OpenLaszlo has some of the most striking examples of cinematic UIs on the web today with some of the most promising technological premises I have seen. I just wanted to put out what I see as the most significant differentiators for OpenLaszlo from other RIA technologies...
Unlike its most potent competitor Flex, OpenLaszlo is not strictly tied down to Flash. The important thing to realize about OpenLaszlo is that it is an XML based UI description language with a dynamic rendering engine that is plug-able. This means that while currently there is mature and capable Flash based engine and a slighty less capable HTML+DHTML aka AJAX engine, these are not the only ones possible(Silverlight?). This, together with its fairly mature data manipulation capabilities, very complete widget library, good documentation and user community size - makes it one of the best candidates for developing attractive UIs quickly.
A problem that I saw though was that there seemed to be some significant problems around seemingly crucial areas - e.g. with SOAP based interactions, debugging in Javascript mode, file uploading, authentication etc for which there seemed to be no authoritative solutions. These are important for any serious real-world applications and OpenLaszlo needs to put out firm recommendations, solutions etc that will keep developers from bouncing around the internet.
All-in-all, this is a really neat open-source alternative to an Adobe based platform leading to the advantages of Flash and AJAX - I look forward to positive changes in this area...
As an active developer, I feel the benefits of open software every day. At the same time, having worked for several large corporations, I'm constantly reminded of the vast chasm dividing those of us who believe in the power of 'open', from those who think it is just another nutty hobby. This weblog is a running commentary on my experiments with various open products and libraries in an attempt to bridge this gap...
Monday, December 15, 2008
Tuesday, December 2, 2008
HammerORA
The Transaction Processing Performance Council (TPPC) provides some of the most comprehensive and industry standard techniques to test/benchmark database/application performance. However, I realized that TPPC only provides the specifications for the standard tests, not their implementations. There are commercial implementations available, but here's , my experience so far with HammerORA.
This free tool implements a large part of the TPC-C benchmarking tests apart from providing some other powerful Oracle bashing treats as well. A lot of articles are available on the internet - some of the following got me going part of the way:
1) It seems necessary to install hammerora inside the Oracle (Oracle Express in my case) installation folders or set up an ORACLE_HOME environment variable. In my case having installed Oracle Express 10g under the C: drive my ORACLE_HOME environment variable needed to be set to "C:\oraclexe\app\oracle\product\10.2.0"
2) The target Oracle RAC instance needs to be correctly set up in the Oracle tnsnames.ora file.
3) The latest version of hammerora may not need the tablespace, user etc to be set up by hand, but if people do need to do it, and if you're following Dave's instructions, then for his 'Create Schema' section you need to include the tablespace creation for TPCC. Dave's handled TPCH correctly I think.
This free tool implements a large part of the TPC-C benchmarking tests apart from providing some other powerful Oracle bashing treats as well. A lot of articles are available on the internet - some of the following got me going part of the way:
- http://www.ba6.us/node/82
- http://www.otmfaq.com/forums/blogs/chrisplough/11-benchmarking-part-2-oracle-db-performance-hammerora.html
1) It seems necessary to install hammerora inside the Oracle (Oracle Express in my case) installation folders or set up an ORACLE_HOME environment variable. In my case having installed Oracle Express 10g under the C: drive my ORACLE_HOME environment variable needed to be set to "C:\oraclexe\app\oracle\product\10.2.0"
2) The target Oracle RAC instance needs to be correctly set up in the Oracle tnsnames.ora file.
3) The latest version of hammerora may not need the tablespace, user etc to be set up by hand, but if people do need to do it, and if you're following Dave's instructions, then for his 'Create Schema' section you need to include the tablespace creation for TPCC. Dave's handled TPCH correctly I think.
Labels:
benchmark testing,
hammerora,
oracle testing,
performance,
tpc-c
Subscribe to:
Posts (Atom)