Section: Software

Wrekavoc

Participants : Jens Gustedt, Emmanuel Jeannot.

Wrekavoc addresses the problem of controlling the heterogeneity of a cluster. Our objective is to have a configurable environment that allows for reproducible experiments on large sets of configurations using real applications with no emulation of the code. Given an homogeneous cluster Wrekavoc degrades the performance of nodes and network links independently in order to build a new heterogeneous cluster. Then, any application can be run on this new cluster without modifications. Therefore, Wrekavoc helps to validate parallel and distributed applications and algorithms. Moreover, on the modeling side, it helps to understand the impact of platform parameters (latency, bandwidth, CPU speed, memory) on application performance.

Wrekavoc is implemented using the client-server model. A server, with administrator privilege, is deployed on each node one wants to configure. The client reads a configuration file and sends orders to each node in the configuration. The client can also order the nodes to recover the original state.

CPU Degradation.

We have implemented several methods for degrading CPU performance. The first approach consists in managing the frequency of the CPU through the kernel CPU-Freq interface. We propose two other solutions in case CPU-Freq is not available. One is based on CPU burning. A program that runs under real-time scheduling policy burns a constant portion of the CPU, whatever the number of processes currently running. The other is based on user-level process scheduling called CPU-lim. A CPU limiter is a program that supervises processes of a given user. On Linux, using the /proc pseudo-filesystem, it suspends the processes when they have used more than the required fraction of the CPU.

Network Limitation.

Limiting latency and bandwidth is done using tc (traffic controller) based on Iproute2 a program that allows advanced IP routing. With this tool it is possible to control both incoming and outgoing traffic. Furthermore, the latest versions (above 2.6.8.1) allow to control the latency of the network interface.

Overlay networking

Thanks to the notion of gateways it is possible to connect islets of nodes and to construct an overlay network matching a user-defined topology. Such network emulates TCP/IP based network with congestion emulation or routing.

Memory Limitation.

Wrekavoc is able to limit the amount of memory available by the processes thanks to the use of the mlock and munlock syscalls.

Configuring and Controlling Nodes and Links.

The configuration of a homogeneouscluster is made through the notion of islet. An islet is a set of nodes that share similar limitations. Two islets are linked together by a virtual network which can also be limited. An islet is defined as a union of IP addresses (or machine names) intervals.

Each islet configuration is stored into a configuration file. At the end of this file is described the network connection (bandwidth and latency) between each islet.