VCWC: A Versioning Competition Workflow Compiler

System competitions evaluate solvers and compare state-of-the-art implementations on benchmark sets in a dedicated and controlled computing environment, usually comprising of multiple machines. Recent initiatives such as [6] aim at establishing best practices in computer science evaluations, especially identifying measures to be taken for ensuring repeatability, excluding common pitfalls, and introducing appropriate tools. For instance, Asparagus [1] focusses on maintaining benchmarks and instances thereof. Other known tools such as Runlim (http:/fmv.jku.at/runlim/) and Runsolver [11] help to limit resources and measure CPU time and memory usage of solver runs. Other systems are tailored at specific needs of specific communities: the not publicly accessible ASP Competition evaluation platform for the 3rd ASP Competition 2011 [4] implements a framework for running a ASP competition. Another more general platform is StarExec [12], which aims at providing a generic framework for competition maintainers. The last two systems are similar in spirit, but each have restrictions that reduce the possibility of general usage: the StarExec platform does not provide support for generic solver input and has no scripting support, while the ASP Competition evaluation platform has no support for fault-tolerant execution of instance runs. Moreover, benchmark statistics and ranking can only be computed after all solver runs for all benchmark instances have been completed. A robust job execution platform is a basic requirement for a competition. During benchmark evaluation, several different kinds of failures may happen, mainly (a) programming errors in the participant software; (b) software bugs in the solution verification programs; or (c) hardware failures during a run, which may be local to a machine (e.g., harddisk or memory failure), or global (e.g., when the server room air condition fails). Moreover, a competition platform must be flexible enough to allow for “late” or updated benchmark and solver submissions. It is not uncommon that anomalies arise during the execution, and changing the course of an evaluation after the platform has started is cumbersome and requires manual effort for the competition maintainers. A fault-tolerant design helps the competition maintainers to perform all steps and minimizes the action required to come back to a safe state. To address these issues, we introduce the Versioning Competition Workflow Compiler (VCWC) system. VCWC uses a two-step approach: first, a workflow for a competition track is generated; a workflow is a dependency description of jobs that need to be executed in order to come