Re: No future for DB2
"bka" <badamache@yahoo.com> wrote in message news:1122515145.936820.249390@g14g2000cwa.googlegroups.com... >>> If only IBM already had a database product hugely respected in the UNIX >>> market, and in fact designed specifically *for* that market, eh? > > 4 in the top 9, including #1: > > http://www.tpc.org/tpcc/results/tpcc_perf_results.asp?resulttype=noncluster&version=5¤cyID=0 I was actually thinking of Informix. DB2 was not designed specifically for UNIX, but for the mainframe. We don't know if Informix would appear at or near the top of TPC benchmarks because IBM won't enter it for them. We used to believe that this because it would embarrass DB2, but personally I think it's more likely to be an extension of the policy of never promoting Informix, except to the existing user base.
"Captain Pedantic" <theharlequin36@hotmail.com> wrote in message news:3krdv3Fv2edcU1@individual.net... > I was actually thinking of Informix. DB2 was not designed specifically for > UNIX, but for the mainframe. We don't know if Informix would appear at or > near the top of TPC benchmarks because IBM won't enter it for them. We > used > to believe that this because it would embarrass DB2, but personally I > think > it's more likely to be an extension of the policy of never promoting > Informix, except to the existing user base. > DB2 for Linux, UNIX, and Windows was not designed for the mainframe. It is a different code base than DB2 for z/OS. It was designed for OS/2 and AIX in its first incarnations. IBM purchased Informix for the patents (which had significant license fees) and the market share. The buyout grew out of discussions with Informix about license fees for the patents and IBM realized that Informix was asking for a significant amount relative to the value of the whole company (Informix stock had seriously depressed in value trying to compete against the big 3).. TPC benchmarks are not exactly cheap since you really need a group of people working full time on it, and lot of expensive hardware. This can easily add up to a few million per year. IBM would rather that customers switch to DB2.
Captain Pedantic wrote: > "bka" <badamache@yahoo.com> wrote in message > news:1122515145.936820.249390@g14g2000cwa.googlegroups.com... > >>>>If only IBM already had a database product hugely respected in the UNIX >>>>market, and in fact designed specifically *for* that market, eh? >> >>4 in the top 9, including #1: >> >>http://www.tpc.org/tpcc/results/tpcc_perf_results.asp?resulttype=noncluster&version=5¤cyID=0 > > > I was actually thinking of Informix. DB2 was not designed specifically for > UNIX, but for the mainframe. We don't know if Informix would appear at or > near the top of TPC benchmarks because IBM won't enter it for them. We used > to believe that this because it would embarrass DB2, but personally I think > it's more likely to be an extension of the policy of never promoting > Informix, except to the existing user base. I think it is both. -- Daniel A. Morgan http://www.psoug.org damorgan@x.washington.edu (replace x with u to respond)
Mark A wrote: > IBM purchased Informix for the patents (which had significant license fees) > and the market share. The buyout grew out of discussions with Informix about > license fees for the patents and IBM realized that Informix was asking for a > significant amount relative to the value of the whole company (Informix > stock had seriously depressed in value trying to compete against the big > 3).. Mark, you got that one reversed: http://news.com.com/IBM+sues+Informix+for+patent+infringment/2100-1001_3-236666.html IBM had been negatioating with Informix for a while to get Informix to pay and likely to cross license in the process. This went nowhere, so IBM sued. IBM Corp has cross licensing agreements with many soft and hardware companies and is deriving billions(!) of dollars from the licenses. I'd love to know how much Oracle and MS are paying for the cross-licensing, btw.... > TPC benchmarks are not exactly cheap since you really need a group of people > working full time on it, and lot of expensive hardware. This can easily add > up to a few million per year. IBM would rather that customers switch to DB2. IBM also doesn't publish TPC on DB2 for zOS or DB2 for iSeries. Neither does it publish IMS. From how I read the TPC-C specs IMS could comply and it would likely run circles around any RDMS on TPC-C (including DB2 and IDS) due to its architecture. To be entirely fair there are several metrics in TPC. Some of them (like price performance for "realistic" systems are not all that expensive. It's teh TB sized TPC-H and millions of TpmC in TPC-C that darin ones wallet. Cheers Serge -- Serge Rielau DB2 SQL Compiler Development IBM Toronto Lab