Supercomputing 2025 (SC25) was bigger than ever, with more than 16,500 attendees and a record 559 exhibitors, including Phison. The team showcased a range of new solutions and technologies, and gained attention in the press for some standout milestones.\u00a0<\/span>\u00a0<\/span><\/p>\n<\/blockquote>\n \u00a0<\/span>Here\u2019s a closer look at what made SC25 such a successful event for Phison.\u00a0<\/span>\u00a0<\/span><\/p>\n The team introduced event attendees to Phison\u2019s newest Pascari enterprise SSDs, the X201 and D201, both designed to support modern workload and storage needs.\u00a0<\/span>\u00a0<\/span><\/p>\n As organizations push deeper into AI, analytics, and high-throughput services, the pressure on the storage layer has never been higher. Enterprises need fast, steady responsiveness and the ability to scale without introducing latency or unpredictability.<\/span>\u00a0<\/span><\/p>\n Phison\u2019s latest additions to the Pascari lineup were built with exactly those realities in mind. With up to 14.5 GB\/s read and up to 3.3 million IOPS, the Pascari X201 is engineered to keep pace with demanding, data-heavy operations, delivering the kind of sustained performance needed for AI pipelines and real-time processing. Although the Pascari D201 delivers the same level of performance as the X201, it includes a different advantage, which is exceptionally efficient density for cloud platforms and large object storage environments where every watt, rack unit, and dollar matters.<\/span>\u00a0<\/span><\/p>\n These drives build on the momentum created by Phison\u2019s earlier breakthroughs, including the ultra-high-capacity Pascari D205V Gen5 PCIe SSD in its distinctive E3.L format, now shipping to select OEM partners. With a long-term roadmap that pushes capacities toward the 245 TB range, the Pascari family is quickly redefining what\u2019s possible for enterprise flash storage at scale.<\/span>\u00a0<\/span><\/p>\n Pascari X201 and D201 enterprise SSDs should be available to select enterprise customers and OEM partners by the end of the year.<\/span>\u00a0<\/span><\/p>\n Get a look at the <\/span>Pascari X201<\/span><\/a> and <\/span>Pascari D201<\/span><\/a>.\u00a0<\/span>\u00a0<\/span><\/p>\n \u00a0<\/span><\/p>\n Phison played a critical role in a historic moment in data transfer history at SC25. The team partnered with VDURA, a leader in AI and HPC data-storage software infrastructure, to achieve the fastest data transfer on record.\u00a0<\/span>\u00a0<\/span><\/p>\n Reigning world deadlift champion Haf\u00fe\u00f3r \u201cThe Mountain\u201d Bj\u00f6rnsson was on hand to do the honors in front of a large crowd. Using VDURA\u2019s flash-optimized data storage infrastructure and Phison Pascari enterprise SSDs, he was able to transfer over 136 PB of information in a mere 27 seconds. The transfer reached a sustained transfer rate of over 5 PB per second, or 303 PB per minute.\u00a0<\/span>\u00a0<\/span><\/p>\n A recent <\/span>press release<\/span><\/a> about the event offered a few estimates to help readers comprehend that kind of blazing speed, stating that 5 PB per second is \u201cequivalent to transferring the entire Netflix library in under 30 seconds\u201d or \u201cmoving every YouTube video ever uploaded in just a few minutes.\u201d\u00a0<\/span>\u00a0<\/span><\/p>\n Last year, Bj\u00f6rnsson had thrilled SC24 crowds by deadlifting the largest amount of data on record, which was over 282 PB of data storage in the form of Phison Pascari enterprise SSDs. This year, he proved that data speed can be as impressive as his unprecedented strength.\u00a0<\/span>\u00a0<\/span><\/p>\n \u00a0<\/span><\/p>\n Phison also showcased its aiDAPTIV+ platform, showing how far AI can go on everyday hardware. In the live demo, a standard laptop equipped with an integrated GPU ran advanced AI agents with the help of aiDAPTIV+ GPU memory extension, proving that serious AI performance no longer requires a rack of servers filled with high-end GPUs.<\/span>\u00a0<\/span><\/p>\n The message was clear: aiDAPTIV+ is lowering the barrier to entry for AI across the board. IT teams, university researchers, and even large-scale cloud and enterprise environments can tap into dramatically faster inference without relying on costly, power-hungry GPUs. And for students and professionals, it opens the door to hands-on AI work on the devices they already own.<\/span>\u00a0<\/span><\/p>\n aiDAPTIV+ is purpose-built to supercharge AI agents on iGPU-powered laptops and desktops, delivering up to 25x faster performance for common AI agent applications. In one recent test, running a GenAI inference chat on YouTube video content, which once took 73 seconds, was completed in just four seconds.<\/span>\u00a0<\/span><\/p>\n In a recent interview with <\/span>VMblog<\/span><\/a> before SC25, Phison GM and President of Phison US Michael Wu was asked about the value proposition of Phison technology in relation to AI. His reply:\u00a0<\/span>\u00a0<\/span><\/p>\n \u201cDo more AI with less GPU. Phison’s aiDAPTIV+ lowers total cost by offloading memory pressure from GPUs onto high-capacity flash, improving price-performance and making on-premises AI practical. The result: faster time to value, predictable costs, and supply-chain stability from a partner with a 25-year track record and ultra-fast enterprise grade Pascari SSDs.