Stuttgart\/Garching, Germany \u2013 July 22, 2025 \u2013 Q.ANT announced it has delivered its Native Processing Server (NPS) to the Leibniz Supercomputing Centre (LRZ), \u201cmarking the first integration of an analog photonic co-processor into an operational high-performance computing (HPC) environment,\u201d the company said.<\/p>\n
\u201cThis milestone deployment enables LRZ to evaluate photonic acceleration for artificial intelligence (AI) and simulation workloads \u2013 offering a path toward significantly higher performance with drastically lower energy consumption,\u201d Q.ANT said.<\/p>\n
The collaboration between Q.ANT and LRZ redefines how data centers approach performance, footprint, energy-efficiency, and system architecture and marks a new chapter in computing \u2013 where analog precision, powered by light, addresses the scaling crisis of AI infrastructure., according to Q.ANT.<\/p>\n
\u00a0\u201cPhotonic processors offer a novel and promising path to accelerate AI and simulation workloads, while sharply reducing our environmental footprint. This deployment marks a milestone in our future computing mission to advance energy-efficient AI and high-performance computing,\u201d said Prof. Dr. Dieter Kranzlm\u00fcller, Chairman of the Board of Directors of LRZ. \u201cQ.ANT\u2019s NPS integrates easily into our existing infrastructure, allowing us to immediately evaluate its performance. Bringing in future technologies and opening up new avenues for our users would, however, not be possible without the continuous support from both the federal and Bavarian government.\u201d<\/p>\n
\u201cOur collaboration with LRZ marks a pivotal milestone: for the first time in history, we\u2019re operating photonic processors in an HPC under practical workloads. With this step, we demonstrate that light-based processors have moved beyond research and into real-world application,\u201d said Dr. Michael F\u00f6rtsch, CEO of Q.ANT. \u201cThis is a decisive step toward integrating photonic computing into the mainstream of next-generation computer architecture by 2030. The long-term commitment of the Ministry for Research, Technology and Space has been instrumental in enabling the breakthroughs that made this success possible.\u201d<\/p>\n
Artificial intelligence is increasing computing requirements and pushing high-performance data centres to their limits in terms of power consumption, heat generation and space. Because the photonic chip generates no heat, costly cooling measures are unnecessary. Thanks to the special properties of light, complex calculations can also be performed faster and more efficiently on the photonic processor. \u202fQ.ANT\u2019s photonic technology enables a new class of high-performance, energy-efficient server rack solutions, with:<\/p>\n
Up to 90x lower power consumption per workload,\u202fdue to the absence of on-chip heat and reduced cooling requirements.
\n100x increase in data center capacity enabled by greater computational density and faster execution of complex operations.
\n16-bit floating point precision with close to 100 % accuracy for all computational operations on the chip.
\nSeamless integration into existing infrastructure via standard PCIe interface and x86 software compatibility \u2013 including support for PyTorch, TensorFlow, Keras.<\/p>\n
Q.ANT\u2019s NPS relies on light instead of electricity for complex computing operations \u2013 such as training and inferencing AI models. It can be seamlessly integrated into existing data centers and is fully compatible with current hardware and software. Thanks to x86 compatibility, no adjustments to the software stack or changes to the server layout are necessary. The NPS consumes significantly less energy, generates no heat on the chip and enables a higher computing density. With this deployment, LRZ is strengthening its role as a pioneer in the development of energy-efficient high-performance computers.<\/p>\n
The collaboration is aimed at researching hybrid digital-analog architectures for future HPC environments and is funded by the German Federal Ministry of Research, Technology and Space. Deploying Q.ANT\u2019s NPS, LRZ expands this scope to include photonic analog computing, an option that was not technically feasible at the start of the program but made viable by Q.ANT\u2019s advancements in photonics. LRZ will use the Q.ANT NPS to establish new benchmarks and real-world use cases for applications such as climate modeling, real-time medical imaging, or materials simulation for fusion research.<\/p>\n
The first evaluation phase at LRZ involves installing several units of the latest generation Q.ANT NPS, selecting benchmark workloads, and testing real-world applications \u2014 particularly in AI inference, computer vision, and physics simulations. Later phases will include second- and third-generation NPS units for deeper evaluation.<\/p>\n
The project was made possible with national funding. The commissioning ceremony was attended by guests from politics and industry. During her welcome address, the Minister for Research, Technology and Space (BMFTR), Dorothee B\u00e4r, acknowledged the significance of this milestone for Germany and Europe.<\/p>\n
\u201cIntegrating Q.ANT\u2019s photonic processor into the Leibniz Supercomputing Centre ushers in a new computing paradigm, moving it from the laboratory to reality, and takes Europe a decisive step closer to an energy-efficient digital infrastructure. Germany is proud to support groundbreaking innovations that strengthen our scientific leadership and technological sovereignty. We support this collaboration because it demonstrates how public research institutions and private deep tech pioneers can collaborate to shape the future right here in Germany,\u201d said Dorothee B\u00e4r, the Federal Minister of Science, Technology, and Space. (To be approved)<\/p>\n
\u201cComputing with light instead of electricity \u2013 what once sounded like science fiction is now becoming reality. For the first time worldwide, a photonic AI accelerator is in operation at a data center here in Garching, Germany: up to 90 percent less energy consumption with 100 times the performance. This shows the tremendous potential of our research institutions \u2013 and what becomes possible when science, industry and government join forces.\u201d Markus Blume, Bavarian Minister of Science and the Arts.<\/p>\n","protected":false},"excerpt":{"rendered":"Stuttgart\/Garching, Germany \u2013 July 22, 2025 \u2013 Q.ANT announced it has delivered its Native Processing Server (NPS) to…\n","protected":false},"author":2,"featured_media":40281,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[46],"tags":[191,32542,32543,32544,32545,19651,74],"class_list":["post-40280","post","type-post","status-publish","format-standard","has-post-thumbnail","category-computing","tag-computing","tag-hpc","tag-leibniz-supercomputing-centre","tag-lrz","tag-photonic","tag-q-ant","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/40280","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/comments?post=40280"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/40280\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media\/40281"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media?parent=40280"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/categories?post=40280"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/tags?post=40280"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}