Using its Second Generation cloud offering, Oracle Cloud Infrastructure (OCI) has reached new heights for Ultrarapid Nanopore Sequencing Analysis. This speed and compute performance is another delivery of Oracle’s support for fast and accurate genome analysis for healthcare and researchers, which can help save lives.
In 2022, NVIDIA released an Ultrarapid Nanopore Analysis Pipeline (UNAP) capable of running an Oxford Nanopore germline workflow for small and structural variants of a 55x coverage sample in just 4 hours and 10 minutes on an NVIDIA DGX A100. In collaboration with NVIDIA, Oracle Cloud engineering has deployed a new workflow featuring upgrades to both software and hardware, including replacing Guppy with the new Dorado basecaller, the latest NVIDIA Parabricks DeepVariant v1.5 and the latest NVIDIA H100 Tensor Core GPUs.
UNAP analysis was processed on an OCI computing environment containing a single node of eight H100 GPUs, each with 80GB memory. On OCI, the workflow benefits from bare-metal GPU compute, an industry-leading 61.44TB of included NVMe storage and time-to-market advantages OCI customers can realize with deep ongoing engagement from Oracle’s world-renowned engineering organization.
Record-setting benchmark results
Utilizing all eight NVIDIA H100 GPUs in a single node, Oracle completed the entire UNAP workflow — including basecalling and alignment as well as multiple variant calling — in just 57.44 minutes. This performance represents a 4.3x speedup compared to the previous workflow on DGX A100.
By combining industry-leading performance for genome analysis with OCI’s lower, simpler and more predictable pricing, Oracle can help organizations deliver better and faster genomic analysis -at a lower cost.
Want to learn more?
Join us at Oracle Health Conference, from September 18-20. Register for the Oracle + NVIDIA sessions to learn more about genomic analysis, these noteworthy results and how OCI is using AI and machine learning to transform healthcare. For more information on NVIDIA DGX A100 and NVIDIA H100 Tensor Core GPUs, see OCI Compute: GPU instances.