As human beings, we often forget we’re not infallible, and we certainly aren’t immune to making mistakes. Remember the guy who accidentally deleted his entire company with one mistaken piece of code?
Or earlier this year, when tech giant Facebook (and its subsidiaries Instagram and WhatsApp) went down for an entire day. There was speculation as to whether this was the result of a hack, but experts later confirmed it was more than likely caused by an internal mistake with configuration. The outage cost Facebook an estimated $100 million in lost online advertising sales, while its shares fell five percent, wiping about $40 billion from its market value.
What do these two incidents have in common? They were both caused by human error. There were no outside forces to blame, someone simply ‘did an oopsie’ with catastrophic consequences, costing the companies involved hundreds of millions, as well as unfathomable reputational damage (ergo, more lost revenue.)
Eliminating human error
When it comes to mission-critical facilities such as data centers, you simply can’t afford to leave everything in human hands, both literally and figuratively.
This is why many data center operators implement data center infrastructure management – also known as DCIM – to help automate labor-intensive activities, such as workflow management. DCIM can also provide audit trails, as well as increased visibility across the IT infrastructure, building management systems (BMS) and business systems, because after all, you can’t manage what you can’t monitor.
That said, a bit like the age old adage, ‘you are what you eat’, as useful as DCIM is, it’s only as effective as the data it’s fed. For a DCIM deployment to be successful, it must receive data that is accurate, detailed and most important of all, automated. This triple threat is the key to reducing the number of people and, therefore, the possibilities of errors in a facility.
“What everyone is looking for, and trying to get to is that autonomous data center, or as automated a data center as you can get,” says Ross Sonnabend, SVP, product and strategy at RF Code.
If you want to maintain an inventory, in fact any type of audit, control or compliance, you have to know what you have. To find this out via traditional asset management practices, typically, someone is going to need to go out and physically count all of said assets.
Whether this is with a barcode scanner or via a spreadsheet; regardless of the method, it is undoubtedly a time consuming and expensive process: people cost money.
To mitigate this problem, RF Code has come up with wire-free monitoring solutions that eliminate the need to send individuals out to site. What sets this solution apart from traditional asset management, is the implementation of real-time, automated information.
Automating the data provided to the DCIM is key. This is because the majority of the information DCIM applications receive is entered manually, therefore the solution is at the mercy of people entering reliable, accurate data on a timely basis.
RF Code estimates that most IT asset inventories are only 80 percent accurate, meaning tens, or hundred of thousands of dollars are at risk when you’re manually entering information. Eliminating as much human entered data as possible – and therefore the potential for human error – greatly improves the accuracy and availability of the data.
And as mentioned earlier, the need for accurate data is obvious. Feeding inaccurate data into a DCIM application promises a classic ‘garbage in/garbage out’ result. Likewise, the need for detailed data goes without saying: fine-grained, detailed data enables more intelligent decisions and actions.
For the DCIM application to function, it is entirely dependent on the information available to it, so if this is inadequate, it will provide little benefit to the organization. A bit like when the Ever Given got stuck, a lack of intelligent decision-making rendered it utterly useless.
RF Code’s wire-free environmental monitoring solutions and real-time asset tracking solutions add critical intelligence to DCIM applications, enhancing them to ensure that they provide the greatest possible business value.
Tag, you’re it
You might be wondering how this real-time, automated data gets into the system in the first place, and the answer to that (excellent) question is tagging. RF Code will go out and perform an audit, whereby tags are installed on assets. Once this is done customers can essentially associate an RF Code active tag with a server ID. This can be done via a mobile application or a flat file import directly onto the system.
“What most of our customers do, is they actually have their server vendors pre-tag servers with our tags, and they receive a flat file email early, and that automatically goes into the system. So once that system shows up on a pallet in a cardboard box, we can actually read that tag through cardboard or wood,” explains Ross Sonnabend, SVP, product and strategy at RF Code.
He continues, “The second it gets to the loading dock, we already know what’s there because it’s already been imported into the system, showing as being received and in the loading dock. This will spur a lot of customers to actually kick-start the payment process because they know they’ve received it.”
Not only is this great news for RF Code, but it also enables customers to benefit from negotiable early payment terms, as well as making the process for receiving new materials a lot easier, not to mention more organized.
Automation > manual processes
A data center is full of critical assets that need to be tracked, reported on and used in an automated process.
From loading dock storage to the whitespace, RF Code’s solution calculates and tracks it all, enabling customers to set thresholds, alerts and alarms, for example, on something being there too long or not enough, or having too much of one thing and not enough of another.
Automation also helps with decommissioning, which is still a very manual process even today. Traditionally, decommissioning falls upon a person and that person then has to go into a database and make any necessary changes which, of course, leaves the door wide open for human error.
“We can automate that process for someone,” says Sonnabend. “So literally all they have to do is take the device out of the existing home location and move it to a new decommissioning location and they’ve automated the decommissioning process right there, so now it comes off their inventory. They know the time, the date and the person who decommed it.”
And it’s not just decommissioning that can be automated. The solution can also be applied to receiving new materials, another highly manual process that RF Code can help customers automate, helping to avoid the significant losses brought about by human error.
Missing the mark
Many organizations consider themselves to be around 90 to 95 percent accurate when it comes to their data, but experience shows that’s just not the case. “Turns out most people are around 80 percent,” says Sonnabend. “So if you’re 20 percent less accurate, that will start to affect your compliance standards, and raise a lot of interest with the finance team.”
And even industry giants can get it wrong. Using RF Code’s solution, IBM increased its average inventory accuracy from 79 percent to near 100 percent. Fourteen percent of IBM’s assets were also moving without any type of change ticket associated with them, therefore the ability to track this aspect provided secondary benefits.
Not knowing what you have can also cause overprovisioning. “One customer was keeping about 15 percent of their total assets and inventory and we reduced that down to about two percent. The reason was that they now know what they have and don’t need to keep so much on hand. It also allowed them to start redirecting assets across multiple data centers, taking them from mass buys to more strategic purchases.”
The devil’s in the detail
It’s clear from the above that in order for a DCIM to really work for your organization, you need to feed it real-time, accurate data, that doesn’t come from a once in a while audit, but an automated solution that will not only help eliminate the human error brought about by manual processes, but free up staff and resources to focus on more high-level tasks.
And if you’re thinking about space planning or capacity planning, it’s no good guessing, you need that kind of information at your fingertips. “RF Code is an elegant answer to making sure you have the best asset information, as well as environmental and power information in your DCIM,” concludes Sonnabend.
When it comes to your DCIM (accurate, real-time) knowledge is power – you might not be as accurate as you think you are.
More from RF Code
With the potential to extend your business beyond the bounds of traditional enterprise or cloud computing, could Edge be the answer?
The biggest security challenges for organizations running edge data centers might not be what you think they are
Sponsored Security at the edge
The biggest security threat to edge data centers might not be cyber miscreants penetrating the firewall, but very ordinary miscreants simply coming through the door, warns RF Code’s Gregg Primm