home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
SunSoft Catalyst CDWARE 1996 May to August
/
Catalyst CDWARE 1996 May to August.iso
/
.products
/
IBM_Canada
/
djspec.html%
< prev
next >
Wrap
Internet Message Format
|
1996-02-27
|
11KB
From redmond@VNET.IBM.COM Wed Nov 22 07:40 PST 1995
Date: Wed, 22 Nov 95 10:38:16 EST
From: "Patrick Redmond" <redmond@VNET.IBM.COM>
To: jasoni@cdware.Eng.Sun.COM
Subject: IBM software pages 7
<!doctype html public "-//IETF//DTD HTML//EN">
<HTML>
<head>
<title>IBM DataJoiner Spec Sheet</title>
<meta name="abstract" content="IBM DataJoiner Specification Sheet>
<meta name="revision" content="1.0">
<meta name="owner" content="jasonk@vnet.ibm.com">
<meta name="keywords" content="datajoiner">
<meta name="review" content="REVIEW_961231">
<meta name="security" content="public">
</head>
<BODY>
<img src="../../images/mastsoft.gif" alt="IBM Software"><br>
<h1><a name="top">IBM DataJoiner</a></h1>
<h2><I>A New Class of Data Access Middleware</I></h2>
<p>Table of contents:
<ul>
<li><a href="#need">The Need for DataJoiner</a>
<li><a href="#reas">Four Good Reasons to Buy DataJoiner</a>
<li><a href="#opti">DataJoiner Optimizer</a>
<li><a href="#vend">Vendor Support</a>
<li><a href="#stds">Standards and Client Platforms</a>
<li><a href="#edu">DataJoiner Education, Consulting, and Services</a>
<li><a href="#concl">Conclusion </a>
</ul>
<HR>
<h3><a name="need">The Need for DataJoiner</a></h3>
<P>
Admit it. You have data scattered from here to there on sources that can't talk to each other. You can usually access the
information you need everyday, along with the few items you need from time to time. But what about solving a really tough problem,
one that requires you to pull information from a variety of sources all at once? It might be left alone, not because it's
unnecessary, but because it requires droves of programmers to figure out where the information is located and how you can use it
once you get it. Maybe you can't use it because your marketing department has Oracle but customer service uses DB2. Good luck.
<P>
Every organization from those in home offices to those in corporate headquarters searches for a competitive advantage. Finding daily
operations data efficiently is no longer a competitive advantage. The best companies today can access any data they want, when they
want it, and use it the way they need to. That's today's competitive advantage and one not to be overlooked. Ignorance might be
bliss in poetry, but it's death in business.
<P>
That's why IBM developed DataJoiner. It's a new class of database middleware that lets you join data from different sources with a
single SQL statement and a single interface. That single interface hides all the differences you struggle with when accessing
information from around your business. Even your application doesn't have to know the differences. One interface to one database
image. It's that simple. You don't even have to know where the data resides.
<p>
<center><img src="djart1.gif"></center>
<p>
<HR>
<h3><a name="reas">Four Good Reasons to Buy DataJoiner</a></h3>
<P>
<UL>
<LI>DataJoiner makes queries more powerful than ever. You don't have to send ten different queries to ten different data sources to
get an answer with DataJoiner. One query can access all ten sources, find the data, and join it together in front of you. Those ten
sources can be IBM or non-IBM, relational or non-relational, local or remote. Database gurus call this heterogeneous access. You'll
call it awesome, and it is. It's so awesome, in fact, that when we demonstrate it at tradeshows, people don't believe that we're
accessing live data. They ask us if we pre-loaded the demo on our machines before we came, labeling one data source Oracle, another
Sybase, and another DB2. Nope. It's all live, and we join it right there at the booth. Honest.
