A bunch of venture funded network management start-ups are storming the enterprise space with pockets full of venture capital money.
What ties them all together? All of the tools are open source, but that isn’t new in network management. Projects like Nagios, OpenNMS as well as a raft of others have been around for a decade or so. What’s new is the combination of open source products and the level of funding going into the new players.
Open source as "new wave" strategyAs open source is a significant part of the new wave's strategy, why is it important? The enterprise management space is mature, the big players like HP OpenView and IBM Tivoli have both been around a long time. In that time they have built a large user base. The kind of user base that is going to be tough to break into.
One thing that is going to be a big part of the buying process is credibility. Has the solution solved problems for lots of other people just like me. With the widespread adoption that open source can bring, the answer will likely be yes.
Open source can help build a user base very quickly. Better than that, you can build a user base of evangelical users quickly. What’s more, you can build a participatory eco-system too, in a way that is very difficult around a proprietary product.
Evangelical users, through access to the source code, can modify the system to suit themselves and others in the community. Of course, not every systems admin has the hard core software engineering skills required to modify some of the new wave products. Systems built upon Java, like Hyperic, can pose significant skills barrier making a select few in the community able to modify them in anything but a trivial way. One of the big wins for Zenoss is likely to be their adoption of Python as their implementation language. Python lowers the skills barrier and should encourage admins who would otherwise be unwilling to climb a steep Java learning curve.
Problems, pitfalls...There are potential pitfalls. None are likely to be fatal, but the new wave companies will have to bear them in mind when they are coming up with their respective strategies.
- Open source traditionally has a bottom up implementation -- the techies get hold of the tool, adopt it without telling any of the higher ups. Enterprise systems are traditionally implemented top down. The CIO decides that a tool is required and then selects a tool based upon a beauty contest between a number of vendors. Enterprise tools cost a lot to implement, not just in license fees, but also in training, consultation and the like. Only senior managers can sign-off the size of budgets that typically go with an enterprise tool. The bill for implementing an open source tool may not include the up front licensing cost, but will still have the training and consulting. The training and consultation bill can still be substantial. The kind of substantial that requires the CIO's signature.
- Open source licenses -- some people don't understand open source software licenses and have very peculiar ideas about them. Some even go so far as to ban open source use completely. Maybe a dual license scheme could be used with companies who aren't comfortable with the open source license. That, or a large dollop of user education about open source licenses. Though, I doubt that the education route would be either easy or quick.
- Without a large product sale, how are you going to give sales people an incentive -- like it or not, sales people are a part of the enterprise sell. Using the established sales channel is going to be tough. The channel is already selling products that are likely to yield higher commission. Proprietary software products have an upfront license fee, some of which will go in commission. The open source company may not have the luxury of the up front license fee so may struggle to match the kind of commissions offered by their proprietary competitors. As a sales person, which are you going to sell, the high commission product or the lower one? I doubt too many sales people will be too excited about the open source part of the sale. Of course, the new wave could just avoid the channel and sell direct. That does suggest that they will need to build a formidable international presence in each country/region. That isn't going to be cheap.
ConclusionIt sure is interesting times in network management at the moment. We don't just get one venture funded enterprise open source tool, we get three all at the same time.
Whilst the open source vendors have their challenges, so do the proprietary vendors. How do they stop their user base slipping away from underneath them to their open source competitors?