Hello,
Just like my other team mates, I've been on the road for the past week. Last week I delivered a DB2 Viper training course for business partners in Brazil, and next week (May 22 - 26), I will be promoting the use of DB2 Express-C at several of the best universities in Brazil. I'm sure they will be impressed by the Viper hybrid data server. If you happen to be in the Sao Paulo area, and would like to attend these events, or meet me in person, send me a note to db2x@ca.ibm.com.
Cheers, Raul.
Sunday, May 21, 2006
Still time for DB2 people to win a 50 inch plasma TV or one of 25 iPods.
I did not want any of our DB2 Express - C community to miss out on a chance to win a really big TV or one of 25 really small iPods. Take a look at my ITToolbox blog post Penny for your thoughts ... or better yet a 50 inch plasma TV or one of 25 iPods. for details of the contest. The important part is not the prizes. This survey gives you a chance to have your opinions heard by the International DB2 Users Group (IDUG) and to influence the programs that IDUG brings to you in the future. If you are not familiar with what IDUG can do for you, take the opportunity to browse their web site. It is a terrific resource for anybody who works with DB2.
Winds of change
I am pleased to announce that DB2 Express - C team is now a part of the DB2 Technical Marketing team. What does this mean? Well, many things will stay the same. Raul, Ian and Ryan will still be blogging here and will continue to facilitate discussion on the DB2 Express - C forum.
You will also notice some changes. DB2 Technical Marketing is a team of product evangelists. Our job is not just teach people how to use DB2 products but to, what I call, demonstrate the art of the possible.
With data servers becoming a mature technology that firmly implanted in the application infrastructure, databases don't get the recognition they deserve. We plan to bring the "cool factor" back to data servers. We have the technology to make this happen. DB2 Viper is changing the landscape of database technology by becoming a hybrid Relational-XML data store. We also have richer Web Services support, RSS, Atom, PHP, Ruby on Rails and other Web 2.0 must-haves. Our focus will be to spread the message and to demonstrate how DB2 servers can deliver value beyond providing a simple data repository.
Oh, and as you may have guessed already, I manage the DB2 Technical Marketing team. My name is Leon Katsnelson and you will be seeing me around here quite a bit.
You will also notice some changes. DB2 Technical Marketing is a team of product evangelists. Our job is not just teach people how to use DB2 products but to, what I call, demonstrate the art of the possible.
With data servers becoming a mature technology that firmly implanted in the application infrastructure, databases don't get the recognition they deserve. We plan to bring the "cool factor" back to data servers. We have the technology to make this happen. DB2 Viper is changing the landscape of database technology by becoming a hybrid Relational-XML data store. We also have richer Web Services support, RSS, Atom, PHP, Ruby on Rails and other Web 2.0 must-haves. Our focus will be to spread the message and to demonstrate how DB2 servers can deliver value beyond providing a simple data repository.
Oh, and as you may have guessed already, I manage the DB2 Technical Marketing team. My name is Leon Katsnelson and you will be seeing me around here quite a bit.
Tuesday, May 16, 2006
DB2 Express-C at Wireless Enterprise Symposium in Orlando, Florida
Hi,
Just like Ryan, I'm away from my desk this week; I'm here in Orlando at the Wireless Enterprise Symposium. I'm working under the "Lotus software" banner here at the showcase, but I'm talking up DB2 and how to use it for your Blackberry mobility backend. Feel free to stop by and chat -- I've got lots of copies of the DB2 Express-C discovery DVD to give away!
Just like Ryan, I'm away from my desk this week; I'm here in Orlando at the Wireless Enterprise Symposium. I'm working under the "Lotus software" banner here at the showcase, but I'm talking up DB2 and how to use it for your Blackberry mobility backend. Feel free to stop by and chat -- I've got lots of copies of the DB2 Express-C discovery DVD to give away!
Monday, May 15, 2006
DB2 Express C at JavaOne conference in San Francisco!
Just a quick note to let everyone know that I'm in San Francisco right now for the JavaOne conference that runs from May 16-19 at the Moscone Center. All the details are at http://java.sun.com/javaone/sf/ .
