Tech News Podcast
A / 文 Search
A / 文 Tech News Podcast Search

Kingston Launches 30.72TB Gen5 NVMe SSD for Data Center Workloads

•

Tech News Podcast
Listen to the article

Kingston’s new 30.72TB DC3000ME SSD shows how enterprise storage is adapting to denser AI, cloud, analytics and edge workloads.

Kingston Digital has officially launched a 30.72TB version of its DC3000ME Gen5 U.2 NVMe SSD, expanding its lineup of enterprise storage solutions tailored for modern data center demands. This new SSD model is designed to meet increasing requirements for higher storage density and performance in servers supporting AI, cloud computing, and analytics workloads. The announcement comes directly from Kingston’s April 27, 2026, press release distributed via Business Wire.

This development is significant because data centers today require not just more computing power but also faster, denser storage architectures to handle the massive datasets behind AI training, inference, real-time analytics, and large-scale cloud services. Kingston’s DC3000ME 30.72TB drive leverages PCIe 5.0 NVMe technology while maintaining backward compatibility with PCIe 4.0, allowing operators to deploy it in mixed-infrastructure environments without wholesale server upgrades.

According to Kingston’s official product specifications, the DC3000ME series is available in capacities ranging from 3.84TB up to 30.72TB, with this largest model achieving sequential read speeds up to 14,000MB/s and sequential write speeds up to 9,700MB/s.

Enterprise SSDs like Kingston’s DC3000ME are critical components behind the scenes, enabling the high-speed data movement that AI and cloud applications demand.

For business and IT operators, this release means more than just a large SSD capacity. It represents an evolution in how enterprise storage is architected to support increasingly data-hungry applications without requiring completely new server platforms. Higher-density drives can consolidate storage infrastructure, reduce physical footprint in data centers, and improve throughput to AI accelerators and analytic engines.

Cameron Crandall, Kingston’s data center SSD business manager, has stated that customers are aiming to maximize storage density amid scaling AI, cloud, and high-performance computing workloads. This feedback framed the development of the DC3000ME series, underscoring the market need.

This announcement echoes a wider infrastructure trend: the AI era depends heavily on storage subsystems as well as GPUs and accelerators. Storage bottlenecks can limit data flow, resulting in performance constraints for model training and other workload pipelines. Solutions like the DC3000ME contribute to the foundational hardware enabling robust cloud services, enterprise software, and streaming platforms by maintaining low latency and high reliability.

What remains to be seen is the drive’s pricing, its performance in independent benchmarks and real-world deployments, and the speed of adoption among data center operators. Kingston’s current materials confirm specifications and availability but do not include external validation or volume shipment data. The eventual market traction will clarify whether this model becomes a mainstream enterprise upgrade or serves niche high-density use cases.

In summary, Kingston’s 30.72TB DC3000ME Gen5 NVMe SSD arrives at a time when faster, denser storage is essential to underpin next-generation AI, cloud, and analytics workloads. Its combination of cutting-edge PCIe 5 technology, enterprise features, and large capacity offers new options for data center design and operation. The next step is observing how customers incorporate this technology into their evolving infrastructure landscapes, balancing cost, performance, and compatibility requirements.

More Stories

More in

  • Foxconn Launches Second-Generation LEO Satellites on SpaceX Falcon 9

    Foxconn Launches Second-Generation LEO Satellites on SpaceX Falcon 9

    Listen to the article

    Foxconn launched PEARL-1A and PEARL-1B into low Earth orbit, extending its electronics manufacturing strategy into satellite infrastructure.

    Foxconn has successfully launched its second-generation low-Earth-orbit (LEO) satellites, PEARL-1A and PEARL-1B, aboard a SpaceX Falcon 9 rocket from California. Both satellites entered their intended orbits and are planned to undertake on-orbit missions for five years, primarily aimed at verifying payload technologies in communications and space science.

    This development is significant not only as a spaceflight milestone but also because Foxconn is principally recognized as a global electronics manufacturer rather than a traditional satellite operator. By sending these experimental satellites into orbit, Foxconn extends its role into a more integrated position within satellite infrastructure, blending manufacturing expertise with communications and space systems.

    The immediate goal is not to launch a consumer-facing satellite internet service but to support industrial-scale technology validation. With these two spacecraft in orbit, Foxconn and its research partners gain a crucial platform to test how payload technologies perform in space, assessing aspects such as communication capabilities, satellite-to-satellite interactions, and the influence of the space environment on components.

