Data center monitoring has evolved from a simple alert system, a “nice-to-have” feature, into a critical element required for maximizing uptime and ensuring the increasingly complex environment within the data center is running effectively and efficiently. It has become vital for organizations to develop and deploy a system capable of monitoring the overall support infrastructure—including the power and environmental conditions of each room, equipment row and cabinet within the data center.

Many limitations exist with the traditional monitoring solutions in use today—including cost, operational complexity and limited focus on a select group of devices. What is needed are data center monitoring solutions that go beyond simply alerting data center management of immediate problems and actually enable real-time data to be captured and analyzed for management to make intelligent decisions regarding the data center infrastructure. The most effective solutions are grounded in best practices and integrate both the IT and Facilities groups into a single, comprehensive unit that provides actionable information in the decision-making process.

The data center has developed into an extremely complex, distributed environment with tremendous computing requirements not even imagined twenty years ago. Just as the data center itself has undergone a dramatic change, data center monitoring has also rapidly progressed. Gone are the days of measuring temperature by whether or not you need to wear a sweater when checking out the mainframe. Simplistic monitoring systems that report a good/bad condition have given way to a multitude of sophisticated protocols for determining the precise state of the critical devices in the data center.

The management structure of the data center has transformed as well. Two disparate groups—IT and Facilities—have arisen, each focused on complementary, yet opposing functions. The IT group encompasses the demand side of the business, requiring increasingly more infrastructure to run its servers, network hardware and storage devices; while the Facilities group, on the other hand, is responsible for supplying the power and cooling to run the data center. As noted by the U.S. Environmental Protection Agency (EPA) while addressing the challenges of energy efficiency within data centers, “In many data centers, those responsible for purchasing and operating the IT equipment are not the same people that are responsible for the power and cooling infrastructure, who in turn typically pay the utility bills. This leads to a split incentive, in which those who are most able to control the energy use of the IT equipment (and therefore the data center) have little incentive to do so,1” causing a significant silo effect within the organization.

The revolution and increased complexity of the overall data center environment has also been dramatically impacted by the introduction of new technologies—including blade servers, intelligent devices and virtualization—removing the commonplace practice of walking the raised floor space to “feel” the environment and conduct ad hoc monitoring. Not only is it impractical for data center personnel to manually assess the energy usage and cooling requirements, the current use of high- density devices generates such a massive amount of heat that affects all infrastructure resources; which is impossible to monitor without robust systems. The availability of energy resources is reaching its upper limits and the energy costs are soaring, causing organizations to consider repositioning their philosophical, strategic and technical principles for the data center to include more flexibility and more cohesive monitoring.

Multiple limitations exist with the traditional monitoring systems in place today. These limitations include operational complexity, limited support for devices in the infrastructure and an incomplete view of the data center as a whole. Most importantly, information regarding the various devices and applications are housed in so many different systems throughout the organization; making the monitoring and assessment of issues occurring within the data center ineffective and proactive evaluation of potential risks or forecasting the needs in the future impossible.

The limitations due to operational complexity generally stem from the design of the monitoring solution—in particular, how multiple communication methods used in the data center are handled. The use of these widely varied communication methods greatly complicates the development of a solution capable of monitoring different devices with disparate protocols.

The two primary approaches to monitoring include a hardware-based and a software-based solution. The hardware solution requires a protocol to be selected as the primary communications method (Modbus and SNMP are the most widely used options). A hardware device—a protocol converter—then transforms the device protocol into the primary communications method protocol.

Another major limitation with traditional monitoring systems is the inadequate support for devices in the infrastructure. Most monitoring solutions were designed by the manufacturers of the hardware, with the primary intent of selling that particular manufacturer’s hardware. Such a vendor-centric approach prevents customers from obtaining support for competitive products, locking organizations into a particular hardware vendor.

The gap between the IT and Facilities group also impacts the limited support available for devices. Traditional monitoring solutions have generally reinforced the separation between these two groups by designing solutions specifically for IT or Facilities, but not both. IT-based monitoring solutions concentrate on the IT hardware—such as servers, storage and rack-level power devices. On the other hand, Facilities-based monitoring solutions are heavily focused on the hardware managed by the Facilities group—such as UPSs, CRAC units and PDUs. This dichotomy of responsibilities and solutions limits the ability to create a cohesive support system to monitor all devices and infrastructure within the data center.

In addition, most monitoring solutions lack support for multiple protocols. If a solution only supports SNMP, for example, it simply is not capable of monitoring devices that only support Modbus. The lack of support for all devices in the infrastructure has forced organizations to implement multiple monitoring solutions for particular sets of devices. The intrinsic costs of purchasing hardware and software—yearly maintenance charges, training expenses and additional personnel to manage the various systems alone—often becomes a significant investment. Yet, another less obvious cost exists, as well. The consolidation of data from the multiple systems into actionable information for timely reporting and decision making is often extremely difficult, if not impossible—leaving the organization without the ability to comprehensively understand the relationship between the various infrastructure components. As a result, data center managers cannot make the intelligent decisions necessary to effectively manage the overall facility.

Many organizations are faced with the task of meeting operational demands in a facility that pushes capacity to the limit. It is essential for organizations to leverage data center monitoring to properly plan and allocate resources, as well as provide a sufficient infrastructure while keeping costs low. Many companies currently rely on multiple systems with unique methods of extracting and storing information, which lacks the ability to make appropriate comparisons. What is needed is a system that provides an IT view into the data center facility to obtain an accurate, real-time picture of the current state of the critical infrastructure and provides the ability to forecast future needs.

Germiller Intelligent Data Center Monitoring is the ideal holistic data center monitoring solution for data center management to monitor the status and impact of the critical infrastructure through a “single pane of glass”. With Germiller Monitoring Solutions, organizations can benefit from an increase in data accuracy, improved capacity planning forecasts and greater availability. Through the use of proactive monitoring, organizations can identify potential problems before they occur