Suvish Viswanathan is the senior analyst, unified IT at ManageEngine, a division of Zoho Corp. You can reach him on LinkedIn or follow his tweets at @suvishv. This is a second part of a three-part series, with this article focusing on “The Evolution of Infrastructure Management.”
In my last post, we looked at the evolution of IT infrastructure management. As demand for data and computing resources grew, so too did the complexity of the IT infrastructure — but so too did the complexity and power of the software we use to manage this infrastructure. Today, we have software tools that can manage enormous numbers of physical servers, each of which hosts hundreds, if not thousands, of virtual servers. The same software can help IT managers track changes in switches and firewalls, storage systems and more.
But the same software rarely keeps tabs on the temperature fluctuations inside the racks supporting all these IT assets. The same software rarely monitors the physical security of the individual racks, the rack cages or even the doors to the data center itself. And can that same software tell you how much diesel is in the tank in case you need your backup generators? You can probably guess the answer.
A Different World for IT
The physical data center itself — the brick and mortar building that surrounds all those physical and virtual machines — consists of electrical supply systems; backup power systems; power distribution units (PDU); heating, ventilation and air conditioning (HVAC) systems; fire detection and suppression systems; biometric security and surveillance systems; water systems and more. Data center managers pay as much attention to monitoring, managing and maintaining these systems as IT managers do to their IT assets. And just as IT management has evolved to deal with the increasing complexity of the IT environment itself, so too has data center facilities management evolved to deal with the complexities involved in that world.
Today’s data center managers have tools and technologies for monitoring each of the systems and services just mentioned. Some of these tools may be brilliantly integrated for security purposes — access control systems, for example, that can collect and validate input from biometric devices and then unlock the doors only to specific data center areas, even to specific data center cabinet enclosures. If the visitor attempts to access an unauthorized area or rack enclosure, the access control system can detect the attempt and automatically retrain the security cameras while alerting data center security. If the unauthorized access attempt persists, the system can automatically alert local law enforcement officials.
Other systems have been developed to optimize data center operations for cost savings. Computer room air conditioning (CRAC) systems, for example, may monitor the temperature from hundreds, even thousands, of locations around the data center and make micro-adjustments to hundreds of air flow and cooling units to distribute heat more effectively through the complex. This can both optimize energy use where cooling is required (which can lower operating costs) and maintain an optimal operating environment for the IT assets whose performance is temperature sensitive. At the same time, the fact that these systems make use of sophisticated automation to monitor and manage the complex features of this environment also means that the data center can keep its personnel costs in check. There is much to be managed and maintained, but if the systems can do the bulk of the work without the need for physical intervention on the part of an administrator, then the data center can operate in a lean and efficient manner.
But It’s the Same Mission
The reality is, though, that all these efforts to enable greater efficiencies — in both the IT and data center management worlds — have only partially succeeded. Each world may operate more efficiently, but they remain two parallel worlds. IT and data center managers might quickly agree that they had a common purpose — to enable the optimal delivery of a service to customers — but the levers they pull and the buttons they push are very different.
The IT and data center worlds should not need to operate in parallel, though. What they need is to operate as one, in a truly integrated manner. That’s what DCIM — data center infrastructure management — should be all about. Instead of the CRAC system simply monitoring temperatures throughout the data center and making micro-adjustments to fans and airflow systems, for example, what if it were integrated to do more than that? The CRAC system should be working intimately with a virtual machine management system so that the broader data center management system — call it the DCIM system — could proactively move virtual machines from a hot server in one part of the data center to a cooler machine in another part of the data center as part of a broader management strategy.
This more fully integrated universe of service delivery management — one that consolidates IT management and data center facilities management — is where the next breakthroughs need to come. The assets under facilities management and the assets under IT management need to be integrated into the same asset management tool so the entire service delivery infrastructure can be tracked and managed throughout its lifecycle. And that means an evolution within both worlds.
An asset management system supporting both worlds needs to be able to capture data from infrastructure components using not just protocols such as SNMP but also protocols such as Modbus, BACnet and others. Similarly, all the management tools must themselves interact with greater communications interoperability — whether that’s building/space management (BSM), security and fire suppression units or any other. With greater communications flexibility, you will not have to rip and replace existing tools as quickly in the future, and you won’t get locked down to a single vendor. Finally, service managers relying on these tools will need an analytics engine that can make sense of the information collected within this system. Only with strong analytical tools will decision-makers be able to make truly well-informed decisions.
If we can integrate the tools we use to monitor and manage the IT assets and the tools we use to monitor and manage the physical facilities in which these assets reside, then we can see levels of operational efficiency and service delivery performance that we have not seen before.
So what might such a world look like in practice? Stay tuned. You can already guess what I’m going to write about next…
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.