Seattle Data Center Insiders Offer Tips to Reduce Operation Costs
At CApRE’s recent Seattle Data Center Summit, we convened a panel of international experts in data center operations and management to offer strategies for reducing cost without increasing risk. To kick off the discussion, we asked the panel — What are some examples of things your firm is doing to reduce costs within your data center? Do you see major differences between new and old data centers?
Cindy Joos, Vice President – Global Shared Services, Cyxtera Technologies. “From the Cyxtera side of the house, we’ve looked at our maintenance standards, obviously,” she replied. “With old data centers, we believe you have to keep your maintenance up, but you don’t want to over-maintain it. So we have standardized across all of the different environments. Obviously, we have a lot of different environments coming in through acquisitions. But for the same or similar environments, we make sure that everyone is doing the same kind of maintenance and not doing too much. So we keep up with it and we stabilize with it.
Next was Rick Noji, Director of Business Operations, Green House Data. “One of the things we are looking at is what maintenance we have to do at night, and what maintenance we can get away with during the day without any of our customers affected,” he offered. “And if we do end up doing our maintenance during the day, we’ll of course notify customers, but I think that has made a big difference, by cutting our costs. Before everything was just pushed through maintenance automatically at night, but we took a look and asked, do we really need this at night? It saved quite a bit for us.”
Ajay Garg, Director of Data Center Management Solutions, Intel, pitched in next. “I can’t speak from the end user perspective, but we work with many data centers. A lot of times what happens is that a data center, from the facilities perspective, might be running in a totally optimized way, but often times what we find is that the IT devices there are just up and running and no one knows what’s going on,” he recalled. “Part of the reason is that no one is actively monitoring it. They say “Oh it’s just some application developer’s job to figure out what’s necessary or not” and to me, one of the best cost-saving elements that can come is if these two limbs work together. To figure out how to drive overall utilization of the data center. Not just from the facilities perspective, but also the IT perspective.”
Gustav Bergquist, COO and Executive Vice President, Multigrid offered his thoughts. “From my perspective, there is plenty of work to be done in terms of how to operate the infrastructure itself,” he said. “In the last 15 years, data cneters have been fortifying themselves, right? We’ve been discussing going off-grid and we’re relying on no one else. But from our perspective, being connected provides operational cost savings that are very large. In my context, we’re selling the heat from our facilities. We’re getting paid to deliver heat to the district heating systems. Secondly, on the electric side, there are plenty of opportunities to be exploited in terms of your UPS for containment reserves.”