SLA-aware scheduling of map-reduce applications on public clouds

The recent need of processing BigData has led to the development of several Map-Reduce applications for efficient large scale processing. Due to on-demand availability of large computing resources, Public Clouds have become a natural host of these Map-Reduce applications. In this case, users need to decide which resources they need to rent to run their Map- Reduce cluster other than deployment or scheduling of map-reduce tasks itself. This is not a trivial task particularly when users may have performance constraints such as deadline and have several Cloud product types to choose with intention of not spending much money. Even though there are several existing scheduling systems, however most of them are not developed to manage the scheduling of Map-Reduce applications. That is, they do not consider things like the number of map and reduce tasks and slots per VM. This paper proposes a novel greedy scheduling algorithm (MASA) that considers the users constraints in order to minimize cost of renting Cloud resources while considering the user’s budget and deadline constraints. The simulation results show 25-60% reduction cost in comparison to current methods by using our proposed algorithm.