<iframe src="//www.googletagmanager.com/ns.html?id=GTM-WHST8N" height="0" width="0" style="display:none;visibility:hidden">

Stay Powered Blog

Threats detected

Posted by Erik Stabile on Aug 27, 2015 12:11:22 PM


Server Technology recently published a case study highlighting one customer's experience utilizing Switched POPS with PIPS PDU and Sentry Power Manager to improve uptime from cyber attacks. 

Here was the challenge:

Our client uses their global lab infrastructure to test various hardware, software and network configurations in order to identify advanced persistent threats and other “zero day” vulnerabilities in the global infrastructure of the internet. For a variety of reasons, our client places some of their development systems in remote colocation facilities. The power and cooling systems of the colocation providers vary by facility, and are different from those within our client’s internal lab environments. Successful deployment of gear at the colocation provider requires a detailed understanding of how much power is consumed while performing a variety of software algorithms. Our client sizes their racks for 15kW maximum power consumption, and roughly 9kW when idle.

When the Senior Infrastructure Engineer joined our client’s team, he was faced with determining how to put a 50 rack expansion of the lab into a colocation facility. The new racks were to have a high number of compute nodes that averaged 1.2kW each.

Inside the internal labs, their infrastructure engineer was seeing a number of potential power related problems that came from having end of row breakers that were incapable of remote monitoring and power measurement. In his experience, he felt he was likely to see the same situation in many of the potential colocation providers. In order to avoid oversubscribing the upstream breakers, he needed to identify power draw and current requirements of each IT rack in his datacenter, along with looking at the hardware that would be required for the colocation deployment. Furthermore, he needed a way of aggregating the rack level data from the internal labs in near real time to provide alarms and warnings whenever the power draw was likely to exceed the capacity of the end of row breakers.

Click here to read the solution! 

Topics: case study

Get In Touch

Latest Posts

Follow Me