Conference ulysse::rdb_vms_competition

1089.0. "Rdb Press Release" by 32250::VOELKER () Wed Feb 19 1992 15:10

    
Subj:	Feb. 19 release

    

Diane Voelker
(603) 884-3899


         DIGITAL'S Rdb/VMS DATABASE PERFORMS 193.76 TRANSACTIONS 
              PER SECOND RUNNING THE TPC BENCHMARK B 
          
          PERSONAL USE LICENSING NOW AVAILABLE FOR Rdb/VMS Version 4.1


MAYNARD, Mass. -- February 19, 1992 -- Digital Equipment Corporation
released performance results today for version 4.1 of Rdb/VMS, which was
announced last year.  Using the industry-standard benchmark, TPC
Benchmark B, which measures disk input/output, system and application
execution time and transaction integrity, Rdb/VMS V4.1 attained a rate
of 193.76 transactions per second (tpsB) with price/performance
calculated to be $5,929 per tpsB on a VAX 6000 model 560.  The results
were audited by the independent consulting firm of KPMG Peat Marwick and
reported to the Transaction Processing Performance Council (TPC). 

"This performance demonstrates that we are ahead of the competition on
the VAX VMS platform.  Rdb/VMS V4.1 offers the biggest performance boost
of any release in the history of the product," said Vickie Farrell,
Digital's NAS Product Marketing Manager.  "The big win for our customers
is that in order to realize performance gains for their applications,
they just have to install version 4.1.  No database redesign or
application tuning is necessary,"  added Farrell. The biggest benefit
will be seen by users with write-intensive applications. 

The TPC Benchmark B is a fully specified benchmark standard and is the
result of Digital and other members of the TPC working together to
create an industry-wide standard for evaluating the performance and
price/performance of transaction processing and database systems. TPC
Benchmark B was approved by the members of the TPC in August, 1990. The
configuration and results of the Rdb/VMS test are given below. 


System                                       TPS      Price per TPS
Configuration   CPU Model    Software       (tpsB)     ($/tpsB)
--------------------------------------------------------------------------
                                                                 
VAX 6000        560          Rdb/VMS V4.1    193.76     $5,929
                             VMS V5.4 


Digital also announced the Personal Use Licensing option for Rdb/VMS
V4.1, which means that a customer only needs to purchase the software
for the specific users of the product.  On October 30, Digital announced
Personal Use Licensing of more than 120 layered software products and
applications.  The licenses are assigned at a user's site and may be
reassigned as needed. Today's announcement further extends the range of
products that now include Personal Use Licensing.  The U.S. price for
the full development Personal Use License is $7,500; the interactive
Personal Use License is $2,500. 

"With the new Rdb/VMS Personal Use License option, Digital is providing
users with the freedom and flexibility to choose licenses based on the
way the software is used, not just on the size and configuration of the
computing resources.  This feature is key to customers being able to take
full advantage of the emerging client/server computing environment,"
said David Stone, vice president of Digital's Software Products Group. 
"When you consider the latest TPC Benchmark B performance numbers and
the new Personal Use Licenses, Rdb/VMS V4.1 is the clear choice for
customers who want a high-quality, high-performance database at a great
value," Stone added. 
                
Digital Equipment Corporation, headquartered in Maynard, Massachusetts,
is the leading worldwide supplier of networked computer systems,
software and services.  Digital pioneered and leads the industry in
interactive, distributed and multivendor computing.  Digital and its
partners deliver the power to use the best integrated solutions - from
desktop to data center - in open information environments. 
           
                             #### 
     
          
Note to Editors:  Digital, Rdb/VMS, VAX, VAXcluster, VAX/VMS, VAX 6000, 
                  and VMS are trademarks of Digital Equipment Corporation.

                  TPC Benchmark and TPC-B are trademarks of the
                  Transaction Processing Performance Council.

CORP/92/489

    
          <<< DATABS::$255$DUA25:[NOTES$LIBRARY]IM_PARTNERS.NOTE;1 >>>
                                -< IM Partners >-
================================================================================
Note 293.9                        TPC-B Results                           9 of 9
DATABS::NEEDLEMAN "today nas/is, tomorrow..."        91 lines  18-FEB-1992 15:28
                            -< Press release 2/19 >-
--------------------------------------------------------------------------------
    Here is the press release. Please note the release date
    
Subj:	Feb. 19 release

    

Diane Voelker
(603) 884-3899


         DIGITAL'S Rdb/VMS DATABASE PERFORMS 193.76 TRANSACTIONS 
              PER SECOND RUNNING THE TPC BENCHMARK B 
          
          PERSONAL USE LICENSING NOW AVAILABLE FOR Rdb/VMS Version 4.1


MAYNARD, Mass. -- February 19, 1992 -- Digital Equipment Corporation
released performance results today for version 4.1 of Rdb/VMS, which was
announced last year.  Using the industry-standard benchmark, TPC
Benchmark B, which measures disk input/output, system and application
execution time and transaction integrity, Rdb/VMS V4.1 attained a rate
of 193.76 transactions per second (tpsB) with price/performance
calculated to be $5,929 per tpsB on a VAX 6000 model 560.  The results
were audited by the independent consulting firm of KPMG Peat Marwick and
reported to the Transaction Processing Performance Council (TPC). 

