A data center, its servers, storage and networking, and the peripheral equipment needed for operation all add to a management challenge. Not only is their use and maintenance expensive, but it is also too often inefficient and difficult. Data center asset management is an area where many enterprises could make significant improvements, especially in solving the following problems:
A data center is a warehouse where goods are plugged into power sockets. Like in any other warehouse, the management team must know what is located where.
The bigger the data center, the more time it takes to run around checking machines, serial numbers, and locations. Beyond a certain size, this becomes unreasonable and the data center becomes unmanageable unless you use a better solution.
Do you know how much capacity is being used or still available for a given server?
Figures of 15% capacity used are not uncommon because IT managers would rather play it safe.
However, this means wastefulness in computing resources, power, and cooling.
Information on capacity usage can help customers using dedicated servers in colocation facilities stretch their computing dollars further.
Whether cabling is neatly tucked out of the way (good) or hanging around like so much spaghetti (not good), it can be difficult or impossible to know where the server connection ends up by just a visual inspection.
Ideally, servers and storage units would declare themselves as coming up to the end of the warranty or being due for maintenance.
A manual approach is to manually log all the information in a spreadsheet and check it daily.
However, this takes diligence and is difficult to share in a properly updated way between data center team members.
Will specific server racks support additional machines or new technology? Are machines close to power or cooling limit points, beyond which reliability and safety could be affected?
Too many data centers have no accurate or real-time data on these key aspects.
Data center asset management often starts out with the best intentions when it comes to power and air conditioning. Then internal changes are made or separating partitions are installed, and the initially optimized design ends up with overly hot and cold spots or passages.
Without knowing where these are, data center managers are doomed to spend more on power and air conditioning, which is unnecessary, and risk breakdown where it is.
Data centers need to be understood as a whole entity and component by component.
Adjustments or new electrical loads in one area should also be evaluated in terms of the complete power system.
Readings were made a week ago, and information for servers when first installed may be out of date today.
Upgrades, workload variation, and power and temperature fluctuations call for a real-time reporting solution if decisions are to be made with accurate data.
What is the biggest data center asset management problem you have faced? Share your experience with us with a note in the Comments area below.
If you're in the data center, mission-critical, or cloud services industries, or you sell to the data center industry, don't miss our weekly update newsletter -- Data Center Sales & Marketing Institute (DCSMI) Update Newsletter. Get notified about new reports, events, podcasts, and blog posts.