On 11/28/2012 02:17 AM, Mark Hahn wrote: > > I don't think either Gb or IB are a good match for the many/little > approach being discussed. SiCortex was pretty focused on providing > an appropriate network, though the buying public didn't seem to > appreciate the nuance. >
... and those who did appreciate the nuance, sadly got orphaned in no time. Maybe the venture capital controlling the levers and pulling the rug didn't appreciate the nuance, or had more mundane goals, or both. I guess most here agree that COTS is a safer place to be [affordable, maintainable, flexible, outlives non-COTS technologies, etc] and the more manufacturers and vendors the better. However, can we take for granted that, even though the technologies do change, they will remain plural, inter-operable, DIY-friendly? Does the current trend indicate this? (A single manufacturer of x86 processors soon to be, about the same for Infiniband, motherboard-soldered processors and memory, etc.) Or should we retire the screwdrivers? The phylogeny of computers has many cases where openness, plurality and inter-operability ensured survival, innovation, and expansion, but the quest for singularity and dominance is also there, right? _______________________________________________ Beowulf mailing list, [email protected] sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
