spilon

Christmas Gifts for JDBC and Web Service Developers

Blog Post created by spilon on Feb 9, 2010

Hello JDBC and Web Service developers!

 

Some of you might have noticed we released PI JDBC 2010 R2 and PI Web Services 2010 R2 around Christmas time, and I would like to talk a little more about these. (I recognize this post is a little late, but I'm sure the beginning of the year was pretty hectic for all of us  )

 

The 2010 R2 releases are incremental releases to the "2010" versions released in August 2010 - they broaden the support for information contained in the PI Asset Framework (AF) and provide a few bug fixes and enhancements. Let me go through the details for each of them.

 

 

 

PI JDBC 2010 R2
The main change for this release is the change from the classic PI OLEDB Provider (PI-centric) to PI OLEDB Enterprise (AF-centric) as the underlying data access layer, which implies a whole new table set and new functionality are available. Note, this release is backward compatible and using the same settings as in PI JDBC 2010 will fall back to the previous, classic PI OLEDB Provider behavior. Here's how:

 

As you may already know, JDBC uses database URLs to connect to databases. The URL for PI JDBC needs to provide the name of the PI SQL Data Access Server (DAS) machine and then some PI OLEDB Enterprise (or classic PI OLEDB) connection string. In this 2010 R2 release and going forward, you should use:
     jdbc:pisql://pisqldas_server/pioledbenterprise_connectionstring

 

As an example, to use the PI OLEDB Enterprise table set and connect to the PI Asset Framework (AF), you could use this URL:
    jdbc:pisql://mySQLDAS/Data Source=myAFServer; Integrated Security=SSPI;

 

If you want to revert back to the classic PI OLEDB Provider table set and connect to the PI Data Archive, you would use:
    jdbc:pi://pisqldas_server/pioledb_connectionstring

 

For example:
    jdbc:pi://mySQLDAS/Data Source=myPIServer; Integrated Security=SSPI;

 

 

 

The 2010 R2 release also includes several other enhancements and bug fixes, including significant performance improvements for writing time-series data and support for multiple connections from multiple threads; please consult the Release Notes for more details. One thing I would like to bring to your attention, however, is a significant performance enhancement with regards to writing data into PI. Indeed, the 2010 R2 version implements a JDBC interface called AddBatch which does batch processing of the write operations. Combined with the FLUSH option (see the "PI OLEDB SQL Optimization" White Paper for more details), it really pushes the limits of what you can do with PI JDBC. Our good friend Andreas Schremmer is currently working on a White Paper about this... you can expect this within the next couple weeks

 

 

 

PI Web Services 2010 R2
Just like for PI JDBC, PI Web Services 2010 R2 brings new data retrieval features, particularly wider adoption of the PI Asset Framework (AF). PI Web Services 2010 R2 offers new features including the following:
     - New methods for basic PI tag search and snapshot retrieval
     - Extended AF Data Reference support
     - Extension of filter expression support
     - Support for additional data insertion modes

 

I talk about these a little more in the "PI Web Services 2010 and Beyond" (see under the Auditorium), but once again you can consult the Release Notes for more details on this particular release.

 

 

 

Conclusion
We hope these changes help you in your projects, and contribute to making more out of your PI System (including making it an "asset-powered" one )

 

In closing, I look forward to hearing from you about what you think is important we work on next - simply comment on this blog post or initiate new discussion threads in the appropriate forum under Discussion Hall.

Outcomes