<LI>DataJoiner makes application development easy. There's only one API to deal with. You tell your application how to talk to
DataJoiner and DataJoiner talks to all the different databases in your business. We call this transparency. The data locations, SQL
dialects, networking protocols, operating systems, data types, error codes, and functional differences are transparent to your
application. Which means you don't have to worry about them.
<LI>DataJoiner strengthens your replication strategy. Using DataJoiner in conjunction with IBM Replication Solution Products, you
can replicate data from Oracle to DB2, and DB2 to Sybase, among many other possibilities. DataJoiner provides access to current as
well as point-in-time data at diverse, remote data sources. It's like buying a handful of new CDs. Same great stereo system, more
music to play.
<LI>DataJoiner provides a full-strength relational database. The power of IBM's DB2/6000 database is the launchpad for DataJoiner's
middleware capabilities. We call this DB2/6000 functionality DataJoiner's local database. You can use it to store query results,
develop and manage snapshots of data from different sources, and support concurrent transaction processing. The local database
offers full backup and recovery, logging, and locking subsystems. No extra charge.
</UL>
<P>
"Wow," you might say after reading all this, "that's just like having one big database." Exactly.
<p>
<center><img src="djart2.gif"></center>
<p><a href="#top">[top]</a>
<HR>
<h3><a name="opti">DataJoiner Optimizer</a></h3>
<P>
If you're impressed by how much DataJoiner can do, wait until you see how fast it can do it. IBM is known for its optimizer
technology and DataJoiner development infused the company's most advanced techniques inside DataJoiner. Some middleware products
have an optimization strategy that consists of materializing each of two tables to be joined on a platform containing the middleware
code. The join occurs there. But wait a minute. Take a moment to consider that one of the tables might contain 5,000,000 rows, while
the other might only contain 10,000 rows. Doesn't it make sense to just materialize the smaller table? It might. Better yet, perhaps
only portions of each table need to be materialized. Some optimizers won't figure this out. Ours will.
<P>
Our optimizer doesn't just look at the number of rows in the two tables to be joined, nor is it satisfied with a simple index
strategy. Instead, the DataJoiner optimizer takes into consideration the relative CPU speed of each data source, the relative IO
speed, and the relative network bandwidth when formulating a fast solution to your query. It combines these factors with statistics
you supply on each data source to blaze a path to the information you need.
<P>
DataJoiner's optimization strategy can lead to improvements measured in days, not hours or minutes. Think about that. DataJoiner
could give you answers by lunchtime. Something else might not give you answers until you drive home that night, watch television,
dream about your query, shower the next morning, eat breakfast, and drive back to work. Hope it wasn't urgent.
<p><a href="#top">[top]</a>
<HR>
<h3><a name="vend">Vendor Support</a></h3>
<P>
"Okay," you're thinking, "who can DataJoiner really talk to?" Glad you asked. DataJoiner supports relational and non-relational data
sources, including:
<P>
<UL>
<LI>The DB2 Family, including:
<p>
<UL>
<LI>DB2 for MVS
<LI>DB2 for VSE and VM (SQL/DS)
<LI>DB2 for OS/400
<LI>DB2 for AIX
<LI>DB2 for OS/2
<LI>DB2 for HP-UX
<LI>DB2 for Solaris Operating Environment
<LI>DB2 Parallel Edition
</UL>
<LI>Oracle
<LI>Sybase
<LI>Microsoft SQL Server
<LI>IMS (through gateway)
<LI>VSAM (through gateway)
</UL>
<P>
Additionally, DataJoiner includes a generic access API that allows you to use existing drivers and/or create new drivers to gain
access to an unlimited set of data sources.
<p><a href="#top">[top]</a>
<HR>
<h3><a name="stds">Standards and Client Platforms</a></h3>
<P>
DataJoiner supports a range of standards including ODBC and XOPEN CLI. Clients can access DataJoiner from any of the DB2 Client
Application Enabler platforms such as OS/2, Windows, and UNIX. Also, you can use one of several popular programming languages like
C, FORTRAN, and COBOL to take advantage of DataJoiner's single-database API.
