HardwareFreaks

Tag: Supercomputing

  • The US government wants a working quantum computer by 2028 and quantum-resistant encryption by 2031

    The Quantum Race: America Sets Its Sights on a Working Quantum Computer by 2028

    The quest to harness the immense power of quantum mechanics is no longer confined to theoretical physics; it is rapidly becoming a strategic national priority. In a significant move to accelerate technological advancement, the United States government has issued a directive aimed at realizing practical, working quantum computing capabilities within the next few years.

    This ambitious initiative calls upon federal agencies to forge powerful partnerships with private industry and leading academic institutions. The objective is clear: to collaboratively develop and deliver a functional quantum computer capable of supporting critical scientific research by 2028. This goal positions quantum technology not just as an academic curiosity, but as a foundational tool for future innovation across numerous sectors.

    To ensure this monumental undertaking remains on track and defines the cutting edge of the field, the Department of Energy has been tasked with a crucial preliminary role. The DOE is responsible for identifying the specific technical benchmarks that will serve as the definitive standards for defining what a successful quantum system must achieve. This foundational work is essential for setting the roadmap for all participants involved in the development process.

    Quantum computing promises to revolutionize fields ranging from drug discovery and materials science to cryptography and artificial intelligence. By aiming for a functional machine by 2028, the government is betting heavily on this technology to unlock unprecedented computational power that can solve problems currently intractable for even the most powerful supercomputers.

    The collaboration required for this effort is as vital as the technology itself. By bringing together the resources of private companies—known for their innovative engineering and scale—and the deep theoretical knowledge of universities, the government is creating an ecosystem designed for rapid, high-impact development. This public-private synergy ensures that research is driven by both commercial viability and fundamental scientific rigor.

    The mandate isn’t just about building a machine; it’s about defining the architecture of the future of computation. By focusing on these specific technical benchmarks, federal agencies are laying the groundwork for an infrastructure that will benefit global scientific understanding and economic growth well into the late 21st century. The race is officially on to turn theoretical potential into tangible reality.

    Buy on Amazon

    Credit: TechSpot

  • China just built the world’s most powerful supercomputer – using Huawei chips and no GPUs

    China Claims HPC Crown with Surprise Debut of LineShine

    The world of high-performance computing has just witnessed a major shakeup, as the biannual TOP500 project released its latest edition, revealing the most powerful supercomputing systems globally. This latest ranking, announced at the prestigious ISC 2026 conference in Hamburg, Germany, didn’t just update the leaderboard; it introduced a game-changer to the field of high-performance computing (HPC).

    The new list is a critical barometer for measuring computational power and technological advancement, identifying systems that push the boundaries of what is possible in processing massive datasets. But this cycle brought with it an unexpected contender that has immediately captured the imagination of tech enthusiasts and industry analysts: LineShine.

    What made this revelation particularly electrifying was LineShine’s entry into the top spot. This powerful new machine, developed in China, is not just another addition to the HPC world—it represents a significant leap forward in national computational infrastructure and engineering capability. Its placement at number one on the global list signals a massive shift in the landscape of supercomputing power.

    LineShine‘s debut is more than just a numerical achievement; it is a powerful demonstration of innovation and ambition. It highlights the rapid acceleration of advanced computing development emerging from Asia, positioning China firmly at the forefront of building world-class computational systems.

    The performance metrics reflected in the TOP500 ranking speak volumes about the complexity and scale LineShine can handle. By entering the elite tier immediately, this Chinese-built system has successfully demonstrated its ability to tackle some of the most demanding computational tasks currently being explored across science, research, and industry.

    This stunning achievement underscores the global competition in the HPC arena and points toward a future where cutting-edge computational power is increasingly driven by ingenuity from around the world. The unveiling of LineShine solidifies its status as a formidable force in the ongoing quest for the world’s most powerful supercomputers.

    Buy on Amazon

    Credit: TechSpot

  • China tops the list of fastest supercomputers with a CPU-only behemoth, ending US champion El Capitan’s reign — 2.198 exaflops of performance without a single GPU

    In a major technological upset, China’s LineShine supercomputer has claimed the top spot on the prestigious TOP500 list, signaling a powerful shift in the global high-performance computing landscape. This achievement isn’t just about raw performance; it represents a bold statement about indigenous technological capability and a deliberate change of posture in an increasingly fractured geopolitical environment.

    LineShine stands out because it is fundamentally built on a domestic stack, achieving this zenith without relying on any Western accelerators or GPUs. Installed at the National Supercomputing Centre in Shenzhen and developed by the Shenzhen Cloud Computing Center, the system pushes 2.198 exaflops on the High Performance Linpack benchmark, setting a new precedent for what can be achieved using only Central Processing Units (CPUs).

    The machine’s architecture is a marvel of domestic engineering. It utilizes 13,789,440 cores of proprietary silicon and operates entirely on CPUs, achieving double-precision performance that eclipses competitors in the standard HPC testing regime. This achievement marks the first time a system based purely on domestically designed components has led the list, demonstrating sophisticated self-sufficiency.

