![]()
NextSilicon, a leader in next-generation computing solutions for AI and high-performance computing (HPC), has delivered four Maverick-2 accelerators to Los Alamos National Laboratory. The systems are installed in the Lab’s Darwin testbed, where researchers will evaluate how the technology runs the laboratory’s scientific applications.
Scientific computing has long asked researchers to reshape their codes around fixed hardware. NextSilicon and Los Alamos are taking the opposite approach: a processor that adapts to the application.
“For decades, scientists have reshaped their codes to fit the hardware. Our pioneering technology in the form of the Maverick-2 accelerator reverses that completely. We are honored to work with Los Alamos to realize the benefits of hardware that adapts to the science,” said Elad Raz, founder and CEO of NextSilicon.
Maverick-2 is an accelerator: a chip that works alongside a system’s CPU to speed up the most demanding parts of an application. It runs on NextSilicon’s Intelligent Compute Architecture, which reconfigures its hardware while a program runs so the chip’s resources match what the application needs at that moment. Unlike conventional CPUs and GPUs with largely fixed execution structures, Maverick-2 maps computational operations onto a reconfigurable execution fabric, optimizing hardware resources for different computational patterns.
What Los Alamos will do with Maverick-2
Researchers at the Lab will run multi-physics simulation codes that support the laboratory’s national security science on the four accelerators. Irregular memory access and complex control flow are long-standing challenges in these codes, and the Lab’s research on them motivated the search for alternative architectures.
“Working directly with technology developers is how we evaluate and affect new computing architectures for our mission. Maverick-2 gives our teams hardware to test with,” said Gary Grider, senior director for Computing Technologies at Los Alamos National Laboratory.
“Maverick-2 can be reconfigured for different application requirements, with tools available to examine how codes map onto the system,” said Galen Shipman, chief architect for Advanced Technology Systems at Los Alamos. “That information can inform both performance analysis and application co-design.”
About the collaboration
NextSilicon and Los Alamos have worked together for several years across multiple generations of NextSilicon technology, pairing NextSilicon’s expertise in reconfigurable computing with the Lab’s experience developing and analyzing large-scale scientific applications. Darwin is a set of small-scale systems where Los Alamos tests emerging hardware and works with vendors to refine it. The Maverick-2 work will focus on multi-physics applications and the NextSilicon toolchain.
About NextSilicon
NextSilicon builds software-defined hardware that adapts to the work, not the other way around. Its adaptable architecture reconfigures to each workload at runtime, so existing code runs faster and more efficiently with no rewrites, and keeps pace as models and algorithms change. Proven for more than eight years in high-performance computing and at Sandia National Laboratories, where it passed a production acceptance on unmodified code, NextSilicon is now bringing the same architecture to AI inference, for future-proof computing as power becomes the limit on scale.
About Los Alamos National Laboratory
Los Alamos National Laboratory is a federally funded research and development center with priorities set by the Department of Energy’s National Nuclear Security Administration (DOE NNSA) and key national strategy guidance. We execute work across all of DOE’s missions: national security, science, energy, and environmental management. Scientific and engineering capabilities developed through LANL’s stockpile research are part of what makes DOE and NNSA a science, technology, and engineering powerhouse for the nation.
View source version on businesswire.com: https://www.businesswire.com/news/home/20261116535776/en/
Media gallery