The single-source shortest path problem is one of the most studied problems in computer science. Nevertheless, it is infamous for being notoriously hard to parallelize due to the so-called transitive closure bottleneck. In 1998 an algorithm called -stepping was proposed to parallelize the single-source shortest path problem. This algorithm formed the basis for further research in even more optimized algorithms, and eventually became to be the de-facto standard in parallelizing the single-source shortest path problem. A second idea by Crauser et al. to parallelize the single-source shortest path problem was published in 1998. This idea did not get a lot of attention, and seems to have been forgotten. The goal of this thesis is to evaluate the viability and performance of Crauser et al.'s approach in comparison to -stepping. Furthermore, several ideas to improve or change this approach are evaluated. In order to do this, these algorithms are first introduced, then proven correct, and finally implemented. There are two implementations of each algorithm. One implementation is to analyze the number of phases each algorithm needs to solve the single-source shortest path problem, a key metric in assessing the quality of such algorithms. The second implementation is optimized to measure the performance on the systems provided by the research group. The phase analysis shows that the improvements to Crauser et al.'s original formulation of their algorithm reduce the number of phases by a large margin, therefore leading to a theoretical gain. Unfortunately, no efficient implementation was found for these improvements, and therefore they cannot enhance real-world performance. Nevertheless, Crauser et al.'s original algorithm performs very well and is competitive to -stepping.