On 10/3/06, John C. Bland II <[EMAIL PROTECTED]> wrote: > > How soon, after release of CF 8, will you upgrade to CF 8? > > > > Note: This is for research, not just a curious question.
Honestly, it is too early to make such a judgement call. What I mean is the decision will be largely based upon the feature set enhancements introduced. In such times that we're banging our heads against limitations of current release (jvm memory caps) and a new release would provide immediate relief to those issues, we would look to be early / immediate adopters. Any potential changes to cpu / license / virtualization model would also provide a cost / benefit analisys and possible motivation to migrate. However, if the new feature set adds further enhancements without remedying any of our current critical issues, we will likely wait till after the first roll-up update, if ever, to upgrade current infrastructure. Quite possibly we wont consider the cost of upgrading and the expensive testing around that until its time to replace the hardware underneath the version as well. In practice we have never 'upgraded' an existing server pool. We've always waited until it was time to allocate new hardware and brought new versions online then. However we've often decided to continue to run old versions at that point as well. The decision is always based upon an audit of how many applications are still running on the pool we will be retiring, and what would be the estimated cost in testing and recoding applications for compatability. A decision is made at this point if absorbing that cost and effort is worth the feature set / benefits of the new version. Working in a cost - recovery model, if you don't have built in budget for these cycles the path to upgrading is very difficult. We would like to think that applications come into a upgrade / feature enhancement cycle frequently enough that we could address moving them to newer version server pools at these times; however, in practice about 30% of applications deployed on a given pool never reach a upgrade / feature enhancement phase and tend to run until the application is no longer requireded. As such, it isn't practicle to assume you can phase out a pool without a testing / migration cost outside of normal development cycles. In short, I think if your research can wait until there is atleast a beta feature list, I think your data might be more meaningful. Although I hope that some description as to the decision process I've been involved in might be meaningful to you. $0.02 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Introducing the Fusion Authority Quarterly Update. 80 pages of hard-hitting, up-to-date ColdFusion information by your peers, delivered to your door four times a year. http://www.fusionauthority.com/quarterly Archive: http://www.houseoffusion.com/groups/CF-Talk/message.cfm/messageid:255679 Subscription: http://www.houseoffusion.com/groups/CF-Talk/subscribe.cfm Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
