> On Thu, Aug 31, 2006 at 09:58:59AM -0700, Matthew Dillon wrote: > : that 75% of the interest in our project has nothing to do with my > : project goals but instead are directly associated with work being done > : by our relatively small community. I truely appreciate that effort > : because it allows me to focus on the part that is most near and dear > : to my own heart. > > Big question: after all the work that will go into the clustering, other than > scientific research, what will the average user be able to use such advanced > capability for?
Average user? Not much. But for anyone who administers server farms, the single-system-image style of clustering is a promise that multiple vendors have tried to deliver, but all have failed (in my memory). Putting on my flame-suit, SCO's Non-Stop Clusters with UnixWare was the best implementation I've seen, but that failed to mature and was axed after the Caldera takeover. (Bill Hacker will likely disagree, but I will point out that I've only been active in the U*ix space for ~10 years at most.) This is where DFly can really steal FreeBSD's thunder. Many of those who use FreeBSD do so in an ISP context. Many of those installations have multiple servers either for load-balancing or redundancy issues. But maintaining many separate servers is a pain. Being able to have multiple DFly boxes that are clustered together and can be administered (at least from the application space) as a single-system-image environment will provide a great boost to productivity. Sure, Xen and virtualization have their place, but there still is a place for huge systems -- whether it's a multi-million dollar 64-way machine (like I use at my "day job"), or a cluster of 32 quad-core Opterons. Regards, -- Matt Emmerton
