Our response to a Data Centre News opinion published on March 6, 2017
On the 6thof March Andrew Waterston wrote an article (LINK) about DCIM where he questioned the future of DCIM along a list of 10 topics. Since there is no option to comment on his article (for some reason ‘comments are closed’) we present you alternative insights here, based on those very topics. Let us start to declare that we agree with Mr. Waterston in many ways. However we think that his conclusions are based on a narrowed focus as he seems to look at the major players only when he fires his ammunition. He concludes that DCIM is an abbreviation of ‘Dead Camels in Motion’, possibly referring to some major DCIM players withdrawing from this market.
When he had widened his view he would have noticed that there are more DCIM providers than those in the quadrants of Gartner and the likes. Many of which offer solutions to the issues he raised. In the following list we go through each of these and fill you in on how Perf-iT offers solutions to each of is arguments.
- What is the user need?
“The author (Mr. Waterston) states that DCIM vendors are increasing their functionality list on paper only without asking the customer what they really need”This is a good question that seems to be overlooked by some of the ‘dead camels’ targeted by Mr. Waterston. They invested immense amounts in creating something that nobody asked for. But there are other DCIM systems that ARE based on customer requirements. A good DCIM is capable of offering solutions to a variety of users, not just the one who shouts the loudest. After all, it’s a matter of presenting the right data in the right order. Good DCIM tools should present solutions to the specific needs of specific users.
- Make mine vanilla?
“Everone has a different view on what DCIM is which depends on the persons background / interest”When the DCIM is flexible enough it can be easily adapted to special user requirements. Whether that is possible depends on how the system was build: top-down, or bottom-up? Perf-iT has build its DCIM suite from the bottom up. Therefore it can be ‘re-routed’ at any level in order to meet the customers requirements. So vanilla, chocolate, mocha or whatever you prefer… when the base is solid and the spine is flexible you can get everywhere you want.
- Questionable vendor neutrality
“all DCIM suppliers are a part of a hardware company, so no neutrality at all”This is a valid point. We too disagree with DCIMs that try to push you in a harness of a specific vendor or brand. You should always have the freedom of choice and that’s why Perf-iT offers bi-directional access to databases and supports over 1500 different protocols to communicate with other systems. Thus ensuring that your system is never limited to a specific brand or protocol.
- Monolithic architecture
“DCIM tools are build up ad-hoc, without a predefined plan but based on customer questions”Waterston was right when he commented on the dangers of a one-size-fits-all solution that covers ‘all’ dynamics of a modern data centre. These are monster applications that are very inflexible and almost impossible to change in any direction. For that reason we have build our system on a industry standard platform Iconics™ which has a wide array of add-ons and features.
- Implementation challenges.
“DCIM has become to complex to implement”A very critical point for many DCIM projects. The ‘dead camels’ of our industry require a very costly implementation process. You either hire expensive consultants or have your own employees struggle for a long period to get it up and running. It doesn’t need to be like that. Our implementations take days, or even a single day if we have all the required data. Our auto-generate tool builds the application ‘on the fly’ and you can start using it within a minimum time frame.
- Support, maintenance and upgrade costs.
“Costs of support, maintenance and upgrade are considerable”
Usually this takes up a big chunk out of your operations budget when you look a the DCIMs from Gartner’s Magic Quadrant. When the DCIM is flexible enough it doesn’t have to be costly at all. Our DCIM is an OPEN system. We even give the source code to you so you can make any adjustments you want. But if you need our support, we are there for you to help you build the software that suits your needs.
It is often easier to get traction (executive focus and internal resources) for a complex, expensive project than a cheap one.More higher ranked officers are getting involved and it therefor generates more prestige. But when all that can be left behind and people just look at what is needed then the right DCIM offers a lot of convenience to its users at an acceptable price. It saves time, saves expenditures and it increases the availability of the systems. The return of investment is often measured in months instead of years.
Outsourcing your data centre infrastructure to a co-location facility does not mean that you can dispose of a DCIM. You still need to manage your assets and coordinate people and processes. You also want to monitor all details of the SLA agreement from your provider. In order to verify that they are within its boundaries or else have the proof when they are violated.
“The need of DCIM disappaires with the transition to the Cloud”We consider cloud as an argument in favour of DCIM and not as an argument against. Cloud providers will make use of DCIM but they will probably not get involved with the ‘dead camels’. They will look for alternatives that can be tailored to their specific needs. The dead camels only offer what is on the table and have no means to adept to these requirements. We see an increase of cloud providers in our customer base because of our ability to specify our product to their requirements.
- Low cost alternatives
“found some point solutions in freeware, low cost asset management platforms, power monitoring Apps and spreadsheets.”All though we don’t want to be considered as ‘cheap’ or ‘low-cost’ we do offer our solution at a much lower price than our big brand competitors. Next to that, our DCIM comes with valuable tools that helps the user to decrease their OPEX and CAPEX significantly and we don’t do multimillion euro advertisement campaigns and will therefore never end up in this ‘magic quadrant’. But we do have a large portfolio of satisfied customers who are happy working with us and our solutions. They happen to the best marketing tool as they are our ambassadors in the field. This keeps our costs low and our product affordable to any data centre, large or small.
Our conclusion is that DCIMs are not a ‘dead camels’ in general. The industry just needs to get rid of some dead wood. As soon as this shake-out is over and DCIM has come down from the cloud it was riding on (fuelled with hot air by market analysts) it will be seen as an invaluable instrument to increase the efficiency and availability of any data centre. Certainly when endusers adjust their DCIM choice biggest list of mentioned functions but on real operational needs.