Reconstructing the software environment of an experiment with kameleon

Abstract

In the scientific experimentation process, an experiment result needs to be analyzed and compared with several others, potentially obtained in different conditions. Thus, the experimenter needs to be able to redo the experiment. Several tools are dedicated to the control of the experiment input parameters and the experiment replay. In parallel concurrent and distributed systems, experiment conditions are not only restricted to the input parameters, but also to the software environment in which the experiment was carried out. It is therefore essential to be able to reconstruct this type of environment. The task can quickly become complex for experimenters, particularly on research platforms dedicated to scientific experimentation, where both hardware and software are in constant rapid evolution. This article discusses the concept of the reconstructability of software environments and proposes a tool for dealing with this problem.

Challenges of Large-Scale Biomedical Workflows on the Cloud -- A Case Study on the Need for Reproducibility of Results

“…The infrastructure layer controls processes related to launching a virtual machine with the desired storage, whereas the software layer is concerned with process on a running virtual machine. Sanabria et al, [17] presented an approach called Kameleon [18] based on a software appliance generator. The authors defined the term appliance as pre-built software that is combined with just enough operating system (jeOS) and can run on a bare metal (real hardware) or inside a hypervisor.…”

Trusted High-Performance Computing in the Classroom

“…Tools and ideas were presented, such as Pegasus [11] and Kameleon [5]: these tools target reproducible experiments, mainly addressing the problem exploiting virtual machines to reproduce the environment where an experiment is held. The main focus of these research groups keeps being on the virtual machine approach combined with cloud computing, as can be seen in [1] and [10].…”

Reproducible Software Appliances for Experimentation

“…We extended our previous work Kameleon [10] which is a very simple software appliance generator that enables the construction and exact post reconstruction of a given software appliance from text descriptions. It is targeted to make easier the reconstruction of custom software stacks in HPC, Grid, or Cloudlike environments.…”

“…This recipe is a higher level description easy to understand and contains some necessary meta-data in form of global variables and steps. For more details [10] 2. The DATA which is used as input of all the procedures described in the recipe.…”

“…The low size of Kameleon engine and Polipo (less than 1MB) makes feasible the distribution of the exact versions used to create the environment, avoiding the incompatibility between versions. The whole process is depicted in Fig 2. More information can be found in [10] or in Kameleon web site 9 .…”