Expanding the horizons of life-science research workloads
When an international leader in genetics, genomics, statistics and structural biology faced rapid growth and the
daunting challenge of accommodating new workload types in their HPC environment, they tested all available options.
The Wellcome Centre for Human Genetics (WHG) at the University of Oxford had reached the limits of their open-source
Grid Engine scheduler: it lacked support for bugs, security vulnerabilities and GPUs; and it did not address the use of
containerized machine learning applications. To serve its 400 researchers and 70 support personnel, the WHG shared HPC
cluster comprises over 4,000 InfiniBand-connected, high-memory compute cores and 4PB of high performance, parallel storage
running 250 applications. WHG needed a robust scheduler to run their workloads (serial-batch, array, MPI, container, Spark)
on the same, shared cluster.
After extensive testing of both open-source and commercial products, WHG selected Univa Grid Engine, citing its modern
scheduler, expert technical support, and minimal user re-training. With the emergence of deep learning and image processing,
WHG can now accommodate GPU and other technologies within the same administrative framework.
WHG retained its investments in skills and applications, reduced time spent on cluster management, and now enjoys new
capabilities like GPU-aware scheduling, DRMAA2, and container support.
"The conversion from the previous scheduler to Univa Grid Engine was virtually painless. Our users are happy that their
hard-won knowledge continues to be relevant, significant scheduler bugs and vulnerabilities were fixed, and we also save on
our own precious system administration time," said Dr. Robert Esnouf, Head of Research Computing Core, Wellcome Centre for
Human Genetics, University of Oxford.
To access the case study .pdf, please complete the form below.
Fields marked with (*) are mandatory.