The TPC held a General Council meeting on August 10th in Burlingame, CA. On the OLTP front, the TPC-E subcommittee continued its review of the comments on the draft benchmark and is on track for an October mail ballot. With respect to decision support, the TPC-H subcommittee addressed new issues on dbgen and qgen, and TPC-H Version 2.5.0 was approved. Review of the DS benchmark also continues although its progress has been slowed by the dbgen bugs.

The TPC-C Maintenance Subcommittee did not convene at this meeting. At this stage the benchmark is functionally stabilized and the subcommittee has no open action items at this time. The subcommittee will continue to monitor the benchmark, along with any TAB interpretations and rulings, for discussion and possible inclusion in future versions of the TPC-C specification.

The TPC-H subcommittee defined Version 2.5.0 of the TPC-H specification, which includes a modified list of audit requirements and an updated executive summary and qualification query outputs.  It also includes Version 2.5.0 of the data generator dbgen, provided by the TPC, which includes bug fixes for parallel data generation. Version 2.4 was invalidated because the tools that were drafted for that revision were found to produce invalid results under certain conditions.

The General Council approved minor revision 2.5.0 of TPC-H with an effective date of January 1st 2007 .  The General Council also passed a motion that benchmarks of scale factors larger than 10,000 cannot be published with TPC-H Version 2.3.0.  Benchmark results can be published using Version 2.5.0 immediately.  Benchmark results can be published using Version 2.3.0 for scale factors up to 10,000 until December 31st 2006.

The TPC-App subcommittee did not convene at this meeting. The subcommittee made changes to the definition of the term Operating System to take into account the existence of virtualization software products. "A system" running multiple guest operating systems is considered to be a single system (not a cluster) for the purposes of this benchmark. The subcommittee also added clarification wording to the spec to require that all durable messages be stored on the SUT.

In May, the TPC-E subcommittee posted a draft version of the specification and TPC provided source code for formal review. Comments on the draft were accepted through July 30th. Over the past two months, the subcommittee has been in the process of doing a detailed review of the specification and responding to comments. The subcommittee received 261 comments and has reviewed and addressed 90 (as of August 10).

No TPC-DS business was conducted at this meeting as the subcommittee focused all efforts on the urgent TPC-H issues.

The Price committee does not have outstanding work items at this time.

The TPC will be making presentations at a number of industry conferences over the next 12 months. We welcome opportunities to present the benefits and status of TPC benchmarks. If you are interested in inviting someone from the TPC for a presentation, please contact Michael Majdalany (majdalany @ tpc.org)

The TPC is also conducting market research to understand the level to which decision makers use benchmarks in their purchasing decisions.

• 2015: May / November
• 2014: July / November
• 2013: January / March / May / November
• 2012: June
• 2011: May / August / November
• 2010: April / May / July / November
• 2009: January / April / June / November
• 2008: February / November
• 2007: February / April / June / September / October
• 2006: February / April / June / August / October / December
• 2005: February / April / May / August / September / December
• 2004: February / March / June / August / October / November
• 2003: January / March / May / August / October / December
• 2002: January / March / May / September / November
• 2001: March / May / July / September / November
• 2000: April / June / August / October / December
• 1999: January / March / May / July / September / November / December
• 1998: November