SC24: A Milestone Moment for NextSilicon
As I look back on the SC24 supercomputing conference in Atlanta, it was clear that this year was unlike any before it. For NextSilicon, SC24 wasn’t just like another event– it marked a turning point in our 7-year journey. We unveiled Maverick-2, the industry’s first Intelligent Compute Accelerator (ICA), and witnessed firsthand how its promise resonated with the HPC community. SC24 provided the perfect stage to showcase the groundbreaking potential of our technology – to the HPC community whose workloads it was designed for.
Yet, as I reflect on the week, I am struck by a mix of emotions. On the one hand, it was an incredibly successful conference for us at NextSilicon, packed with countless productive meetings with customers and partners. As thrilling as it was to see our technology spark so much excitement, I couldn’t help but feel a bit of sadness. I missed the spontaneous moments that make conferences like these so special—dropping in on a talk or a BoF, running into an old friend for coffee, or simply roaming the floor and soaking up the community spirit. These moments remind me of why I fell in love with HPC in the first place. Still, SC24 was a monumental event, and I’m excited to share some highlights with you.
From the moment the doors opened, the energy and excitement were palpable. Seeing the reactions to our Maverick-2 ICA launch reinforced why we do what we do—to push the boundaries of high-performance computing and enable our customers to solve their most complex challenges. Over the course of the event, we had the privilege of connecting with an incredible array of customers, partners, analysts, investors, and media. The enthusiastic discussions with everyone I spoke to, further validated the importance of the need for a new way forward for HPC and AI—that NextSilicon is uniquely positioned to lead.
A key takeaway from these interactions was how the potential of the Maverick-2 ICA—and the new possibilities it unlocks—truly resonated with customers. They consistently shared how traditional architectures are increasingly struggling to deliver efficient performance at scale. The limitations of GPU architectures underscore the urgent need for a more flexible, adaptable, and intelligent acceleration approach. Customers repeatedly expressed their challenges with vendor lock-in, the burden of porting code, and the lack of architectural adaptability to support converging HPC and AI workloads. Maverick-2’s ability to empower developers to bring their own code and move beyond these constraints is striking a powerful chord, tapping into their long-standing and unmet demand for a fresh approach. The future of computing demands innovation that addresses these pain points, and NextSilicon is proud to deliver solutions that meet these needs. Sharing our vision at SC24 was deeply rewarding, as it validated the need for a fresh approach to enabling modern HPC and AI workloads. The rapidly evolving algorithms, models, and methodologies surrounding these workloads demand systems that can adapt with them. Maverick-2 stands out not just for its groundbreaking performance but its novel architecture and intelligent adaptability that will empower researchers to innovate for years to come. At NextSilicon it has always been our mission to help researchers break free of the limits of compute to unlock human knowledge and to deliver a more sustainable, and collaborative future. The discussions at SC24 reaffirmed this mission for me and that with our Maverick-2 ICA, we have taken a major step forward in helping make that vision a reality for our customers.
Honoring the Legacy, Shaping the Future
As it goes with supercomputing conferences, these shows are as much if not more about the vibrant community as it is about the technology. To that end, we were fortunate to welcome some of the most influential minds in the field to our booth. A heartfelt thanks to someone I am proud to consider my friend and industry colleague, David Barkai, who shared poignant reflections on 50 years of supercomputing progress from his book Unmatched. His insights into the architectural eras of HPC—from vector processors to accelerators and cloud computing—serve as a powerful reminder of the incredible journey our industry has undertaken. Likewise, as a software engineer, it was an honor to host John L. Gustafson. His pioneering work on Posit computing—featured in his book Every Bit Counts: Posit Computing—is inspiring new ways of thinking about numerical representation and efficiency. The contributions pioneers like David and John have made are the cornerstone of HPC’s legacy. I – like many others – am humbled to stand on the shoulders of such giants as we help write the first chapter for the next decade of supercomputing.
As I look back on SC24, I’m filled with gratitude for everyone who contributed to making this event a success. To our team, customers, partners, investors, analysts, and the media—thank you for your passion, ingenuity, and collaboration. The future of HPC is bright, and NextSilicon and our Maverick-2 ICA are poised to play a central role in shaping it. As we look ahead, I’m excited about the opportunities to deepen our partnerships, accelerate innovation, and tackle the challenges that matter most to our customers and the world.
Here’s to the next chapter of HPC and AI—and to continuing this incredible journey together!
Stay Connected
Stay tuned as we share more details in future blogs and announcements via our Nextletter. Sign up here
About the author:
Elad Raz is the founder and CEO of NextSilicon, a company pioneering a radically new approach to HPC architecture that drives the industry forward by solving its biggest, most fundamental problems.