<P>
DataJoiner works with hundreds of tools that support the IBM Client Application Enabler API. Now you can extend the power of
familiar tools like Microsoft Access, IBM Visualizer, IBM Visual Age, and PowerSoft PowerBuilder to work with data across your
business through the DataJoiner interface.
<p><a href="#top">[top]</a>
<HR>
<h3><a name="edu">DataJoiner Education, Consulting, and Services</a></h3>
<P>
Want to learn how to use DataJoiner? We'll teach you all about it. IBM offers customer education for DataJoiner, with classes
available now.
<P>
We can even come to your place to configure your existing systems and install DataJoiner. IBM offers two phases of DataJoiner
consulting:
<P>
<UL>
<LI>DataJoiner Planning Phase. This service helps you plan for the installation and configuration of DataJoiner and the
configuration of your network systems. Issues addressed include an assessment of general readiness, client and data source
definitions, and a strategy for backup and recovery.
<LI>DataJoiner Implementation Phase. This service gets DataJoiner up and running. Steps include installing DataJoiner and
configuring access to data sources, installing and configuring remote clients, and documenting your environment.
</UL>
<p>
<center><img src="djart3.gif"></center>
<p><a href="#top">[top]</a>
<HR>
<h3><a name="concl">Conclusion</a></h3>
<P>
So that's DataJoiner. It accesses your data, IBM or non-IBM, relational or non-relational, local or remote, like it was all part of
one big database. It makes your queries more powerful than ever, makes your application development easy, strengthens your
replication strategy, and provides a full-strength relational database. And thanks to its intelligent optimizer, DataJoiner gets
your work done in a hurry.
<P>
It's a whole new class of data access middleware.
<P>
For more information about DataJoiner, including information about education and services,
contact your IBM authorized software reseller or IBM marketing representative.
</BODY>
<!-- end ibmbody--------------------------------------------------------- -->
<HR>
<b><pre>
<A HREF="../../index.html"><IMG BORDER=0 SRC="../../images/home.gif" alt="Software home page"></A> <A
HREF="http://sq.ibm.com/sq/"><IMG BORDER=0 SRC="../../images/sqx.gif" alt="Software Quarterly"></A> <A HREF ="../../dl.html"><IMG
BORDER=0 SRC="../../images/download.gif" alt="Download"></A> <A HREF="mailto:ibmsoftware@vnet.ibm.com"><IMG BORDER=0
SRC="../../images/feedback.gif" alt="Feedback"></A> <A HREF="../../software/index_all.html"><IMG BORDER=0
SRC="../../images/index.gif" alt="Index"></A>
<A HREF ="../../index.html"><IMG BORDER=0 SRC="../../images/l_homex.gif"></A><A HREF="http://sq.ibm.com/sq/"><IMG BORDER=0
SRC="../../images/l_sqx.gif"></A><A HREF ="../../dl.html"><IMG BORDER=0 SRC="../../images/l_downx.gif"></A><A
HREF="mailto:ibmsoftware@vnet.ibm.com"><IMG BORDER=0 SRC="../../images/l_feedx.gif"></A><A HREF="../../software/index_all.html"><IMG
BORDER=0 SRC="../../images/l_indexx.gif"></A>
</pre>
<br>
[
<a href="datajoin.html">IBM DataJoiner home page</a>
]
<br>
[
<a href="http://www.ibm.com/">IBM home page</a> |
<a href="http://www.ibm.com/Orders/">Order</a> |
<a href="http://www.ibm.com/Search">Search</a> |
<a href="http://www.ibm.com/Assist/">Contact IBM</a> |
<a href="http://www.ibm.com/Finding/">Help</a> |
<a href="http://www.ibm.com/copyright.html">(C)</a> |
<a href="http://www.ibm.com/trademarks.html">(TM)</a>
]
</b>
</body>
</html>