Data center management: What does DMaaS deliver that DCIM doesn’t?
- 24 May, 2018 20:00
Data-center downtime is crippling and costly for enterprises. It’s easy to see the appeal of tools that can provide visibility into data-center assets, interdependencies, performance and capacity – and turn that visibility into actionable knowledge that anticipates equipment failures or capacity shortfalls.
Data center infrastructure management (DCIM) tools are designed to monitor the utilization and energy consumption of both IT and building components, from servers and storage to power distribution units and cooling gear.
DCIM software tackles functions including remote equipment monitoring, power and environmental monitoring, IT asset management, data management and reporting. With DCIM software, enterprises can simplify capacity planning and resource allocation as well as ensure that power, equipment and floor space are used as efficiently as possible.
What is DMaaS?
Datacenter management as a service, or DMaaS, is a service that’s based on DCIM software. But it’s not simply a SaaS-delivered version of DCIM software. DMaaS takes data collection a step further: Equipment and device data is collected from scores of data centers, and then that data is anonymized, pooled and analyzed at scale.
DMaaS could potentially transform the way data centers are managed and operated, says Rhonda Ascierto, research director for the datacenter technologies and eco-efficient IT channel at 451 Research.
“DMaaS aggregates and analyzes large sets of anonymized customer data and can be enhanced with machine learning,” Ascierto says. “A key goal is to predict and prevent data-center infrastructure incidents and failures and to detect inefficiencies or capacity shortfalls.”
Two early players in the DMaaS market are Schneider Electric and Eaton. Both vendors mined a slew of data from their years of experience in the data center world, which includes designing and building data centers, building management, electrical distribution, and power and cooling services.
Access to that kind of data, harvested from a wide range of customers with a wide range of operating environments, is the real value of DMaaS, which enables an enterprise to compare its own data center performance against global benchmarks. For example, Schneider’s DMaaS offering, called EcoStruxure IT, is tied to a data lake containing benchmarking data from more than 500 customers and 2.2 million sensors.
“The key to DMaaS' value – and what differentiates it from on-premises and SaaS-delivered DCIM – is the pooling of anonymized data to run big-data statistical analysis using a combination of machine learning, anomaly detection and event-stream playback,” Ascierto says.
DCIM vs. DMaaS
Since its emergence, DCIM software has been long on hype and shorter on delivery. It’s difficult to deploy, and ROI can be unclear.
“DCIM can reduce risk while enabling new efficiencies, better capacity forecasting and improved business agility. Yet it remains an under-deployed technology owing to challenges such as implementing operational changes to support DCIM and difficulties in measuring overall ROI. DMaaS promises to mitigate these challenges and extend DCIM's value,” Ascierto says.
Data-center management is trickier when computing environments are spread among traditional on-premises data centers, cloud and colocation sites, and edge computing applications. A traditional DCIM model requires software installations at every data-center site, which can be challenging when trying to compare performance across environments. DMaaS is a cloud-based remote monitoring service that’s designed to support a range of deployment sites.
Still, DMaaS has a long way to go to reach its potential. The long-term goal of DMaaS is to integrate physical data-center infrastructure management with many other services, including IT workload management, energy management, connectivity and business costing, Ascierto says. But it will take time to get there.
“DMaaS does not yet match the breadth of capabilities available from 'traditional' DCIM, but over time, we expect it will go further,” she says. “DMaaS takes the data-center world beyond DCIM, and beyond single-site, proprietary management. Over time, other data and services will likely be added, including integrated workload management, energy management, customer relationship and business systems, weather, staff services, and security and network management.”
Choosing between on-premises DCIM and cloud-based DMaaS isn’t a clear-cut decision. It may make sense to keep some DCIM monitoring capabilities on-premises for security and latency reasons, Ascierto says. Sending data to a centralized cloud facility and back isn’t practical for time-sensitive monitoring and alarming, for example. Forecasting, on the other hand, isn’t essential to day-to-day operations and makes sense to do off premises.
“I don’t think they’re an either/or proposition. I think they’re an ‘and,’” Ascierto says.