"This performance demonstrates that we are ahead of the competition on
the VAX VMS platform.  Rdb/VMS V4.1 offers the biggest performance boost
of any release in the history of the product," said Vickie Farrell,
Digital's NAS Product Marketing Manager.  "The big win for our customers
is that in order to realize performance gains for their applications,
they just have to install version 4.1.  No database redesign or
application tuning is necessary,"  added Farrell. The biggest benefit
will be seen by users with write-intensive applications. 

The TPC Benchmark B is a fully specified benchmark standard and is the
result of Digital and other members of the TPC working together to
create an industry-wide standard for evaluating the performance and
price/performance of transaction processing and database systems. TPC
Benchmark B was approved by the members of the TPC in August, 1990. The
configuration and results of the Rdb/VMS test are given below. 


System                                       TPS      Price per TPS
Configuration   CPU Model    Software       (tpsB)     ($/tpsB)
--------------------------------------------------------------------------
                                                                 
VAX 6000        560          Rdb/VMS V4.1    193.76     $5,929
                             VMS V5.4 


Digital also announced the Personal Use Licensing option for Rdb/VMS
V4.1, which means that a customer only needs to purchase the software
for the specific users of the product.  On October 30, Digital announced
Personal Use Licensing of more than 120 layered software products and
applications.  The licenses are assigned at a user's site and may be
reassigned as needed. Today's announcement further extends the range of
products that now include Personal Use Licensing.  The U.S. price for
the full development Personal Use License is $7,500; the interactive
Personal Use License is $2,500. 

"With the new Rdb/VMS Personal Use License option, Digital is providing
users with the freedom and flexibility to choose licenses based on the
way the software is used, not just on the size and configuration of the
computing resources.  This feature is key to customers being able to take
full advantage of the emerging client/server computing environment,"
said David Stone, vice president of Digital's Software Products Group. 
"When you consider the latest TPC Benchmark B performance numbers and
the new Personal Use Licenses, Rdb/VMS V4.1 is the clear choice for
customers who want a high-quality, high-performance database at a great
value," Stone added. 
                
Digital Equipment Corporation, headquartered in Maynard, Massachusetts,
is the leading worldwide supplier of networked computer systems,
software and services.  Digital pioneered and leads the industry in
interactive, distributed and multivendor computing.  Digital and its
partners deliver the power to use the best integrated solutions - from
desktop to data center - in open information environments. 
           
                             #### 
     
          
Note to Editors:  Digital, Rdb/VMS, VAX, VAXcluster, VAX/VMS, VAX 6000, 
                  and VMS are trademarks of Digital Equipment Corporation.

                  TPC Benchmark and TPC-B are trademarks of the
                  Transaction Processing Performance Council.

CORP/92/489

    
    

    OK,
    
    This is great stuff and we all should congratulate the engineering
    team. I wonder how Oracle and their auditors Codd&Date feel, because
    they measured and submitted 60.1 tpsB to the TPC AND GOT IT ACCEPTED
    on the same hardware (153 tpsB for Oracle). Or does global buffering
    and undo-redo make Rdb/VMS suddenly 3 times faster?
    
    Any idea's for ad's in the newspapers with this?
    
    Thanks,
    Henny Olthof, TP-DB Netherlands

�       <<< Note 1089.1 by IJSAPL::OLTHOF "Henny Olthof @UTO 838-2021" >>>
�                       -< What did Oracle measure then? >-

Henny,

The database that was used for this benchmark did have fast commit (aka 
undo-redo) but did not use global buffering.

Probably more significantly, we used ACMS as a front end to handle
contention while Oracle handles all contention within the database so the
implementation model used between Digital and Oracle is fairly different. 
However, the K$TPS does accurately reflect the cost of ACMS as part of the
total solution/benchmark. 

Ummm, I don't believe this test used ACMS.

Hal

    
    weas this the test that had 3100's at the front end?
    
    I read about a benchmark for rdb on the 6000 that in truth had some
    3100's doing some front-end work.
    
    Simon

	Re .4

	Simon,

	No, this test was a single system test. There were no front-end systems
	used. Also, ACMS was NOT used (to appropriately route the transactions)
	in this test; instead, an internal driver was used.

	Front-ends are only used for TPC-A like tests, where one needs to 
	emulate terminal I/O. We found that by doing this, not only the 
	performance improved slightly, but the price/performance improved
	substantially, since the cheaper and smaller machines were doing
	all the front-end work. This is perfectly legal from the specification's
	point of view. All our TPC-A tests have been done with frond-end 
	systems.

Re:  .3 - right, no ACMS (this was a mis-communication between me and one
		of the folks who ran the benchmark)

Re:  .4 - I don't believe that this benchmark used 3100's (or anything
		else) for front ends...

    
    we price in development licenses and hardware on a small VAX and
    runtime software on the 6560. This lowers our cost of development on
    the 6560. User based licensing may change this in the future.

    Barry