Making efficient use of modern multi-core and future many-core
CPUs is a major challenge. We describe a new compiler-based plat-
form, Prospect, that supports the parallelization of sequential appli-
cations. The underlying approach is a generalization of an existing
approach to parallelize runtime checks. The basic idea is to gener-
ate two variants of the application: (1) a fast variant having bare
bone functionality, and (2) a slow variant with extra functional-
ity. The fast variant is executed sequentially. Its execution is di-
vided into epochs. Each epoch is re-executed by an executor us-
ing the slow variant. The approach scales by running the executors
on multiple cores in parallel to each other and to the fast variant.
We have implemented the Prospect framework to evaluate this ap-
proach. Prospect allows custom plug-ins for generating the fast and
slow variants. With the help of our novel StackLifter, a process can
switch between the fast variant and the slow variant during runtime
at arbitrary positions.