This is one of the biggest Java developer's conferences in the world, so I'm sure there will be at least a few DB2 Express C users attending. Feel free to drop by the DB2 pedestal and say hi, we'll have some fun stuff as well as some cool demos.
Ryan Chase
This is one of the biggest Java developer's conferences in the world, so I'm sure there will be at least a few DB2 Express C users attending. Feel free to drop by the DB2 pedestal and say hi, we'll have some fun stuff as well as some cool demos.
Ryan Chase
Thursday, May 11, 2006
Easy Linux Desktop and Server Operating Systems - Xandros
Cool - I just learned that not only has Xandros Linux been certified as "Ready for IBM DB2 Software for Linux", meaning that it is officially a supported OS for DB2, but that the current version of Xandros Server includes DB2 Express-C!
This is fabulous -- now we'd like to get Express-C included in Red Hat...and SLES...and Fedora...and Ubuntu...and CentOS...and....
This is fabulous -- now we'd like to get Express-C included in Red Hat...and SLES...and Fedora...and Ubuntu...and CentOS...and....
Thursday, May 04, 2006
Hierarchical Data (DB2 vs. MySQL)
I attended the MySQL User's Conference in Santa Clara last week -- big thanks to the organizers, it was an excellent event. While there, I attended a number of session talks, including one on hierarchical data handling in MySQL, presented by Mike Hillyer. I came away somewhat puzzled by the methodology of the handling of hierarchical data.
The method proposed to handle this kind of data was to used nested sets to encapsulate the data so that it can be processed through SQL query methods. In nested set theory, essentially you surround a root node with numerical values starting at 1, and going up to a given maximum number that allows you to give a left (1) and right (max) number to surround all your child nodes.
For example, if we had a parent with two child nodes, and each of those child nodes had two leaf nodes (leaf nodes are child nodes, but with the smallest possible data element), then the numbers would be:
parent: left => 1, right => 14
child: left => 2, right => 7
leaf: left => 3, right => 4
leaf: left => 5, right => 6
child: left => 8, right => 13
leaf: left => 9, right => 10
leaf: left => 11, right => 12
Then you can programmatically determine leaf nodes by a simple alegbraic formula that states that the leaf nodes are the only children where the left value + 1 = right value.
As interesting an implementation as this is (and you can read a more detailed exampled about it here), it still leaves me confused. Why not use a database that handles XML data trees (which is essentially what a hierarchical data set is) natively? The nested set theory is a much more inefficient means of data storage and retreival, and means that you have to continually rebase your node numbers if you add elements, and therefore you also need to keep re-indexing your database.
What I would recommend is that developers look into the native XML storage facilities of DB2's Viper release and see how much faster both data queries and database administration can be when you can handle hierarchical data in a more straightforward manner.
The method proposed to handle this kind of data was to used nested sets to encapsulate the data so that it can be processed through SQL query methods. In nested set theory, essentially you surround a root node with numerical values starting at 1, and going up to a given maximum number that allows you to give a left (1) and right (max) number to surround all your child nodes.
For example, if we had a parent with two child nodes, and each of those child nodes had two leaf nodes (leaf nodes are child nodes, but with the smallest possible data element), then the numbers would be:
parent: left => 1, right => 14
child: left => 2, right => 7
leaf: left => 3, right => 4
leaf: left => 5, right => 6
child: left => 8, right => 13
leaf: left => 9, right => 10
leaf: left => 11, right => 12
Then you can programmatically determine leaf nodes by a simple alegbraic formula that states that the leaf nodes are the only children where the left value + 1 = right value.
As interesting an implementation as this is (and you can read a more detailed exampled about it here), it still leaves me confused. Why not use a database that handles XML data trees (which is essentially what a hierarchical data set is) natively? The nested set theory is a much more inefficient means of data storage and retreival, and means that you have to continually rebase your node numbers if you add elements, and therefore you also need to keep re-indexing your database.
What I would recommend is that developers look into the native XML storage facilities of DB2's Viper release and see how much faster both data queries and database administration can be when you can handle hierarchical data in a more straightforward manner.
Subscribe to: Posts (Atom)