Section: New Software and Platforms

SimGrid/VMPlaces

Participants : Takahiro Hirofuchi, Adrien Lebre [correspondent] , Jonathan Pastor, Flavien Quesnel, Mario Südholt.

Simulation, Virtualization, Cloud computing, VM placement

SimGrid is a toolkit for the simulation of algorithms executed on large-scale distributed systems. Developed for more than a decade, it has been used in a large number of studies described in more than 100 publications. In 2013, ASCOLA with the support of the SimGrid core-developers, designed and implemented additional capabilities, in particular the Virtual Machine abstraction, enabling to address Cloud Computing related concerns.

Developed, first, in an experimental repository, the integration of these extensions into the master branch of SimGrid has been achieved during Summer 2014. The principal role of ASCOLA is now to ensure the maintenance of this portion of the code with respect to the evolutions of the SimGrid toolkit (such as for instance the recent port of the SimGrid kernel in C++).

Although the virtualization extensions are recent, several projects leveraging them have been already proposed. (The list of the projects is available at : http://simgrid.gforge.inria.fr/contrib/clouds-sg-doc.html ) Among them, ASCOLA is working on dedicated framework to evaluate and compare VM placement algorithms. Entitled VMPlaces, this framework is composed of two major components: the injector and the VM placement algorithm. The injector is the generic part of the framework (i.e. the one you can directly use) while the VM placement algorithm is the part you want to study (or compare with available algorithms). Currently, the VMPlaceS is released with three algorithms:

• Entropy, a centralized approach using a constraint programming approach to solve the placement/reconfiguration VM problem

• Snooze, a hierarchical approach where each manager of a group invokes Entropy to solve the placement/reconfiguration VM problem. Note that in the original implementation of Snooze, it is using a specific heuristic to solve the placement/reconfiguration VM problem. As the sake of simplicity, we have simply reused the entropy scheduling code.

• DVMS, a distributed approach that dynamically partitions the system and invokes Entropy on each partition.

SimGrid is available at http://simgrid.gforge.inria.fr .

VMPlaces is available at http://beyondtheclouds.github.io/VMPlaceS/