Get Started With NVIDIA Parabricks
NVIDIA® Parabricks® is a scalable genomics analysis software suite that leverages full-stack accelerated computing to process data in minutes. Compatible with all leading sequencing instruments, it supports diverse bioinformatics workflows and integrates AI for accuracy and customization. A cost-effective and space-saving solution, Parabricks is ideal for large-scale genomics projects focused on advancing disease understanding and management.
Benefits of Parabricks
Parabricks Packaging and Availability
Containers
Parabricks is available as a public container on NVIDIA NGC™ for use on premises or any cloud service platform or provider. For customers who need support for Parabricks in clinical workflows, large sequencing projects, high-throughput platforms, or other workflows that require immediate assistance, NVIDIA offers NVIDIA AI Enterprise.
Compare the benefits of Parabricks with NVIDIA AI Enterprise
Microservices
Parabricks microservices allow users to effortlessly access applications for accelerating genomics analysis. To deploy anywhere and self-host AI applications and data, NVIDIA NIM for an accelerated runtime on inference models and CUDA-X™ Microservices for NVIDIA-accelerated SDKs and tools provide the fastest and highest-performing production AI.
The Cloud-First Way to Get the Best of NVIDIA AI
Parabricks is powered by NVIDIA DGX™ Cloud, an end-to-end AI platform for developers that offers scalable capacity built on the latest NVIDIA architecture and co-engineered with the world’s leading cloud service providers.
Start Your AI Journey With NVIDIA LaunchPad
Get access to NVIDIA AI Enterprise solutions with a free hands-on lab on NVIDIA Parabricks.
Parabricks Ecosystem
Resources
Access documentation, tutorials, and community support for Parabricks.
FAQ
NVIDIA Parabricks is a GPU-accelerated software suite for secondary genomic analysis. Parabricks is compatible with all leading sequencing instruments and supports diverse bioinformatics workflows.
Parabricks increases speed, reducests cost, and boosts accuracy for secondary genomic analysis.
- Boost accuracy: Deploy high-accuracy deep learning for all major sequencers.
- Increase speed: Experience 108X faster analysis of whole-genome sequencing (WGS) compared to CPU-only solutions.
- Reduce cost: Lower compute cost by up to 50 percent for WGS compared to CPU-only solutions.
Yes, Parabricks matches the results of open-source tools, so you can switch out open-source tools and get the exact same results with Parabricks. The latest benchmarks from the 4.3.1 release can be found in the table below.
| NVIDIA H100 Tensor Core GPU
Fastest Speed |
NVIDIA L4 Tensor Core GPU
Lowest Cost Per Sample |
||||
| 2x GPUs | 4x GPUs | 2x GPUs | 4x GPUs | 8x GPUs | |
| FQ2BAM | 17.18 | 9.88 | 47.35 | 21.77 | 13.60 |
| BWA-Meth | 27.43 | 15.12 | 77.35 | 39.77 | 22.47 |
| DeepVariant | 9.67 | 5.82 | 23.48 | 13.10 | 7.8 |
| HaplotypeCaller | 10.57 | 4.90 | 12.00 | 7.73 | 4.27 |
| Mutect2 | 25.80 | 13.60 | 55.8 | 32.50 | 17.5 |
Performance time in minutes.
30x whole genome sequenced for FQ2BAM, BWA-Meth, DeepVariant, and Haplotype Caller with Illumina data.
50x tumor-normal whole genome sequenced for Mutect2 with Illumina data.
Parabricks is available as a public container on NVIDIA NGC™ for use on premises or any cloud service platform or provider.
For customers who need support for Parabricks in clinical workflows, large sequencing projects, high-throughput platforms, or other workflows that require immediate assistance, NVIDIA offers NVIDIA AI Enterprise.
Parabricks can be run on public cloud providers, platforms and partners, and on premises.
- Public cloud providers: AWS, Google Cloud, Oracle Cloud, Microsoft Azure, Naver Cloud Platform, Baidu AI Cloud, Alibaba Cloud
- Platforms and partners: DNAnexus, AWS, Terra, Velsera, FormBio, Lifebit
- On premises: NVIDIA DGX systems, NVIDIA-Certified servers, Dell Technologies, Hewlett Packard Enterprise, Lenovo, Cisco, Atos, Supermicro
Hardware Requirements
- Any NVIDIA GPU that supports CUDA architecture 70, 75, 80, 86, 89, or 90 and has at least 16 gigabytes (GB) of GPU RAM. NVIDIA Parabricks has been tested on the following NVIDIA Tensor Core GPUs:
- V100
- T4
- A10, A30, A40, A100, A6000
- L4, L40
- H100, H200
- Grace Hopper™ Superchip
- The fq2bam tool requires at least 24GB of GPU memory by default; the low-memory option will reduce this to 16GB of GPU memory at the cost of slower processing. All other tools require at least 16GB of GPU memory per GPU.
- System Requirements:
- A two-GPU system should have at least 100GB CPU RAM and at least 24 CPU threads.
- A four-GPU system should have at least 196GB CPU RAM and at least 32 CPU threads.
- An eight-GPU system should have at least 392GB CPU RAM and at least 48 CPU threads.
Software Requirements
- An NVIDIA driver with version 525.60.13 or greater
- Any Linux operating system that supports nvidia-docker2 Docker version 20.10 (or higher)
- Fastest speed: NVIDIA H100 Tensore Core GPUs
- Lowest cost per sample: NVIDIA L4 Tensor Core GPUs
- Brand new support: NVIDIA Grace Hopper Superchip
Parabricks is supported on most NVIDIA enterprise GPUs, including A100, A6000, A40, A30, A10, L40, T4, and V100 Tensor Core GPUs.
Please contact us with your question.