Mark Rittman's Oracle Weblog
This is the weblog for Mark Rittman, a developer working on Oracle Data Warehousing technology based in Brighton, England. You can contact me at mark@rittman.net.
        

04 August 2003

One of the new features that's expected to be included in Oracle 10G is HTML DB, a new rapid application development environment for Oracle web-based applications.

Previously known as Project Marvel, details of what is now called HTML DB are available at the Project beta website, together with an overview of the technology and development process used in developing Marvel applications.

HTML DB appears to be different to working with Portal, JDeveloper and the PL/SQL Toolkit, in that you work within a declarative environment (i.e. it's point and click) using three modules - Flow Builder, SQL Workshop, and Data Workshop - to build your applications.

It looks like it's aimed at existing Oracle PL/SQL developers who wish to take advantage of web technologies such as Web Services, XML, SOAP and XSL, wrapped up in an environment that seems a little like a web-based version of Oracle Designer.

All development on HTML DB is carried out online and up until now, to get to use the product you had to use a hosted service made available by Oracle and then download a runtime version to run locally; however, it looks like the final version, now known as HTML DB, will be available bundled with the 10G database server


8:36:44 PM    

Ronan Miles, UK Oracle User Group Chairman, on Oracle 10G:

"I think grid computing will do for the computer what disk farms have done for disks"

"Once, we were in awe of one disk and now it's a resource in a Raid array. Grid can take processing and memory down the same road. You [will] have a computing surface you just use."

"There has been concern about the degree of investment Oracle has put into applications," said Miles.

"With 10G, it is indicating its commitment to database technology. If the next product had been 10i, it would have sounded like it couldn't think of anything other than bigger, faster, better.

"Whether [the technology] really works or not doesn't matter because it will do by 11G."

More from the article at VNUnet.com


8:21:06 PM    

Mark Van De Wiel at Oracle has pointed out a new viewlet demo of the OLAP functionality in OWB9.2.

It uses an early version of the upcoming 9.2.0.4 database patch (a.k.a. the 9.2.0.3x patch) that enables the 9i OLAP cube and dimension load process from OWB, and shows how a 9i OLAP cube can be created and populated using OWB, with the resultant Analytic Workspace displayed through a BI Beans JDeveloper project.


7:00:44 PM    

Nigel Pendse's OLAP Report gives his view on the forthcoming purchases of Crystal Decisions by Business Objects, and Brio Software by Hyperion Solutions.

In Nigel's view, the Crystal Decisions / Business Objects deal is more to do with a defensive move on both sides to avoid the threat of Microsoft Reporting Services t to Crystal’s core reporting market and Cognos’ increasing strength against Business Objects, and it has parallels with the merger between Hyperion and Arbor that was largely driven by the immenent release of Microsoft SQL Server 7 and OLAP Services.

He's more enthusiastic though about the Brio / Hyperion merger, noting that there is little product overlap between the companies' products, they share a similar geographic location and company culture, and the Brio tools fill a relational reporting gap in Hyperion's product mix. He also notes that Hyperion have learnt many lessons from their merger with Arbor, which was generally seen as a bit of a disaster, and they've made several smaller aquisitions since that have worked out well.


5:10:05 PM    

Roby Sherman has recently published a technical comparison of Oracle 9i release 2 running on RedHat Linux 7.2 and Windows 2000 Server, on identical, Pentium-based hardware.

A series of RDBMS tests were coded to produce a mix of I/O characteristics similar to those found in typical, single instance OLTP, DSS and hybrid environments. These tests were specifically designed to stress the database instance in areas such as Parallel / Bulk DML, Large transactional and query performance against range, hash, and list-partitioned tables, Redo generation (and subsequent LGWR flushing), and Parallelized table scans, joins, and intercommunication.

The result? RedHat Linux 7.2 demonstrated an average performance advantage of 38.4% compared to Windows 2000 Server in a variety of operational scenarios. The study didn't take account of clustering and it probably didn't represent the maximum performance you could get out of each platform (and wasn't there an issue with mainstream versions of Linux not being able to handle asynchronous I/O, which was fixed in RedHat Advanced Server 2.1?) but it's an interesting read nonetheless.

On the same theme, there's also a paper by Jay Houghton on ittoolbox.oracle.com on "Building a Scalable Low-Cost Data Warehouse Using Oracle9iR2 on Windows 2000 Advanced Server". It concentrates more on enabling the Advanced version of Windows 2000 Server to address more than 2GB of memory, and how to take advantage of the new ETL functions in 9i, with a view to opening up this commodity platform for large-scale data warehouse use. I wonder how Advanced Server would have stacked up in the previous comparison.


2:27:52 PM    

© Copyright 2003 Mark Rittman.
 
August 2003
Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31            
Jul   Sep






Click here to visit the Radio UserLand website.

Subscribe to "Mark Rittman's Oracle Weblog" in Radio UserLand.

Click to see the XML version of this web page.

Click here to send an email to the editor of this weblog.