normalization != bad performance
in [OOP]
|
Prev: Creating Full Name from given, initials, sn, when some mightbe null
Next: Call for Papers with Extended Deadline: 2007 International Conference on Multimedia Systems and Applications (MSA'07), June 25-28, 2007, USA
From: Thomas Gagne on 27 Feb 2007 12:31 Daniel Parker wrote: > On Feb 27, 9:50 am, Jerry Coffin <jcof...(a)taeus.com> wrote: > >> A typical SQL DBMS requires _drastically_ faster hardware than a >> hierarchical database to provide even roughly equivalent response time >> -- to the point that a SQL DBMS running on current hardware is about the >> same speed (or slighthly slower) than a hierarchical database was around >> 25 years ago, running on hardware that was current at the time. >> >> > Which makes it all the more impressive that the RDBMS essentially > wiped out the hierarchical and network databases as living, growing > entities around the mid 1980's, when network speeds and CPU speeds > were much slower :-) And pre-empted OO databases, so that they never > even became a contender. Wouldn't you agree? Can you think why that > might have been the case? Even though, at the time, they were > unquestionably slower? > Flexibility during design. When the software explosion began in the 90s the best place to play with database models was inside relational databases. Another thing they had going for them was a repeatable theory to how they should be designed. Add to that (Sybase's) invention of the database server accessible to clients using only thin procotol libraries and everyone else's investment in RDB technologies and you have the recipe for a winner. The curious thing, is, if RDBs are a great and flexible place to model your data, is there any value in moving data to a more rigid but faster model? Or, do most practitioners accept the that hardware will overcome software's performance issues? > But I don't believe that your statement is true. I haven't had the > pleasure of working much with CODASYL or IMS databases, but my limited > exposure doesn't suggest that it's true, in general. In the early 90s (or was it late 80s?) I testing proprietary, network, and relational databases and found (surprisingly for me) that network databases were immodestly faster for both transaction processing and queries. -- Visit <http://blogs.instreamfinancial.com/anything.php> to read my rants on technology and the finance industry.
From: topmind on 27 Feb 2007 13:00 Thomas Gagne wrote: > Daniel Parker wrote: > > On Feb 27, 9:50 am, Jerry Coffin <jcof...(a)taeus.com> wrote: > > > >> A typical SQL DBMS requires _drastically_ faster hardware than a > >> hierarchical database to provide even roughly equivalent response time > >> -- to the point that a SQL DBMS running on current hardware is about the > >> same speed (or slighthly slower) than a hierarchical database was around > >> 25 years ago, running on hardware that was current at the time. > >> > >> > > Which makes it all the more impressive that the RDBMS essentially > > wiped out the hierarchical and network databases as living, growing > > entities around the mid 1980's, when network speeds and CPU speeds > > were much slower :-) And pre-empted OO databases, so that they never > > even became a contender. Wouldn't you agree? Can you think why that > > might have been the case? Even though, at the time, they were > > unquestionably slower? > > > Flexibility during design. When the software explosion began in the 90s > the best place to play with database models was inside relational > databases. Is there now a better place? > Another thing they had going for them was a repeatable > theory to how they should be designed. Add to that (Sybase's) invention > of the database server accessible to clients using only thin procotol > libraries and everyone else's investment in RDB technologies and you > have the recipe for a winner. > > The curious thing, is, if RDBs are a great and flexible place to model > your data, is there any value in moving data to a more rigid but faster > model? Or, do most practitioners accept the that hardware will overcome > software's performance issues? Most businesses *want* flexibility. Due to "free trade" and the arrival of cheap educated 3rd-world labor, the U.S. must specialize in change. If change is your comparative advantage, then sacrificing flexibility to gain speed is not very logical. -T-
From: Daniel Parker on 27 Feb 2007 13:21 On Feb 27, 12:31 pm, Thomas Gagne <tga...(a)wide-open-west.com> wrote: > > In the early 90s (or was it late 80s?) I testing proprietary, network, > and relational databases and found (surprisingly for me) that network > databases were immodestly faster for both transaction processing and > queries. > Sorry, but by "immodestly faster", do you mean "much faster" or "not much faster" :-) Daniel
From: Thomas Gagne on 27 Feb 2007 22:29 Daniel Parker wrote: > On Feb 27, 12:31 pm, Thomas Gagne <tga...(a)wide-open-west.com> wrote: > >> In the early 90s (or was it late 80s?) I testing proprietary, network, >> and relational databases and found (surprisingly for me) that network >> databases were immodestly faster for both transaction processing and >> queries. >> >> > Sorry, but by "immodestly faster", do you mean "much faster" or "not > much faster" :-) > I mean *much* faster. We measured CPU and disk IOs as well, and network was *much* lower on both accounts. -- Visit <http://blogs.instreamfinancial.com/anything.php> to read my rants on technology and the finance industry.
From: Dmitry A. Kazakov on 28 Feb 2007 03:49
On Wed, 28 Feb 2007 07:15:49 GMT, JXStern wrote: > On Tue, 27 Feb 2007 07:50:02 -0700, Jerry Coffin <jcoffin(a)taeus.com> > wrote: > >>In article <CpydnTUKeJGdsXnYnZ2dnUVZ_uygnZ2d(a)wideopenwest.com>, >>tgagne(a)wide-open-west.com says... >>> Jerry Coffin wrote: >>> > <snip> >>> > >>> > The relational model is a bit like the ISO 7-layer model for networks. >>> > Each makes a great reference model and in theory, each should work >>> > perfectly well. OTOH, real implementations of either are few and far >>> > between. The few that do make the attempt all share one characteristic: >>> > stunningly awful performance. >>> > >>> Sorry, can't let that one go. > > Well, good, because it's gibberish. I can't tell if he's talking > about the RDBMS engine or a particular database implemented on it, but > either way it's ridiculous. Every production system out there has at > least adequate performance. It may be stunningly inefficient in some > cosmic sense, but computer power is so cheap these days, it's there to > waste. In these times it is no more true. The computer application field is expanding very rapidly into the area of small mobile networked devices. The size counts there. This is a fast growing market segment, which will change many things soon. For example, traditional client-server architectures don't scale there. -- Regards, Dmitry A. Kazakov http://www.dmitry-kazakov.de |