    Foxconn’s deployment of PEARL-1A and PEARL-1B underlines a pivotal move by an electronics manufacturing giant toward participating directly in the satellite communications ecosystem, reflecting broader shifts in connectivity infrastructure planning.

    Reuters reported on May 3, 2026, that Foxconn confirmed the satellites’ successful insertion into intended orbits aboard a SpaceX Falcon 9. SpaceX’s CAS500-2 mission manifest lists the PEARL satellites as part of a rideshare deployment facilitated by Exolaunch, illustrating a cost-efficient launch approach where multiple payloads share a single rocket flight.

    The PEARL program is steered by Hon Hai Research Institute, tying into ongoing research on LEO communications and non-terrestrial networks (NTN), which are expected to be critical elements in future 6G connectivity frameworks. The LEO satellites are envisioned as complementary infrastructure to terrestrial networks, extending coverage to geographically challenging or underserved areas.

    Technical documentation from the Small Satellite Conference describes PEARL-1A and PEARL-1B as 6U XL CubeSats developed with National Central University’s integration. Their payloads encompass inter-satellite links, a Compact Ionospheric Probe, and Perovskite Solar Cells, with Ka-band communications facilitating both space-to-Earth and inter-satellite experiments.

    These payloads emphasize technology validation rather than a finished commercial product. Inter-satellite links are fundamental research for dynamic mesh networks in orbit, ionospheric probes improve understanding of radio signal behavior, and advanced solar cells assess energy efficiency and durability in space conditions.

    For users and enterprises, the evolution of such satellite infrastructures could eventually translate to enhanced connectivity in remote communities, marine environments, emergency networks, industrial sites, and logistics operations. Such applications depend on advances in technical reliability, standards, regulatory approvals, ground equipment availability, and viable commercial business models.

    Foxconn’s entry into the satellite manufacturing and validation arena poses strategic questions about the role of electronics manufacturers in the expanding space infrastructure market. Precision manufacturing, integrated components, and systems engineering are critical to satellite performance, areas where companies like Foxconn hold substantial expertise. This launch exemplifies a convergence of traditional electronics manufacturing with satellite communications emerging as a key connectivity layer.

    However, important questions remain unanswered in public records. There is no confirmation of an active commercial satellite service linked directly to PEARL-1A or PEARL-1B, nor details on customer engagements, capacity, pricing, or timeline for deployment to end-users. The clearest verified fact is Foxconn’s successful orbital deployment of these experimental satellites for technology validation.

    Going forward, Foxconn’s PEARL program transitions from development and launch phases into operational testing and evaluation in orbit. The principal future milestone will be the public release of mission data, communication performance metrics, and whether experimental results lead to a defined civilian satellite infrastructure offering.

    Read full article: Foxconn Launches Second-Generation LEO Satellites on SpaceX Falcon 9
  • U Power Plans Hydrogen-Backed Energy Venture for Thai Data Centers

    U Power Plans Hydrogen-Backed Energy Venture for Thai Data Centers

    Listen to the article

    U Power plans a hydrogen-backed AI energy management joint venture for Thailand’s data center market as power demand rises.

    U Power Limited announced on April 27, 2026, that it has entered into an agreement with Guofu Hydrogen Energy (Hong Kong) Development Co., Limited and Cloud Digital Chain Limited to form a U Power-controlled joint venture. This new venture aims to develop AI-driven energy management solutions tailored for intelligent data centers, starting in Thailand. U Power is expected to hold the majority equity stake, with plans to provide capital for Hong Kong subsidiaries by July 31, 2026, and to begin operations within approximately 90 days thereafter.

    This initiative highlights an emerging trend where growing data center demand, driven by cloud computing and artificial intelligence services, is prompting energy firms to explore hybrid power systems. These systems integrate AI management tools, hydrogen energy infrastructure, and grid support to optimize electricity consumption and support sustainable operations.

    For everyday users, the development is significant because data centers underpin many digital services such as streaming, online payments, cloud storage, and mobile applications. Increasing data center energy needs impact electricity grids, infrastructure planning, digital service costs, and environmental outcomes.

    The venture represents a convergence of rising digital demand and innovative energy solutions, seeking to enhance data center sustainability and operational efficiency.

    U Power indicated that this joint venture will leverage its AI-integrated energy platform, combine Guofu Hydrogen Energy’s expertise in hydrogen equipment, and Cloud Digital’s background in data center power supply solutions. Initially focusing on Thailand, the company plans to expand gradually into other markets over time.

