HP3000-L Archives

December 1998, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Simonsen, Larry" <[log in to unmask]>
Reply To:
Simonsen, Larry
Date:
Thu, 3 Dec 1998 07:27:39 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (53 lines)
the tools you mentioned are great.
however, Adager and dbgeneral require exclusive access to the database
and except for mapping of one record to another format do not change the
data in the database.  datanow only uses standard image intrinsics for
updating the data in the database.

perhaps you should look at other ways to improve performance.  perhaps
additional memory or splash compiler to convert code to run in native
mode would better serve your purposes.


> -----Original Message-----
> From: Les bureaux de credit du nord [SMTP:[log in to unmask]]
> Sent: Thursday, December 03, 1998 6:50 AM
> To:   [log in to unmask]
> Subject:      HELP -FAST READ/WRITE TURBO IMAGE ACCESS
>
> My computer programmer would like to know how to read and update
> records of an
> image database using a multiple record access  (in blocks) for fast
> updating in S.P.L.
> language.
>
> From what I understand, records are usually updated one record at a
> time on the
> disks which slows down updating process in batch mode.   Utilities
> such as Adager, DbGeneral, Datanow
> can update records much faster.   Correct me if I'm wrong, but
> read/write block access only applies to
> serial access and nothing can be done about random key access except
> maybe activating autodefer of the image
> database at the risk of jeoperdizing data integrity.
>
>
>
> I'd like to know what general sequence of intrinsics I can use for
> serial updating of an image database
> and what parameters I need to use.
>
>
>
> Our batch programs are all written in S.P.L.   We are looking ways at
> improving the speed at which
> records can be processed.  I figure that if records can be processed
> in memory as a block of data,
> there will be less chances of an IO bottleneck coming from disks.  If
> you could attach an S.P.L. example, I would appreciate it.
>
>
> Jean Huot
> Northern Credit Bureaus Inc.
>

ATOM RSS1 RSS2