Hi Sean, Thanks! I've made note of this, and will change it via https://issues.apache.org/jira/browse/TRAFODION-2799.
Dave -----Original Message----- From: Sean Broeder [mailto:[email protected]] Sent: Tuesday, November 7, 2017 1:05 PM To: [email protected] Subject: RE: Question about transaction manager on Trafodion architecture page Hi Dave, Thanks for pointing this out. I think stating a specific release of HBase support in the architecture is a mistake because it's just begging to become obsolete. Perhaps a better replacement for the first bullet is something like "Upgraded to support HBase coprocessor mechanism" because this will not likely change in the near future. I think the third bullet remains true. We do support global transactions that span multiple rows, regions and tables within an HBase cluster. Thanks, Sean -----Original Message----- From: Dave Birdsall [mailto:[email protected]] Sent: Tuesday, November 7, 2017 8:49 AM To: [email protected] Subject: Question about transaction manager on Trafodion architecture page Hi Trafodioners, On the Trafodion architecture page (http://trafodion.apache.org/architecture-overview.html), in the section on Transactions, there is the following text: Trafodion supports distributed ACID transaction semantics using the Multi-Version Concurrency Control (MVCC) model. The transaction management is built on top of a fork of the HBase-trx project implementing the following changes: * Upgraded it to work on HBase version 0.98.1 (for CDH 5.1) or 0.98.0 (for HDP 2.1). * Added support for parallel worker processes doing work on behalf of the same transaction. * Added support for global transactions, that is, transactions that can encompass resources (regions/HTables) across an HBase cluster. * Added transaction recovery after server failure. I believe the first and third bullets above are out of date. Can you suggest up-to-date text for these? I will then update the web page. Thanks, Dave
