Keeping IT infrastructures purring like new in a business world of decreasing budgets and increasing user demands is no simple task.
To determine what was happening in the infrastructure management space, CBR asked two experts from two very different service providers for their take on the subject. The first is Andy Brauer, CTO of Business Connexion, the second, Greg Montjoie, hosting business unit manager at Internet Solutions.
Andy Brauer, CTO of Business Connexion
How has the discipline of managing infrastructure changed over the years?
Brauer: There has always been an emphasis on change control, but if one looks at governance requirements one will see that it is not just a question of change control but also of who is changing what. It is about governance, compliance and change control as opposed to just being about change control.
There is also a lot more automation. ITIL (Information Technology Infrastructure Library) is driving catalogues, which are driving services around these areas, so the management of infrastructure is becoming automated. Things like root cause analysis and event correlation have been around for a while, so what is happening is actually a reinvention and redefining of infrastructure management rather than a quantum leap. The whole emphasis today, in this area, is becoming more service focused. Infrastructure management is growing up now and focusing on the service: the service to the end user and the service to the business user. The focus has to move away from what the product can deliver to what the industry requires as a service.
Are the big players managing to do the necessary or are they too big, bloated and old?
Brauer: BMC seems to have a very good vision, but I see a lot of pressure coming from new niche players that are leapfrogging the bigger players by focusing on service. It is difficult for these big companies, even though they are doing a lot. IBM, CA, BMC - all of them offer a lot of direction but it is more difficult for them to evolve than it is for a new system on the block which can identify specific problems. Open source is also maturing rapidly in this area.
If you do not innovate, it is very difficult to compete.
Montjoie: The big players offer stability and broad functionality, but this comes with a high cost and potential slow reaction time to customise a solution to a business' exact requirements. Niche management tool scan provide solutions but there is an inherent risk going with product that may have a limited view of your operation. By inference, niche tools will not have the scope or scalable functionality that a larger player may have.
With this in mind it is very important to understand what the businesses endpoint need is, as a niche tool may not grow to give your business the total functionality the larger players ultimately will. The flipside is the larger system tools may be overkill for your requirements and may not deliver on the expected ROI if all the available functions are not utilised.
Who are the smaller (maybe even local?) players?
Brauer: It depends on the business requirements. Does a small to medium sized company need all of the functionality that these major systems give them, or are there specific areas that they want to manage? There are different models that we use in the industry.
One is called FCAPS (fault, configuration, accounting, performance, security), which looks at the faults, accountability, configuration, performance and security, among other things. But do all companies need all of this functionality? Open source is a big player in the small to medium enterprise space. There are very interesting tools on the market, which are quite sophisticated - both open source and less well known solutions (not part of the big four).
It is important to ask what it is that a company needs to do. This is a question that the industry has made a mistake on many times. Companies often have a solution first and then look for the problem, as opposed to first identifying the problem that needs to be solved.
Montjoie: Smaller players provide a fantastic service for monitoring/assigning a particular function or set of functions within your environment. When looking at the bigger picture you may find that niche products fail to give the complete overview of the environment and may lack integration into the larger products as you grow. You may be able to bolt together enough niche products to give you this functionality but with this will come a large cost to develop an integration layer.
What about the SMB market?
Brauer: One needs to look at the context in which the words 'small' and 'medium' are used when talking about small and medium sized enterprises. In America, where the terminology comes from, a medium sized enterprise is what we would consider large, and a small business is what we would consider medium sized. We do not have the same scale as the Americans. So the enterprise tools that we are skilling and gearing meet the medium sized enterprise requirements in the States. I would also say that the dynamics of our country are different at present because we are a Third World country that is quickly becoming First World.
Again, companies must ask what the requirement is, what the problem is, what the risk is, and what the compliance issues are. A small company that is not listed on the NYSE has no need for SOX (Sarbanes Oxley) compliance, and therefore does not need certain components in its infrastructure. It all comes down to the business requirements - not what the tools can do.
Montjoie: The market is moving to an outsourced environment where small businesses can leverage off shared technologies and centralised data centres and networks to give them enterprise class functionality at an affordable price.
Is there a market for web-based managed infrastructure management services?
Montjoie: Absolutely, it is becoming more and more critical to manage and monitor environments from a central console that can be accessed anywhere. The Web provides an ideal conduit for this type of interface.
Brauer: Service orientation is aimed at getting things done as quickly as possible. The most common way of doing this is via the web. Web 2.0 is more interactive and the understanding of what the Web can deliver is changing. Also, the dynamics of the way that people are engaging with the Web are changing the way that it works. The community is closing and there is more peer-to-peer technology coming out of the Internet's evolution and there will be more direct interaction. The point is that the Web has become so widely adopted and accepted that the interfaces and management components are all gearing themselves around these trends.
Traditional systems used to have something running in the background with a web front end where users accessed information. This typically made systems very heavy, slow, cumbersome and unreliable. The new approach is to gear and write the system design using web-type technology from scratch.