Session
Schedule FOSDEM 2020
HPC, Big Data, and Data Science

Facilitating HPC job debugging through job scripts archival

UB5.132
Andy Georges
SArchive is a lightweight tool to facilitate debugging HPC job issues by providing support teams with the exact version of the job script that is run in the HPC job in an archive either on the filesystem, in Elasticsearch, or by producing it to a Kafka topic.

HPC schedulers usually keep a version of the user’s job script in their spool directory for the lifetime of the job, i.e., from job submission until the job has run to completion — either succesfully or failed. However, once the job has completed, the job script and associated files are removed to avoid stacking up a large number of files. HPC systems typically run several millions of jobs, if not many more, over their lifetime -- it is not feasible to keep them all in the spool directory. In case the job failed, user support teams are often asked to help figure out the cause of the failure. For these occasions, it often is helpful if the exact job script is available. Since a typical scheduler setup will make changes to every submitted script through, e.g., a submission filter, simply obtaining what the user submitted requires an extra hoop to run the given script through the filter(s). Furthermore, users may have tweaked, changed, or removed the job script, which may add to the difficulty of debugging the issue at hand.

SArchive aims to address this problem by providing user support teams with an exact copy of the script that was run, along with the exact additional files that are used by the scheduler, e.g., to set up the environment in which the jobs runs. It can be argued that making a backup copy is actually the job of the scheduler itself, but we decided to use a tool outside the scheduler. This has the advantages that (i) one need not have access to the scheduler’s source code (not all schedulers are open source) and (ii) sites running multiple schedulers need not make any changes to each of them, but only to SArchive — which should be a fairly limited effort, if any at all. SArchive is currenly tailored towards the Slurm scheduler (hence the name), but it also supports the Torque resource manager. Adding support for other schedulers should be fairly straightforward — pull requests are welcome :)

Currently, SArchive provides three archival options: storing archived files inside a file hierarchy, ship them to Elasticsearch, or produce them to a Kafka topic. File archival is pretty feature complete, the code for shipping to Elasticsearch and Kafka is still under development and only has what is needed in our (HPCUGent) specific setup — which may evolve.

Additional information

Type devroom

More sessions

2/2/20
HPC, Big Data, and Data Science
Colin Sauze
UB5.132
This talk will discuss the development of a RaspberryPi cluster for teaching an introduction to HPC. The motivation for this was to overcome four key problems faced by new HPC users: The availability of a real HPC system and the effect running training courses can have on the real system, conversely the availability of spare resources on the real system can cause problems for the training course. A fear of using a large and expensive HPC system for the first time and worries that doing something ...
2/2/20
HPC, Big Data, and Data Science
Adrian Woodhead
UB5.132
This presentation will give an overview of the various tools, software, patterns and approaches that Expedia Group uses to operate a number of large scale data lakes in the cloud and on premise. The data journey undertaken by the Expedia Group is probably similar to many others who have been operating in this space over the past two decades - scaling out from relational databases to on premise Hadoop clusters to a much wider ecosystem in the cloud. This talk will give an overview of that journey ...
2/2/20
HPC, Big Data, and Data Science
Félix-Antoine Fortin
UB5.132
Compute Canada provides HPC infrastructures and support to every academic research institution in Canada. In recent years, Compute Canada has started distributing research software to its HPC clusters using with CERN software distribution service, CVMFS. This opened the possibility for accessing the software from almost any location and therefore allow the replication of the Compute Canada experience outside of its physical infrastructure. From these new possibilities emerged an open-source ...
2/2/20
HPC, Big Data, and Data Science
Moritz Meister
UB5.132
Maggy is an open-source framework built on Apache Spark, for asynchronous parallel execution of trials for machine learning experiments. In this talk, we will present our work to tackle search as a general purpose method efficiently with Maggy, focusing on hyperparameter optimization. We show that an asynchronous system enables state-of-the-art optimization algorithms and allows extensive early stopping in order to increase the number of trials that can be performed in a given period of time on ...
2/2/20
HPC, Big Data, and Data Science
Suneel Marthi
UB5.132
The advent of Deep Learning models has led to a massive growth of real-world machine learning. Deep Learning allows Machine Learning Practitioners to get the state-of-the-art score on benchmarks without any hand-engineered features. These Deep Learning models rely on massive hand-labeled training datasets which is a bottleneck in developing and modifying machine learning models. Most large scale Machine Learning systems today like Google’s DryBell use some form of Weak Supervision to construct ...
2/2/20
HPC, Big Data, and Data Science
Frank McQuillan
UB5.132
In this session we will present an efficient way to train many deep learning model configurations at the same time with Greenplum, a free and open source massively parallel database based on PostgreSQL. The implementation involves distributing data to the workers that have GPUs available and hopping model state between those workers, without sacrificing reproducibility or accuracy. Then we apply optimization algorithms to generate and prune the set of model configurations to try.
2/2/20
HPC, Big Data, and Data Science
UB5.132
Predictive maintenance and condition monitoring for remote heavy machinery are compelling endeavors to reduce maintenance cost and increase availability. Beneficial factors for such endeavors include the degree of interconnectedness, availability of low cost sensors, and advances in predictive analytics. This work presents a condition monitoring platform built entirely from open-source software. A real world industry example for an escalator use case from Deutsche Bahn underlines the advantages ...