AGRF's sequencing times reduced 86% with FlashBlade

Pure Storage has announced that the Australian Genome Research Facility (AGRF) is leveraging its FlashBlade unified fast file and object (UFFO) solution,

The AGRF is using the solution to speed up data performance and reduce genomic pipeline turnaround times for global clients.

AGRF caterers to over 15,000 customers globally. It partners with the genomics community to provide innovative and integrated capabilities across the biomedical, clinical, agricultural and environmental sectors.

AGRF’s genomic solutions are focused on making improvements to people’s everyday lives, demonstrated through their involvement in COVID-19 research during the pandemic and critical conservation efforts for threatened species in the Australian ecosystem.

Of course, which such a large task, there is an enormous amount of data and metadata developed through genomic sequencing. This requires scalable, reliable and high-performance storage infrastructure to deal with it.

To be more specific, AGRF generates over ten terabytes of raw data each week, something that increases roughly four-fold before making its way back to clients.

“The simplicity and scalability of FlashBlade not only enables us to run our sequencing analysis workflow quickly and frictionlessly, but the increased performance allows us to introduce new technology to help with overall compute functionality. Additionally, FlashBlade can directly connect with Sequencers, including Illumina, to ingest the raw sequenced data, which dramatically improves AGRF’s workflows in the field,” says Douglas Morrison, ICT Manager, Australian Genome Research Facility (AGRF).

Before implementing Pure Storage’s FlashBlade, AGRF’s existing scale-out file system technology was difficult to manage and was unstable, complex and expensive.

Now, FlashBlade has increased AGRF’s metadata performance and storage capacity and, ultimately, optimized its end-to-end sequencing pipeline to deliver rapid, cost-effective sequencing services.

By migrating its existing workloads to FlashBlade, AGRF obtained higher I/O and bandwidth, all at lower latency, allowing compute jobs to finish faster and increasing the company’s computing capability without compromising capacity. 

AGRF’s clinical genome sequencing times have reduced by as much as 86% since being supported by FlashBlade, going from 28 days to just ten. 

The company has also reduced the time it takes to complete the key initial phase of its analysis workflow from 18 hours to just three - a six-fold improvement in performance. This is while reducing another subsequent phase of the workflow from 10 hours to just 23 minutes, a 26x improvement in performance.  

“Now, we have the ability to deliver faster results, leading to critical decisions being made sooner, while taking on more life-saving projects,” adds Morrison.