Executing test runs in parallel is not uncommon when many test runs need to be executed. HTPar [8] is an extension to HTTrace so that test runs can be executed concurrently on a single or several machines. The goal is to exploit the available resources as well as possible. If several machines are available, the goal is to achieve linear speed-up; that is, the running time of executing the test decreases linearly with the number of machines. In order to achieve this speed-up, it is important to balance the load on all machines just as in all parallel applications [5]. At the same time,however, it is also important to control the state of the test database(s) and to execute the test runs in such a way that the number of database reset operations is minimized - just as for non-parallel testing in HTTrace. As a result, parallel testing involves solving a two-dimensional optimization problem: (a) partitioning: deciding which test runs to execute on which machine; and (b) ordering:deciding in which order to execute the test runs on each machine.