Participant : Franck Cappello [ correspondant ] .
This project started officially in September 2004, under the name V-Grid. V-DS stands for Virtualization environment for large-scale Distributed Systems. It is a virtualization software for large scale distributed system emulation. This software allows folding a distributed systems 100 or 1000 times larger than the experimental testbed. V-DS virtualizes distributed systems nodes on PC clusters, providing every virtual node its proper and confined operating system and execution environment. Thus compared to large scale distributed system simulators or emulators (like MicroGrid), V-DS virtualizes and schedules a full software environment for every distributed system node. V-DS research concerns emulation realism and performance.
A first work concerns the definition and implementation of metrics and methodologies to compare the merits of distributed system virtualization tools. Since there is no previous work in this domain, it is important to define what and how to measure in order to qualify a virtualization system relatively to realism and performance. We defined a set of metrics and methodologies in order to evaluate and compared virtualization tools for sequential system. For example a key parameter for the realism is the event timing: in the emulated environment, events should occur with a time consistent with a real environment. An example of key parameter for the performance is the linearity. The performance degradation for every virtual machine should evolve linearly with the increase of the number of virtual machines. We conducted a large set of experiments, comparing several virtualization tools including Vserver, VMware, User Mode Linux, Xen, etc. The result demonstrates that none of them provides both enough isolation and performance. As a consequence, we are currently studying approaches to cope with these limits.
We have made a virtual platform on the GDX cluster with the Vserver virtualization tool. On this platform, we have launched more than 20K virtual machines (VM) with a folding of 100 (100 VM on each physical machine). However, some recent experiments have shown that a too high folding factor may cause a too long execution time because of some problems like swapping. Currently, we are conducting experiments on another platform based on the virtualization tool named Xen which has been strongly improved since 2 years. We expect to get better result with Xen than with Vserver. Recently, we have been using the V-DS version based on Xen to evaluate at large scales three P2P middleware  .
This software is available at http://v-ds.lri.fr/