ABAP Freak Show – June 17th – Large Database Objects in 7.02


This episode of the ABAP Freak Show looks in detail at large database objects – STRINGS and RAWSTRINGS. Although we have had the ability to use unlimited length STRING and RAWSTRING fields in database tables in ABAP since WebAS 6.10, NetWeaver 7.0 Enhancement Package 2 brings several new features which help to optimize access to these fields. In this eLearning we study both the new Locator and Stream object for reduced resource consumption when working with just such large fields from the database.

You can download a high resolution version MP4 version of the video here:
http://blip.tv/file/get/Thomas_jung-AFSJune17th2009LOBStreamsLocators496.mp4

Tags: , ,

2 responses to “ABAP Freak Show – June 17th – Large Database Objects in 7.02”

  1. Simon Kemp says :

    Hi Thomas,

    You mention in this that you are offloading some of the workload onto the database so although you are freeing up memory on the application server you may be pushing that onto the database. I was wondering if there was some underlying database technology that allowed you to do this now that was not available previously?

    Thanks,
    Simon

  2. Thomas Jung says :

    >You mention in this that you are offloading some of the workload onto the database so although you are freeing up memory on the application server you may be pushing that onto the database.

    Yes and no. Keep in mind that if you were to read the entire STRING/XSTRING directly you would still consume resources on the database server to get the data to the Application server. Using these new techniques we remove the resource usage on the application server and the need to transmit the data from the database server, but still need resources on the database server. So for the most part we aren’t necessarily offloading work to the DB as much as we are limiting the work to only the half that needs to be done on the DB. It can however mean that while the same amount of resources are needed on the DB, that they can be retained for slightly longer since they aren’t released as soon as the transfer to the application server occurs.

    >I was wondering if there was some underlying database technology that allowed you to do this now that was not available previously?
    Perhaps. We don’t really release much information about how the kernel DB interface works. All of this functionality is done within the kernel. Even the methods of the locator and stream classes are just kernel methods with no ABAP code in them. We might very well be taking advantage of database vendor specific optimizations behind the scenes. However to retain database independence, we never implement major features in ABAP and Open SQL until they can be supported by all the Databases that run under our App Server.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: