The data center industry experienced rapid growth over the last few years in order to support the sudden uptick in demand for cloud services. This resulted in an impressive global network of facilities. However, most of them were designed and built too quickly and are managed to service levels based only on existing systems and processes. If you operate one of these average portfolios, then the quality of information you have available to effectively manage your assets has diminished over time, leaving you ill-prepared to meet the increased production requirements that come along with the advent of AI and ML. 

Judging by assessments made during the last decade, the majority of data centers suffer fundamental asset management issues, including bad or missing equipment data and erroneous or nonexistent procedures for critical equipment. This is a huge problem that Salute is working to solve for the industry. 

When asked why, Lee Kirby, co-founder and chairman of Salute, responded. 

“It is so important to have accurate equipment data and procedures, because you simply can’t manage what you can’t find, and you certainly can’t automate what you can’t manually manage,” he said. “So many data centers are stuck — they can only use IoT sensors to monitor and respond when the solution is available today.” 

The issue with that is, as these data centers look to take advantage of what AI has to offer, the most they can initially accomplish is rolling out IoT sensors that enable AI solutions for environmental systems. But that is just the tip of the iceberg in terms of the capabilities and savings that AI solutions can bring to the data center.

Introducing STEP

Salute Technology Enabled Program (STEP) ensures on-site data sharing in the most efficient manner possible, which serves as a catalyst to reduce waste and improve safety by providing knowledge across all portfolio assets globally. 

Salute performs key functions that are imperative to achieving the flow of information by introducing the systems from the very beginning to store accurate data that can be leveraged globally and benchmarked against industry performance metrics. 

The mindset and culture of the Salute team members is what enables them to offer game-changing services to companies that are seeking ways to fully optimize all of the assets in their portfolios, according to Erich Sanchack, CEO of Salute. 

Imagine life cycle planning with an incorrect number of assets. Take, for example, a UPS refresh. If you’ve planned for a capital spend to replace hundreds of UPS batteries and find out later that there were dozens more needing replacement that weren’t tracked in the asset management system, what do you do? The simple solution would be to increase the budget to account for the additional batteries. But, it’s not always that easy — inaccurate asset data can lead to catastrophic events. Think undocumented breakers during planned maintenance activity. This could cause unplanned outages for customers, severely damaging the data center’s reputation. And, worse, it poses life-threatening risks to the technicians performing the work. 

“That’s why we’re all about optimizing data within the data center asset and transforming it into applied learning across global topographies,” Sanchack said. 

At the site level, Salute is either engaging a new build or taking over an existing data center. The key in both cases is to have a pristine data set prior to the go-live date, but the process of achieving this is slightly different depending on the scenario. Let’s explore each in further detail for clarity.

New Builds

It’s been proven that when you have the same company perform QAQC and commissioning before handing over to the operations team, it improves continuity and knowledge transfer. That’s because the resources with intimate knowledge of the infrastructure and how it works don’t leave to work on their next project for another customer. Communication between commissioning and the handover to operations is more efficient because they both work for the same company and are in direct contact with each other. It allows operations teams to go back days, weeks, and even months later to clarify anything that might have been unclear or misunderstood when the project progressed from one phase to the next.

The two teams will be more invested in ensuring everything functions correctly, as their co-workers will be responsible for operations and they will, therefore, be held accountable for the work they performed. In the Salute model, the QAQC and commissioning teams use the same data repository platforms and operations templates for asset data gathering and procedures. When a system is implemented and used by the same team from the beginning and throughout, the processes and data not only eliminate inefficiencies that lead to enormous operational cost savings, they also enable automation and reduce risk.

Existing Data Centers

It’s easy to envision how Salute QAQC and operations would work together as one when the model is implemented at the start of a new build. However, when it comes to existing data centers, we’ve found that the asset data, processes, and procedures are only accurate about 50% of the time. We approach these projects by starting over, so to speak. To begin, we inventory all assets to correct inaccuracies. We then digitize and integrate them into the tools used by Salute operations to ensure efficiency and ongoing integrity. We continue by updating and digitizing the processes and procedures, so they can be integrated into the playbook for technicians to improve maintenance records and trending tools. 

The outcome of these reset projects drives immediate value for the client by reducing risk, establishing accurate financial accounting, and setting the stage for automation.

Globally Local

To achieve a global model with local execution, you need a service model that uses the data for all activity throughout the life cycle of the asset. That requires intelligent systems that span from a global command center to each individual system at every site. The ability to view local levels from a global perspective and act in real time gives your portfolio the flexibility it needs to respond to market demands as well as threats. The advantages of AI/ML can only be exploited with this model, and those hoping to outpace the competition — or at least maintain their foothold in the race — will need to adopt this service model now in order to take advantage of automation later.

Conclusion

At Salute, tomorrow is already happening, which is why we’re implementing solutions that use the most advanced technologies (drones, robotics, AI, and ML) with our clients today. The early deployments of automation are producing extraordinary efficiencies without human intervention — reducing CO₂ emissions, PUE, and downtime in addition to accommodating capacity growth without the need for additional staff. With these algorithm-driven technologies, we’re speeding up failure detection and predictions, performing root cause analysis, lowering power usage, and optimizing resource capacity to maximize efficiencies at far greater levels than we could ever do manually. Simultaneously decreasing downtime and increasing system reliability translates into enormous savings and expansive growth opportunities — companies that achieve this are the ones that will surpass the competition. The future is arriving quickly, and the pace of change is only going to increase. To learn more about how STEP™ can help you achieve maximum performance from your digital infrastructure by fully implementing AI-enabled operations, contact Salute — the time to act is now.