TPC-A
TPC-A measures performance in update-intensive database environments typical in on-line transaction processing applications. (Obsolete as of 6/6/95)

Summary
Issued in November, 1989, the TPC-A benchmark measures performance in update-intensive database environments typical in on-line transaction processing (OLTP) applications. Such environments are characterized by:

  • Multiple on-line terminal sessions
  • Significant disk input/output
  • Moderate system and application execution time
  • Transaction integrity
This benchmark uses a single, simple, update-intensive transaction to load the system tested. Thus the workload is intended to reflect an OLTP application, but does not reflect the entire range of OLTP requirements typically characterized by multiple transaction types of varying complexities. The single transaction type provides a simple, repeatable unit of work, and is designed to exercise the key components of an OLTP system.

The TPC-A benchmark measures how many transactions per second a system can perform when driven from multiple terminals. While TPC-A does not specify the number of terminals, it does require companies to scale (raise or lower) the number of terminals according to the throughput or tps rating. TPC-A can be run in a wide area or local area network configuration, with the performance described by the two metrics "TPC-A local Throughput" and "TPC-A wide Throughput," measured in transactions per second (tps). These two metrics are different and cannot be compared. The TPC-A benchmark specification requires that all companies fully disclose the details of their benchmark run, their system configuration and its cost (including 5 years maintenance cost).

Detailed Description
TPC Benchmark A exercises the system components necessary to perform tasks associated with that class of on-line transaction processing (OLTP) environments emphasizing update-intensive database services. Such environments are characterized by:

  • Multiple on-line terminal sessions
  • Significant disk input/output
  • Moderate system and application execution time
  • Transaction integrity
This benchmark uses terminology and metrics which are similar to other benchmarks, originated by the TPC and others. Such similarity in terminology does not in any way imply that results are comparable to benchmarks other than TPC Benchmark A.

The metrics used in TPC-A are throughput as measured in transactions per second (tps), subject to a response time constraint; and the associated price-per-tps. Comparison of price/performance results disclosed in one country may not be meaningful in another country because of pricing and product differences.

The throughput metric is "tpsA". TPC-A can be run in a wide area or local area network configuration. No distinction is made between results for wide area and local area configurations.

This benchmark uses a single, simple, update-intensive transaction to load the system under test (SUT). Thus the workload is intended to reflect an OLTP application, but does not reflect the entire range of OLTP requirements typically characterized by multiple transaction types of varying complexities. The single transaction type provides a simple, repeatable unit of work, and is designed to exercise the key components of an OLTP system.

The extent to which a customer can achieve the results reported by a vendor is highly dependent on how closely TPC-A approximates the customer application. Relative performance of systems derived from TPC-A do not necessarily hold for other workloads or environments. Extrapolations to unlike environments are not recommended.

A full disclosure report of the implementation details must be made available along with the reported results.

Benchmark results are highly dependent upon workload, specific application requirements, and system design and implementation. Relative system performance will vary as a result of these and other factors. Therefore TPC-A should not be used as a substitute for specific customer application benchmarking when critical capacity planning and/or product evaluation decisions are contemplated.