Accelerating High Performance Computing (HPC) for Population-level Genomics

By Mileidy Giraldo, Ph.D.

September 30, 2019

The development of Next-Generation Sequencing (NGS) technologies in the late 2000’s led to a dramatic decrease in the cost of DNA sequencing. The advent of NGS coupled to the advancements in HPC storage and computing technologies at the time created the perfect storm for a deluge of genomics data. This confluence of events led to a pressing question: how best to put all this data to use?

A genome is an organism’s complete set of DNA and, as such, it ultimately determines all biological functions and the myriad of variations that make some of us susceptible while others immune to different diseases. Therefore, it is of great interest to the biomedical community to determine an individual’s genome, which is a process much like deciphering scrambled letters (genome sequencing) so one can assemble them into words (genome assembly) to write a book, (variant analysis).

Given the new affordability of NGS methods and the increased computing and storage capacities of the last decade, genomics can now be performed at the population level. Large national genomics initiatives such as the “UK Biobank,” the “All of Us program” in the US, Singapore’s “GenomeAsia,” “Genomics Thailand,” etc. are emerging all around the world. With goals of sequencing 500K to over 1M participants in a few years’ time, these country-wide efforts aim to capture the genetic variation of their people to make Precision Medicine a reality. With Precision Medicine the hope is to deliver individualized prevention, diagnosis, and treatment by leveraging knowledge from a person’s genetic background.

The greatest challenge such population-level genomics efforts face is scale: scaling up in input data from exomes (the portions of a genome that code information for protein synthesis) to whole genomes, as well as scaling up production (from a handful to tens of thousands of samples), and the corresponding challenges it creates for the HPC infrastructure. Exomes correspond to 1% of whole genomes and are small regions in genes dictating important biological functions. Yet, exomes cannot provide the comprehensive picture found in the remaining 99% of whole genomes. Exomes were traditionally sequenced because of their smaller size, lower cost, and faster processing. Today, many genomics centers around the world are making the transition from exome to whole-genome sequencing, while also trying to tackle unprecedented volumes of data from hundreds of thousands of patients.

Three out of the four analysis stages in genomics take place in the HPC environment of a cluster or supercomputer, including genome assembly (assembling the DNA letters into words), variant analyses (comparing how a word/gene is spelled in different people), and downstream bioinformatics (measuring effect of variations on function/disease). Therefore, scaling out genomics productions largely depends on the HPC technologies made available to the genomics applications.

With these dependencies in mind, Lenovo set out to identify which technologies bring the most acceleration to genomics workflows. To that end, we conducted a systematic study of the performance of hundreds of parameters on 30+ tools in the Broad Institute’s Genome Analysis Tool Kit (GATK) Germline Variant Calling Workflow against hundreds of permutations of hardware building blocks, system tunings, data types (exomes, whole genomes), execution modes (latency vs. throughput), and software implementations (e.g. standard vs. Spark, etc.).

Today, it still takes a typical datacenter 150-160 hrs. to process a single whole genome and 4-6 hrs. for an exome. In 2017, Intel’s work on BIGstack (Intel’s reference architecture for GATK workflows) reduced processing times to 10.8 hr. and 25 min, respectively. As a result of Lenovo’s permutation tests of the hardware, software, and system factors affecting the performance of genomics workflows we identified an optimized architecture that can process 1 whole genome in 5.5 hrs. and 1 exome in 4 minutes with no specialty hardware. With Lenovo’s genomics optimized hardware, a data center can expect to process 4.5 genomes or 343 exomes per node per day. Some genomics solutions out there promise processing times around 3-4 hr. for a whole genome but require expensive, specialized hardware that does not scale well for large volumes and licensing proprietary software. Lenovo’s optimized genomics architecture on the other hand, provides a 27X to 40X performance improvement on non-specialty hardware, and does so in a manner that is more affordable, more scalable, and reduces costs by leveraging open-source software that is validated and widely-accepted by the scientific community.

Another byproduct of Lenovo’s systematic genomics performance testing was the ability to generate a fluid rather than a static refence architecture for genomics as is the norm in the HPC industry. Every genomics data center adopts a different mix of workloads, analyses workflows, has different active and archiving storage needs, a different mix of research types to support, and therefore needs a customized architecture tailored to their specific needs. Thus, we converted the lessons learned from our genomics benchmarking and systematic testing into formulas captured in an industry-first Genomics Sizing Tool.

Lenovo’s Genomics Sizing Tool calculates the projected HPC usage for an expected workload; for example, it outputs the compute nodes, active, and archive storage needed to meet a workload quota (e.g., 50K genomes/yr.). The Sizing Tool can also be used to size the current production capabilities of an existing cluster: e.g., to answer the questions of “[H]ow many genomes can I process with my current cluster?,” Or, “[H]ow many genomes/yr. can this year’s budget afford me?”

We are leveraging both Lenovo’s optimized architecture and the Genomics Sizing Tool to help data centers around the world accelerate their workflows and plan their HPC resources more effectively as they embark on ever increasing workloads from cohort-level and population-level genomics projects. Lenovo’s team of Genomics experts work together with the data center’s researchers, developers, and HPC experts to create custom HPC usage designs projecting data growth over time, designing data flow, storage, and management across the cluster. These exercises in HPC usage and projections are proving invaluable in workload management, budget planning, IT expenditure justification and allocation, and resource accountability. Through its commitment to developing and adopting cutting-edge technological innovation, Lenovo is enabling the worldwide movement to sequence entire populations, bringing such initiatives closer to making precision medicine a reality, and delivering on its promise of Smarter Technology for All. A white paper will soon follow with a detailed description of the systematic permutation tests and benchmarks alluded to here as well as the resulting optimizations and Genomics Sizing Tool accelerating and sizing the HPC resources for deploying genomics at scale.

 

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire