Very Computer

What is the nature of the data base?  Would you believe we're not
sure?  The amount of data will probably be very large, and it may or
may not be based on the relational model.  Why am I asking such a
vague question?  Because we are trying to develop corporate funding
for a project that is still in the jello stage of conception (you can
see it, and it glistens, but you still can't get a good grip on it).
What we want to do is to be able to talk about existing Unix-based
solutions to very large data sharing problems.

Names we know about:
 Gould -- fast disks, but is there data-sharing software?
 Most other Big Unix Boxes in the world -- ditto above comment.
 Tandem -- is this Unix based?
 Stratus -- is this Unix based,  is this a good database machine?

I'm nervous about posting this, because I expect every Unix box maker
to tell me about their great machines.  That's fine, but are there
proven very-large data-sharing/data-base applications on those
machines?  Are the disks very high performance?  SCSI ports probably
won't hack it.   We will consider both uniprocessor and multiprocessor
solutions.  Non-Unix solutions, while interesting, are not the topic
of this posting.  Please mail to me, don't post.  If others express an
interest, and if the response is informative, I'll post a summary.

Bill O'Farrell, Northeast Parallel Architectures Center at Syracuse University

Bill -

Please get in touch with me if you can.  The return mail path I got
out of news was completely worthless, as usual, and I don't want to
clutter the net with this stuff.  You should be able to get in touch
with me through the uucp paths in my signature - they're the only way,
as far as I know, to get here from anywhere else.

BTW, in case you couldn't guess, we're doing gigabyte-database OLTP
applications on UNIX boxes here at Hopkins...

                                                                 Phil Kos
...!decvax!decuac!\                                   Information Systems
  ...!uunet!mimsy!aplcen!osiris!phil           The Johns Hopkins Hospital
...!allegra!/                                               Baltimore, MD

Quote:> Help! By Friday we need to know if there is a Unix-based box that
> can work as a very high-performance data-base server.  

Mark <>

PS  If anyone would like a full write up of this project send me
your postal address.
<>                                  <>                                  <>
Mark P. Diamond                  {sun, cbosgd, amdahl, mtxinu}!rtech!markd
from Sequent Computer Systems onsite at Relational Technology

Quote:> Help! By Friday we need to know if there is a Unix-based box that
> can work as a very high-performance data-base server.  Yes, there are a

-------------
 Stephen Samuel
  {ihnp4,ubc-vision,vax135}!alberta!edm!steve

--
-------------
 Stephen Samuel                         Disclaimer: You betcha!
  {ihnp4,ubc-vision,seismo!mnetor,vax135}!alberta!edm!steve

Quote:

>Help! By Friday we need to know if there is a Unix-based box that
>can work as a very high-performance data-base server.  Yes, there are a

Quote:>In a recent project with Relational
>Technology a six* processor Symmetry achieved 104 Transactions
>per second running the Debit Credit Benchmark (TP1) on
>a fully sized 1.1G Byte database, at a price performance ratio
>at about 1/8 that of Tandem  (all of this was verfied by
>the independent Codd & Date consulting group).  This is the  
>fastest (by a factor of about three) any UNIX box has achieved
>for this benchmark.

Quote:>RTI, Oracle, Informix and Unify run Sequent
>for their in-house applications.

        Obviously I am biased, but claiming that any computer is "optimum"
for so broad a range of possible uses as relational database applications
seems a little questionable.

Quote:> ...................... In a recent project with Relational
> Technology a six* processor Symmetry achieved 104 Transactions
> per second running the Debit Credit Benchmark (TP1) on
> a fully sized 1.1G Byte database, at a price performance ratio
> at about 1/8 that of Tandem  (all of this was verfied by
> the independent Codd & Date consulting group). ................

> PS  If anyone would like a full write up of this project send me
> your postal address.

    greg pavlov, fstrf, amherst, ny

[the above opinions are strictly mine, if anyone's.]
[the above signature may or may not be repeated, depending upon some
inscrutable property of the mailer-of-the-week.]

I can understand Eric's skepticism because if someone told me 6 months
ago that INGRES would exceed 100 TPS I might have asked them if they
bumped their head on the way to the office.

