OMG you would think I would LOVE an announcement that Illumina and NVIDIA were working together but honestly this sounds like this morning's second case of Illumina having to give up a lot to get a little. Deal has two parts: (1) Illumina's "get" is to host a bunch of Nvidia tools for LLMs and data analytics in on the Illumina Connected Analytics platform. Also MONAI for spatial cell imaging workflows though not sure what they have to do with NGS. Anyway, this seems like a nice-to-have, maybe a convenience to customers - but I don't think it opens up any new functionality. Doesn't cost Nvidia anything, and they get to add another logo on their partners slide 🤷♂️ (2) Illumina (this is the "give" part) will be "enabling DRAGEN algorithms on NVIDIA GPUs". Not sure why this is good for Illumina, and the PR doesn't explain this benefit at all 🤯 DRAGEN is built on a hardware architecture - FPGA - that has nothing to do with GPU at all, in fact it's kind of the opposite. So Illumina is going to have to make a major investment in porting DRAGEN to Nividia GPU for no stated benefit to Illumina. But the benefit to Nvidia is clear: they get to claim yet another customer for their already vast GPU customer base, and they get to big to displace FPGAs in Illumina instruments. Must be missing something here 🤔🤔🤔 (Don't forget to check out our JMP podcast! https://lnkd.in/gU_qpY3S)
Just a thought, but one of the main objections to using DRAGEN has always been its rigidity. Like you can’t make any changes to the workflows, because FPGAs are programmed at the hardware level, not the software level. That’s fine in the clinic, but doesn’t fly so easily in research. Moving DRAGEN to GPUs potentially removes that drawback, even though I imagine if customers want to run it from a GUI, unfortunately they’ll still be limited to just ICA. Oh, and Basespace. 😤☹️
Jane Theaker
2w
FPGA reliance is reduced with this deal. Might be handy if supply chain redundancy/ accessibility from the US , out-competing China, faster development of GPU technology and flexibility of GPUs is important to Illumina. Note: GPUs are manufactured in both the USA and (secondary supply) South Korea. This is a deal which buys Illumina supply chain and flexibility and reduces nonUSA (Chinese?) competition by making GPUs easily accessible to only them.
I think the missing factor is the marketing / perception / market multiple of partnering with NVIDIA. Investors love it, hence +5% today, because it screams AI. The three life science partnerships i've seen announced today by NVIDIA seem super shallow and inconsequential, more co-marketing than co-product. And it tracks with the partnerships that were announced a year ago, most have gone nowhere.
Opening up their proprietary hardware and move it onto CUDA cores might make it a bit more enticing for enterprise omics adaptation in-house. Providing the software implemented on existing hardware may increase the sales of the DRAGEN platform.
I remember meeting with NVIDIA and you about 10 years ago in Boston at BioIT....
Waxing philosophical: this symbolizes the changing of the guard for what will now lead advancements in the “ bio century” GAI. And demonstrates genomics is now to GAI what PCR was to arrays and NGS— enabler
Layne Sadler
1w
NVIDIA acquired Parabricks. FPGA was the wrong decision to begin with because the writing was already on the wall that GPU was here to stay. How were they supposed to build an ecosystem and find talent around that? You can’t even find FPGA instance pricing for AWS instances anymore.
Alex C. Chan
2w
jurassic park quotes again (just replace [frog] DNA with AI): "Thinking machine super-computers and gene-sequencers break down the strand in minutes. And virtual-reality displays show our geneticists the gaps in the DNA sequence. We used the complete DNA of a frog to fill in the holes and complete the code."
Freelance | Data Engineer | Cloud | Infrastructure-as-Code
2wHere is my (cloud-centric) take after working with Dragen and Nvidia on Azure. I would guess Illumina is more motivated to make this happen than Nvidia: - virtual machines with FPGA accelerator are more scarce than virtual machines with GPU accelerator - moving to GPU makes Dragen more accessible and cheaper. - Nvidia Parabricks is a competitor to Dragen, it is also open source and runs on GPU. Running it in the cloud costs a fraction of what Dragen in the cloud costs. I think its a smart move by Illumina, again, Im wearing a cloud hat.