This should be the stuff of an international appeal; please send money – there are millions of IT servers out there that are slowly becoming orphaned. They have no-one to care for them. No one to look out for their interests. In short, they are about to become a lost generation of IT equipment.
Before you ask – No! I don’t think I’m the one who’s losing it…
It’s an odd thought though, that as workloads become increasingly abstracted from physical servers, the continued and main concern for most IT professionals – from CIO to project leader – is to ensure sufficient compute capacity to serve the workload requirement of the business.
How scaling that resource is to be achieved would seem to be of little account, and the physical machines which actually provision it are clearly becoming a secondary consideration. We all know that IT loads, particularly in highly virtualized and cloud data centers, can vary in both time and location. In order to ensure availability in such a system, ongoing monitoring of rack-level power and cooling health are obviously critical.
However, with a few notable exceptions, those running the facilities which power and cool the physical servers, and who are dependent upon their consumption of resources for their own raison d’etre, are themselves abstracted from this equipment.
Facilities professionals can see the servers, and they ensure that they are properly powered, protected, cooled and housed. But like the proverbial ‘one man and his dog’ factory, they stand to get bitten should they get caught touching any of the IT devices.
Essentially, virtualization transforms physical IT servers to a virtual capacity which must be monitored, analysed and managed. As a physical capacity, there is therefore a solid argument for it to be managed by the people who manage all the other physical aspects of the data center – the data center manager.
In fact, some of the customers we work with now are adopting this more progressive view of data center operations. If it becomes pervasive – and logic supports the idea – it’s not unlikely that in the future the IT department will relinquish control of physical server management to the facilities guys.
Without going into sales pitch mode, one of the areas of functionality that DCIM has brought to the table (to differentiate it from asset management and other pre-DCIM tools), is the protection of virtual machines. By monitoring the data center environment, emerging threats to operations can be mitigated to ensure continuity of operations.
By integrating DCIM software with the hypervisor, VMs can be safely and automatically moved to areas known to have sufficient power and cooling to handle the load. Conversely, VMs can be moved away from racks that develop power or cooling problems. If there’s a power outage at rack level, or a cooling fan stops working, the hypervisor can be notified of the event and “at risk” VMs can be moved to a healthy rack elsewhere in the data center. This can happen fully automated and without staff intervention.
In my experience it is the capability to ‘anticipate’ threats and move applications to a more available environment in real time which is becoming increasingly sought after. In this context perhaps the best candidates to foster the IT equipment are going to be those who already have responsibility for data center operations, and not IT operations. Problem solved – thank you DCIM!