However, I know what Mark Diamond says is true because I was in the room
with him, along with Tom Sawyer from Codd & Date consulting, when INGRES
hit 104 TPS.

To appease any cynics I'll list the one caveat of the benchmark first:

        * The INGRES system (running on a Sequent Symmetry machine) which
          hit 104 TPS was running a prototype version of RTI's next release.
          As part of normal prototyping activity we asked ourselves "Just
          how fast can this go?"  We convinced Sequent to let RTI use a
          large Symmetry machine, and we were off...

Eric was surprised about the 1+ Gigabyte database size.  In fact, the
benchmark was run with a DebitCredit defined 100 TPS sized database.  Before
continuing, a little background on the DebitCredit benchmark is in order.

DebitCredit is a well-defined standard benchmark and was written in the
late 1970's  by Jim Gray and about 20 other database professionals.  The
paper was eventually published in DATAMATION under the title "A Measure of
Transaction Processing" by the authors "Anon et al."  Rumour has it the
authors wished to remain secret due to flame-ups that occurred after Dave
DeWitt and Dina Bitton wrote their paper on DBMS benchmarking.

DebitCredit is one of three benchmarks described in the paper, and various
degenerate forms of DebitCredit  have become loosely known in the industry
as "TP1."  The problem with TP1, and the ensuing "TPS" (transactions/second)
measurements, is that most vendors size the databse irregularly (i.e. smaller
than DebitCredit defines).  Thus, as Eric points out, when comparing TPS
measurements one is often comparing apples and oranges.

For the Silver Bullet benchmarks, to which Mark refers, the database was
sized at 100 TPS, or 10 Million 100 byte account records, 10,000 100 byte
teller records, and 1,000 100 byte account records.  Thus, a real purist
would rob RTI of the 104 TPS (and grant only 100 TPS) because the database
was sized for 100 TPS.  (If you do the multiplication you'll see that
the account relation alone is 1 gigabyte of data.)

Overall, the benchmark conformed to DebitCredit standards quite well,
including the submission of tranasctions via a network.  However, there
were a few things we didn't do 100% to the DebitCredit spec.  But then
again, we did a couple to exceed the spec.  In any case, the auditor's
report is being published tommorrow so all DebitCredit whizzes can take
a look.

In conclusion, I saw one "pop-off" on the net (flame semi-on) which questioned
the integrity of the auditor since "Codd and Date and RTI have always had
a good working relationship..." I'll reply to that with

        * If we wanted to pay someone to lie we wouldn't have paid
          Codd and Date's rates!  ;-)

        * Mr. Sawyer was the auditor of Tandem's 208 TPS benchmark.

        * I personally hope that he is not on the net to see this random
          * on his character.

        * If you read his report you'll see that he is certainly impartial.

Finally, if you're interested in seeing the benchmark report you can reply
to this message with a postal address and I'll do my best to get you a copy.

Dave Kellogg
ucbvax!rtech!davek (might need a mtxinu before the rtech)

"Hmmm.  We hit 100 TPS, can I go to bed now??"

-- Cs

    ;-D on  ( Bliss is Bliss, Ignorance is Ignorance, I'm happy )  Pardo

Significant performance gains can be had using cpu/io improvements
that these vendors offer.

Pyramid is also offering significant features in a balanced fashion
(cpu, disk i/o and terminal i/o) that INGRES can take advantage of.

No selling here, but INGRES is hot, and can take advantage of the
multiprocessing capabilities that Pyramid and Sequent and others offer.

** Opinions are my own and not necessarily RTI's.

jim Milbery, RTI Technical Support Burlington,MA

Is your database mainly for reads or reads and writes. If its mainly for
reading information, then having a large disk cache, (eg 8MB) will far
outweigh the speed of the disk.

If you really are going to try to support hundreds of users, then one
major problem will be finding a network interface that can reliably
support this many virtual circuits. One of the biggest problems with
all machines seems to be the limit on the number of concurrent sessions.
Not only is a large amount of memory (ie several K) needed per
session, but also one has to consider things like the size of the
ARP tables.

=====================
     //        o      All opinions are my own.
   (O)        ( )     The powers that be ...
  /    \_____( )
 o  \         |
    /\____\__/