-provisioning power and cooling. Years back this was the only way to apply relatively big safety margins into the capacity. Back then, advanced software for power and cooling management and optimization did not exist, but introducing such products into the daily operation of data centers has significant impacts including energy savings and increased efficiency.
Another mistake is supplying the same level of capacity in all areas. Power and cooling needs to go hand-in-hand, and by "zoning" the data centers with high density and lower density zones, power and cooling can be provided to these zones more effectively, making the infrastructure much more efficient. This can be obtained by using UPSes that match the power needed and by placing InRow cooling units as close as possible to the heat being produced by the compute load.
Finally, not implementing a proper data center infrastructure management (DCIM) software to monitor, plan and report on what is happening in real time makes the capital investment in the data center less attractive. You lack the ability to adjust the operation in due time to meet new requirements, react to changes and provide the best service for your business.