The TPC-E model
TPC-E is modeled on a brokerage with its database of customers, accounts and securities. In a typical brokerage setting, customers generate data research requests, account inquiries and trade requests. The brokerage house receives trade requests, sends them to the market and returns the results to the customer. When developing TPC-E, TPC consulted with well-known brokerage houses to develop a model that is as realistic as possible. TPC-E uses multiple trade types -- market orders, limit orders, stop orders and more. There are several tiers of simulated customers, each with distinct trading behaviors. Simulated companies issue both preferred and common stock. When customers make
The resulting model comprises 33 tables, 22 check constraints and referential integrity constraints, including 33 primary and 50 foreign keys. TPC-E has three dimensions of data, posing realistic partitioning challenges. The model implements business logic using 10 transactions, many with various modes of execution. The rich schema and diverse transactions of TPC-E provide new opportunities for vendors to optimize system performance.
Modern database features
TPC-E is designed to match several additional features of contemporary databases:
Low barriers to TCP benchmark adoption
Benchmark sponsors will find that running TPC-E is relatively easy and affordable. TPC-E has been carefully designed and tested to balance CPU demands with physical I/O, minimizing the number of disk drives required to run it. In addition to shrinking the cost of running TPC-E, lower I/O requirements will result in a closer resemblance between benchmark configurations and those used in real data centers. Also, TPC-E is focused on benchmarking the database server and requires fewer front-end systems to emulate user actions. Finally, TPC has provided code that generates test data for the initial database as well as implementations of key parts of the workload logic. This lets test sponsors focus on optimizing their products rather than developing and optimizing the benchmark application. All of these design choices reduce cost and complexity, putting TPC-E within reach of a broader spectrum of hardware vendors. The more broadly a benchmark is used, the more helpful it is for both buyers and manufacturers.
Transitioning to TPC-E
In many industries that use OLTP, TPC-E will deliver highly relevant performance measurements and drive practical system improvements. While lower cost and simplified infrastructure will allow TPC-E to be adopted rapidly, TPC-C remains a crucial part of the technology industry. Vendors can continue maximizing the value of their investment in TPC-C while preparing to transition to or add TPC-E to their environments. Customers can gain expertise in interpreting TPC-E results and shift their decision frameworks gradually to include the new benchmark. Just as TPC-C has done, TPC-E will help the transaction processing industry achieve new innovations and new heights of performance.
ABOUT THE AUTHOR: Matthew Emmerton is the chairperson of the TPC-E Maintenance Subcommittee. He has been involved in the development and maintenance of the TPC-E benchmark standard since 2005. He is also a member of the TPC's Technical Advisory Board.
What did you think of this feature? Write to SearchDataCenter.com's Matt Stansberry about your data center concerns at firstname.lastname@example.org.
This was first published in February 2009