    Beneath the surface, LineShine is a complex system built from the ground up. It utilizes the LingKun platform, featuring 20,480 compute nodes, each equipped with advanced LX2 processors based on Armv9 architecture. These nodes are tightly integrated through a proprietary interconnect and run on custom-developed operating systems, showcasing a complete, homegrown solution from core to OS.

    The system’s design incorporates high-bandwidth memory, featuring on-package HBM rated up to 4 TB/s, which offers a level of data flow comparable to high-end systems in Japan. This focus on integrated architecture demonstrates ingenuity in leveraging domestic silicon and advanced packaging techniques to maximize performance.

    While the achievement is undeniable in terms of raw FP64 throughput, the story goes deeper when examining where it truly competes. LineShine excels in workloads that reward massive memory access and communication, but the landscape shifts dramatically when considering specialized workloads like artificial intelligence training. In benchmarks designed to approximate AI processing, other systems utilizing Western accelerators maintain a significant advantage.

    This distinction is crucial: while LineShine leads on FP64 performance—the benchmark where CPU-only systems can genuinely contend with accelerators—it trails in mixed-precision benchmarks that define modern AI capabilities. This suggests that the current race to dominate high-end HPC is evolving beyond simple floating-point speed and into specialized acceleration.

    The emergence of LineShine forces a reevaluation of what constitutes supercomputing dominance. It underscores that technological leadership is not solely determined by access to external components, but by the capacity for innovative, domestic design and execution. The move to place this system on the TOP500 list signals a strong assertion that self-reliance in cutting-edge computation is achievable, regardless of external pressures.

    Ultimately, LineShine represents more than just a leaderboard entry; it’s a powerful demonstration of national technological ambition and ingenuity, positioning domestic computing capabilities firmly on the global stage.

    Credit: Tom’s Hardware

  • China’s LineShine supercomputer dethrones US’ El Capitan, secures first place in Top 500 list — first machine in the rankings to sustain more than 2 ExaFLOPS of double-precision performance using only CPUs

    China’s LineShine Supercomputer Redefines the Limits of Computing

    A new titan has entered the world of High-Performance Computing (HPC), and it’s hailing from China. The LineShine supercomputer has successfully dethroned El Capitan to claim the top spot globally, proving that domestic technological prowess can compete with the world’s most advanced supercomputing systems.

    The achievement is backed by staggering raw power. In the Linpack benchmark, LineShine delivered an impressive 2.198 FP64 ExaFLOPS. What makes this feat even more remarkable is that it achieved this level of double-precision performance using only CPUs—establishing it as the first machine in the Top 500 list to sustain over 2 ExaFLOPS using solely CPU power.

    This formidable system is anchored at the National Supercomputing Centre in Shenzhen and represents a powerful collaboration between the Shenzhen Cloud Computing Center and the National Supercomputer Center in Shenzhen. It operates by harnessing 13.79 million cores, interconnected by a proprietary LingQi network, and demands 42.2 MW of energy to perform its calculations.

    Under the hood, LineShine utilizes semi-custom 304-core LX2 processors built on the Armv9 instruction set architecture, clocked at 1.55 GHz. Each core is packed with advanced features like Arm SVE and SME units, designed to accelerate vector and matrix operations essential for both scientific computing and emerging AI workloads, supporting various data formats including FP64, FP32, BF16, FP16, and INT8.

    The memory architecture is equally ambitious, pairing 32 GB of on-package HBM with up to 256 GB of external DDR5 memory, providing massive bandwidth of up to 4 TB/s to feed the hungry processors.

    While the sheer FP64 performance is astounding, an analysis of efficiency paints a more nuanced picture. LineShine delivered 52.07 GFLOPS/W. Although this figure sits just below El Capitan’s benchmark of 60.94 GFLOPS/W, it still demonstrates exceptional power management capabilities.

    Crucially, when compared to other CPU-only supercomputers like Fugaku—which struggles to deliver efficiency between 14.78 and 16.84 GFLOPS/W—LineShine clearly pulls ahead in terms of performance per watt. This efficiency gap highlights the sophisticated engineering behind the LineShine architecture.

    Despite its dominance in traditional supercomputer tasks, the system shows limitations when tackling modern mixed-precision AI workloads. In the HPL-MxP test, LineShine achieved 7.92 mixed-precision EFLOPS, placing it behind giants like El Capitan, Frontier, and Aurora. This suggests that while raw computational muscle is massive, incorporating specialized accelerators remains key for maximizing performance in contemporary AI training and inference.

    Nonetheless, the very fact that a supercomputer developed domestically has achieved such extraordinary FP64 performance is a landmark achievement. Furthermore, LineShine‘s submission to the Top 500 ranking signals confidence in its homegrown technologies, affirming a pathway for independent supercomputing development free from external technological constraints.

    Credit: Tom’s Hardware