    Thailand has become a noticeable data center market in Southeast Asia. Market research estimates Thailand’s data center market size at approximately $1.45 billion in 2025, expecting growth to over $6 billion by 2031, with a compound annual growth rate nearing 28%. The country’s Board of Investment reported 36 data center project applications in 2025, representing investments above $23 billion, reflecting strong momentum in digital infrastructure development.

    Hydrogen energy’s application to data centers remains a complex subject. While hydrogen can serve as a low-carbon fuel carrier and backup energy source, its environmental and economic viability depends on factors such as production methods, storage, transport, and integration. Key considerations for data center operators include reliability, uptime, cost, fuel availability, and emissions accounting.

    The joint venture plans to apply AI as the central management layer overseeing energy demand and supply within data centers. AI energy management involves predictive analytics, load balancing, and automation to optimize power use amid fluctuating computing workloads and cooling requirements. This can improve efficiency and support grid resilience.

    Beyond data centers, U Power’s broader business scope covers hydrogen refueling and commercial vehicle applications. However, the focus of this collaboration remains the civilian data center energy sector due to its direct influence on daily digital services and infrastructure.

    While the framework and intentions are clear, several uncertainties remain. The venture’s success depends on finalizing incorporation, securing customers, achieving commercial viability, and proving the cost competitiveness of hydrogen-powered energy systems in the Thai market.

    The next key milestones will be the formal establishment of the joint venture entities before the summer deadline and the commencement of operational energy management projects at data centers. Close observation will be necessary to evaluate how the AI-driven hydrogen hybrid model performs in practice and whether it can become a scalable solution for the expanding Southeast Asian data center ecosystem.

    Read full article: U Power Plans Hydrogen-Backed Energy Venture for Thai Data Centers
  • Meta Adds Noon Energy Storage Deal for 24/7 Data Center Power Strategy

    Meta Adds Noon Energy Storage Deal for 24/7 Data Center Power Strategy

    Listen to the article

    Meta reserved up to 1 GW / 100 GWh of Noon Energy storage as AI data centers push clean power strategies toward long-duration systems.

    Meta has announced a partnership with Noon Energy to reserve up to 1 gigawatt (GW) and 100 gigawatt-hours (GWh) of ultra-long-duration energy storage aimed at powering its next-generation AI data centers. The agreement, disclosed on April 27, 2026, marks a significant step in Meta’s broader strategy to secure cleaner and more reliable electricity for its digital infrastructure.

    This commitment begins with an initial pilot project of 25 megawatts (MW) and 2.5 GWh capacity, which is expected to be completed by 2028. The pilot will serve to evaluate the performance and viability of Noon Energy’s storage technology before potentially scaling up to the full 1 GW / 100 GWh capacity as per the reserved contract.

    As AI workloads grow rapidly, technology companies like Meta face an increasing challenge: their data centers require continuous power, yet renewable sources such as solar and wind produce electricity intermittently. This makes energy storage a critical component for bridging the gap and enabling 24/7 clean power supply.

    Renewable energy generation alone cannot guarantee uninterrupted data center operation without storage systems capable of delivering energy across longer time periods.

    Noon Energy’s technology is designed for ultra-long-duration energy storage exceeding 100 hours. It utilizes modular, reversible solid oxide fuel cells paired with carbon-based storage methods, differing notably from conventional lithium-ion batteries used for shorter-duration applications. This approach reduces dependency on scarce metals and aims to facilitate scalable, cost-effective storage solutions.

    For everyday users and businesses relying on AI and cloud services, the underlying infrastructure’s energy source profoundly affects sustainability and reliability. Data centers worldwide consumed approximately 415 terawatt-hours (TWh) in 2024, representing about 1.5% of total global electricity use. The International Energy Agency projects this consumption could more than double to 945 TWh by 2030, driven largely by AI server demand.

    The U.S. Department of Energy has underscored how expanding data center facilities, particularly for AI applications, impact regional electricity grids. These centers’ substantial loads and need for dependable power make firm and flexible energy resources essential for grid stability.

    Meta’s Noon Energy partnership complements another initiative with Overview Energy for space-based solar power, collectively aimed at ensuring clean, reliable power for its AI infrastructure. Noon Energy characterizes this deal as one of the largest ultra-long-duration storage commitments presently announced, reflecting the increasing priority of such technologies in tech industry energy planning.

    Despite the promising potential, certain uncertainties remain. It is not yet verified how the pilot project will perform at hyperscale data center requirements, or whether the full-scale contract will be fulfilled on the proposed timeline. Financial details of the deal have not been publicly disclosed.

    As Meta transitions from announcement phase to practical implementation, the broader AI and data center industry watches closely. The emergence of ultra-long-duration storage could prove pivotal in meeting the growing energy demands of AI while enabling a cleaner power footprint.

