MattR
05-29-01, 10:48
| http://www.mcjones.org/System_R/SQL_Reunion_95/sqlr95-Teradata.html Explains a little bit about the history and beginnings of Sybase. INGRES which is now PostGRES I think and Sybase apparently have a lot in common; although Sybase has enjoyed much larger commercial success. I guess this is due to, as they said, the threading that was so revolutionary at the time AND it wasn't $150,000 (like Oracle is today). What Sybase needs, in my opinion as a staunch supporter and ASE Zealot ;), is something new and revolutionary to keep them going. I don't foresee them going away anytime in the near (and hopefully distant) future but I think a couple things that could help would be: * Technical revolution; a way to squeeze more performance while keeping the same reliability and stability. We’ve been lucky to have 99.999% uptime (EBFs aside of course) but ways to achieve higher performance can be had somewhere (cursors perhaps? :) ) * Clustering / Replication; The days of the 64-way Sun box are quickly fading. The future will be NAS, SAN, Fibre Channel, InfiniBand, iSCSI storage feeding many smaller SMP clusters of 2 or 4 way machines. * Continue Linux excellence; Linux runs quite a few of the servers in the world. We run 11.9.2 on RedHat and love it; the low-cost of the Intel platform combined with the performance of Linux make it a win-win solution. Compare the prices for a dual PIII Xeon vs. a Sun 220 or 280R and you see why it’s desirable. I’m not sure what the prices for the 64-bit Intel chips will be, but I’m sure they won’t be nearly as expensive as a Sun machine. (Simply put: many customers can’t afford a $80,000 Sun machine + RAID array). * Enterprise Portal? This seems more like a buzzword than anything else. Most customers have their own applications interfacing with ASE. Any other thoughts? |