S!aP for Data Centers and Communication Networks
Data centers are the engine of the IoT/IT-OT ecosystem, where petabytes of data streaming data is being processed and exchanged in real-time across billions of devices. ICT infrastructure has large carbon and energy footprints - roughly 10 percent of global usage and 2 percent of global carbon emissions. With ever-increasing load on data centers [from revenue generating IT equipment] and support infrastructure [expenses], the corresponding uptake of resources must be monitored, trends predicted, and operating parameters controlled, to minimize downtime and maximize productivity.
Data centers require high availability and reliability for their daily operations, while enterprises are under pressure to reduce TCO i.e. recurring costs of IT and facility management, while improving utilization with reduced management complexity.
The operational costs of data centers are primarily [ref. A. Greenberg et al. 2008] due to Servers (45 % of the amortized costs), infrastructure (25 %), power draw (15 %) and networking equipment (15 %).
Analysis of multi-disciplinary data center metrics enables optimum performance with control of operating costs. Prescriptive analytics yield a holistic mechanism to improve utilization and availability, gain insights on possible inefficiencies, thus enabling capacity planners and architects to optimize the facilities design for effective, scalable and reliable performance.
S!aP COGNUS for Data Centers provides advanced intelligence to advise and optimize the complex interplay across IT systems and operating infrastructure. For instance, it is critical to monitor server utilization conditions, vs. power quality, and load-efficiency curves. Power Utilization Effectiveness and other critical KPIs must be optimized against multiple targets and constraints.
Revenue-generating IT equipment viz. servers, network and security devices consume only about 30-45% of power drawn, while operating infrastructure: electro-mechanical, electrical control and distribution systems - PLCs, Drives, Field IOs, Digital Switchgear, UPS, and power-intensive cooling requirements like CRAC, AHUs, chillers and pumps - draw a whopping 50-70% of input power.
Ultimately, for compliance to global standards such as ISO/IEC 30134 and LEED, GRIHA, IGBC for sustainable green buildings, it is imperative to compute the cross-correlations of metrics for prescriptive advisories in the nine core dimensions of data center operations [ref. V D Reddy et al. 2017, IEEE Trans. Sust. Comp.]: energy efficiency, cooling, greenness, performance, thermal and air management, network, security, storage, and financial impact.
We invite you to assess your Data Center for ISO/IEC 30134 and Green Building/Sustainability compliance.