HammerDB is a free, open-source, database-load test and benchmarking tool that works with databases such as Oracle and SQL Server. HammerDB covers many aspects of database testing including generating the database schema, generating test data, and loading the database tables with test data. We used HammerDB to simulate both transaction-based and analytic scenarios on a specific Dell EMC reference architecture. Performance metrics that are generated from HammerDB load tests are used to validate features of the database and to characterize the performance of the overall system.
The HammerDB TPC-C data generation is scaled by specifying several fictitious warehouse facilities from which a company can supply products to fulfill customer orders. We used a scale factor of 15,000 warehouses when generating data for the TPC-C schema, which resulted in approximately 1.5 TB of data without compression or deduplication. Because the HammerDB graphical tool supports a maximum of 10,000 databases, we then loaded an additional 5,000 databases using SQL Loader. For additional clarity, the term “warehouse” in the context of a HammerDB TPC-C test has no relationship to any data warehousing workloads or decision support systems (DSS). TPC-C defines a transactional-based system and not a DSS.
We used HammerDB to create a transactional workload on the Oracle RAC database that represented our production system while we created a second copy using the PowerStore T. We also used the HammerDB TPC-C workload generator to validate that the test database was operational at the end of the copy process that is described in the following test cases.