    In summary, Meta’s investment in extended-duration energy storage signals a shift in data center energy strategy, underscoring the importance of integrating large-scale storage solutions alongside renewable generation to achieve ambitious sustainability and operational goals.

    Read full article: Meta Adds Noon Energy Storage Deal for 24/7 Data Center Power Strategy
  • Nature Medicine Examines Digital Twins and In Silico Trials in Drug Development

    Nature Medicine Examines Digital Twins and In Silico Trials in Drug Development

    Listen to the article

    Nature Medicine examines digital twins and in silico trials as drug development moves toward more computational evidence.

    On April 27, 2026, Nature Medicine published a Comment article titled “The arrival of digital twins and in silico trials in drug development,” authored by Ashley L. Eadie, Holly Fernandez Lynch, Naomi Scheinerman, and Ravi B. Parikh. This piece places computational patient models and in silico trial methods at the forefront of ongoing discussions on how future medicines might be evaluated and approved. The article situates these technologies within drug development, regulation, experimental disease models, and health policy contexts.

    This publication is significant because digital twins and in silico trials are moving beyond theoretical concepts and laboratory research into practical tools with potential to influence pharmaceutical study design, safety evaluation, patient response modeling, and preparation of regulatory evidence. Nature Medicine emphasizes that the regulatory landscape around drug evaluation is evolving, with digital tools requiring coordinated input from regulators and public agencies to establish their trustworthiness as evidence for drug approval decisions.

    In the realm of health technology, a digital twin typically refers to a computer-based representation of a patient, an organ, a disease process, or a biological system that integrates data and simulations. In silico trials use computational models to simulate aspects of clinical or preclinical studies. These approaches do not aim to outright replace human evidence but their value depends critically on transparency, validation, clinical relevance, and suitability for particular regulatory questions.

    The key challenge lies in ensuring that these complex models align with real-world patient data and can be independently evaluated before serving as reliable decision-making tools.

    The timing of this development is pertinent. The U.S. Food and Drug Administration (FDA) has been expanding its framework around New Approach Methodologies (NAMs), which include in silico modeling and other advanced human-relevant techniques designed to assess safety, efficacy, and product quality. The FDA states that NAMs could improve the predictive relevance of nonclinical testing and reduce or replace certain animal experiments.

    This regulatory context is not simply theoretical. In March 2026, the FDA released draft guidance to assist drug developers in validating NAMs for potential use as alternatives to animal testing, underscoring a broader shift away from default reliance on animal studies. Simultaneously, the FDA's Center for Drug Evaluation and Research has noted a rise in submissions incorporating artificial intelligence (AI) components across drug development stages and issued draft guidance on the use of AI to support regulatory decision-making.

    For patients and healthcare systems, these computational advances hold practical promise. Improved models could enable earlier hypothesis tests, optimize clinical trial design, enhance dose selection, identify emergent safety concerns earlier, and increase study efficiency. However, challenges remain concerning informed consent, equity, data privacy, data quality, and legal and ethical governance.

    Drug developers face the pivotal question of whether virtual evidence can gain sufficient credibility to influence patient-centric decisions. FDA’s Model-Informed Drug Development Paired Meeting Program already provides selected developers opportunities to engage regulators on modeling approaches, including clinical trial simulation and predictive safety assessments. This initiative reflects regulatory engagement but stops short of embracing digital twins or simulated trials as routine regulatory evidence.

    Ultimately, the decisive factor is trustworthiness. Advanced models may nevertheless be unsuitable if they incorporate incomplete data, obscure assumptions, poor patient representativeness, or lack independent validation against biological and clinical evidence. The conversation initiated by Nature Medicine signals that digital twins and in silico trials are entering crucial regulatory discussions, with the next step involving establishing rigorous guidelines to determine when computational evidence can complement, refine, or potentially reduce traditional clinical studies while safeguarding patient safety and public trust.

    Read full article: Nature Medicine Examines Digital Twins and In Silico Trials in Drug Development

Podcast

Technology stories, made to listen.

Explore narrative episodes with context, clarity and perspective on the technologies shaping everyday life.

Podcast category for Tech News Podcast
Listen Now
Tech News Podcast newspaper print edition

Newspaper

Discover the stories that shape our world, better enjoyed on paper.

Our print edition delivers a calmer reading experience—curated analysis, in-depth features, and expert perspectives you can hold in your hands. Also available digitally, wherever you are.

Print on Amazon
Digital on Amazon