A number of people have asked why I decided to write the RAP/400 program. After all there are a number of data replication products out there already why bring another one to market?
Before I can explain why RAP/400, I need to give a bit of background into the product and why I think it should be considered a serious product for those who need data replication for DR. The product has evolved from a white paper I wrote many years ago entitled Remote Journalling and Data Recovery, due to other projects I did not take it to the next level and automate the apply of the entries from the receivers. A recent conversation with a colleague prompted me to look at this again, I was working on another project and came across new API’s which allowed me to manage the automation better than before so I started the development process. Early tests had some issues with the API’s I was using but after a couple of PTF’s the tests showed the technology worked.
After further research and many questions later I produced a simple apply program which did what I needed it to do. I then wrapped a few help screens, a manual and a couple of commands together to produce the initial product offering. I did not intend to take it further, but after more research (Thanks for the replies Larry!) I discovered a few short comings in the original code which I could overcome using newer technology and commands. So I set about developing a new product (new release really) which will encompass the new technology and build a more flexible solution for the customer. That project is ongoing but should be completed by the end of May 2007. The major differences will be the ability to manage up to 100 journals from a single install and support the File level DDL changes. I have also added a number of extra screens to help with recovery of errors that could occur as part of the apply process.
One big question I have been asked is, why a single price regardless of the IBM Processor Group, normally we price all IBM System i5 by P Group. I thought this product did not meet the criteria to allow me to price on those terms. The market I am going after is the smaller i5 user, I expect most of the bigger users to be more interested in a HA solution than a DR solution.
Another question which is related to the price is one which I struggle with often, how do you price something to ensure it sells? I have been told, if I price it too low no one will buy it because they won’t perceive it has value! I don’t really know how to answer that one, but I feel it is priced according to the market I am trying to break into. I set the $2,000 tag for a number of reasons, firstly I wanted it to be attractive to the low end users. Second I didn’t want to price the product where I would invite criticism from the user community. IBM is running a session and a lab at Common in California next month to show exactly how you can build a very simple solution based on this technology, so I had be be aware that others will produce a similar offering and some customers may decide its worth trying on their own. I worked for one of the HA providers where I was always presented with price as a problem to overcome. Other solutions on the market which were lower priced, regularly competed for the same business. Our argument was we would offer our solution at the competitors cost, that meant they would get a Rolls Royce for the price of a Mini! However as those who break that analogy down know the initial cost may be the same but the running and servicing costs are what kills you!
I feel I am offering the solution at a cost which is both justified and affordable.
Some have asked why now and not before? surely there are enough HA solutions out there? I list many of them on my site! Thats easy, I don’t compete in that area, I am only competing for those customers who either don’t have the budget or requirements for a full blown HA solution. One area I think this will be of use is the Data Vaulting providers who will keep a copy of your data on their systems ready for a disaster. Today many of those offering that service either rely on regular saves or one of the HA products, I am offering a much cheaper and easier alternative. Its using IBM all the way under the covers, I don’t have any proprietary apply technology I have to maintain. IBM has access to the lower code so they should apply much quicker that an apply process built over the top of the external API’s.
Eventually I will build in more features, not to compete with a HA solution because I will never do that using this type of technology. These features are not new technology, I have products built which already contain the basics which will allow easy integration into this solution. The product sits on the target system now, eventually I will have to build a source system module which will be responsible for replicating the objects not journalled today. JobQGenie will be re-engineered to give a perspective of job activity on the source for this technology and built into the source module.
I hope that this goes someway to showing why I feel the product has a niche it will fill and why its been developed now and not before. The price will be a constant challenge and one which I think every developer has to struggle with, I strongly believe it is priced at a fair value and should attract customers who need this type of solution.
I will be in Partner World hoping to discuss this other Business Partners. If you would like to discuss the product etc with me please contact me.