Another basic tenet of Kaizen is that inventory is evil. Inventory is what folks use to cover up the problems and inefficiencies in their production process. In traditional analyses of inventory, the cost of inventory is generally determined based on the cost of the capital that is tied up and the space for the inventory. Kaizen considers a significant additional cost that is much harder to quantify – the cost of allowing problems and inefficiencies in production to persist. Kaizen thinks of inventories like a lake. When you drain the swamp you will find the “bodies” that lie buried below the surface of the water.
I believe in software development the analogy to inventory is excessive hardware. Too many companies simply throw ever more and more expensive hardware into their data centers (or rent more from their managed host). That approach is costly not just in the short term cost of those additional machines, but in the long term cost of not moving the companies to more efficient algorithms and implementations. Often there is the potential for order of magnitude improvements in software. But you have to go look for those and that takes time and smarts. Some companies are forced to look because they experience hyper growth and throwing more hardware at it is either not an option at all or a cost prohibitive one. But this would be good disipline for everyone.
Obviously, the equivalent of the no inventory rule from Kaizen cannot be “no hardware." One good alternative is to establish a level of growth that should be accomplished with no "additional” hardware. Another alternative is to experiment at a given level of hardware. Flickr, for instance, periodically takes machines out of the server pool during low load times until they see where stuff begins to slow down. That effectively amounts to a periodic draining of the swamp.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=3d9a9451-dc5a-4132-8818-328a38bd968e)