\u201d<\/span>\u00a0<\/span><\/p>\n<\/blockquote>\n \u00a0<\/span><\/p>\n <\/p>\n SC25 proved to be more than a simple showcase for Phison. It was the company\u2019s statement of where AI and high-performance storage are headed. From redefining enterprise flash with the Pascari X201<\/a> and D201<\/a>, to setting a new world record in data transfer, to proving that powerful AI can run on everyday hardware with aiDAPTIV+, the event highlighted Phison\u2019s ability to innovate across the entire data pipeline.<\/span>\u00a0<\/span><\/p>\n The excitement from customers, partners, and the broader HPC community underscored a clear takeaway. Phison is pushing the boundaries of what\u2019s possible in speed, scale, and efficiency. And with even more breakthrough technologies on the horizon, the momentum from SC25 is only the beginning.<\/span>\u00a0<\/span><\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row disabled_on=”on|on|on” _builder_version=”4.16″ _module_preset=”default” width=”100%” max_width=”100%” custom_margin=”||||false|false” custom_padding=”0px||||false|false” disabled=”on” saved_tabs=”all” global_colors_info=”{}”][et_pb_column type=”4_4″ _builder_version=”4.16″ _module_preset=”default” global_colors_info=”{}”][et_pb_text _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n [\/et_pb_text][et_pb_toggle title=”What workloads benefit most from the Pascari X201?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n The Pascari X201 is engineered for high-velocity, low-latency environments where data throughput dictates compute efficiency. AI model training, real-time analytics, HFT engines, large-scale simulations, and GPU-accelerated pipelines benefit significantly. With up to <\/span>14.5 GB\/s read<\/b>, <\/span>12 GB\/s write<\/b>, and <\/span>up to 3,300K \/ 1,050K IOPS<\/b>, it feeds GPU clusters without bottlenecks, supporting faster epoch cycles, reduced compute idle time, and predictable QoS for latency-sensitive tasks.<\/span><\/p>\n [\/et_pb_toggle][et_pb_toggle title=”How does PCIe Gen5 improve AI training and inference performance?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n PCIe Gen5 doubles bandwidth compared to Gen4, allowing GPUs, vector databases, RAG pipelines, and simulation frameworks to ingest and stream data continuously. The X201 and D201 sustain <\/span>Gen5 throughput at scale<\/b>, preventing queue backlogs. For inference workloads\u2014such as recommendation engines or real-time decision systems\u2014this means lower inference latency and higher request concurrency without performance degradation.<\/span><\/p>\n [\/et_pb_toggle][et_pb_toggle title=”Why is low latency crucial for large-scale analytics and financial modeling?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n Analytics clusters ingest terabytes of telemetry, logs, or transactional data. In financial services, milliseconds impact risk scoring, fraud detection, and liquidity analysis. The X201\u2019s ultra-low latency, high IOPS profile ensures immediate random access to distributed datasets, supporting sub-second analysis cycles and stable response times under pressure. This improves both decision accuracy and computational efficiency.<\/span><\/p>\n [\/et_pb_toggle][et_pb_toggle title=”What makes the Pascari D201 suitable for cloud and hyperscale environments?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n The D201 combines Gen5 throughput with <\/span>15.36 TB density in the compact E1.S 15mm form factor<\/b>, enabling cloud operators to consolidate nodes, reduce rack footprint, and increase storage per RU. Its predictable QoS, 1\u20133 DWPD endurance, and power-efficient architecture align with hyperscale requirements for multi-tenant storage, object stores, backup tiers, CDN infrastructure, and large database clusters.<\/span><\/p>\n [\/et_pb_toggle][et_pb_toggle title=”How do enterprises decide between using the X201 vs. the D201?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n Use the <\/span>X201<\/b> when workloads require <\/span>maximum IOPS, minimal latency, and sustained throughput<\/b>, such as AI training labs, HFT systems, simulation clusters, and GPU-accelerated inference pipelines.<\/span> [\/et_pb_toggle][et_pb_toggle title=”How does Phison\u2019s in-house controller and firmware improve workload predictability?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n Phison integrates controller architecture, firmware, and NAND management in-house, ensuring full vertical control over latency behavior, QoS tuning, wear-leveling, and power optimization. This eliminates performance jitter, enhances sustained throughput, and ensures consistent response times during high-traffic conditions\u2014critical for AI pipelines, cloud multi-tenant environments, and database clusters.<\/span><\/p>\n [\/et_pb_toggle][et_pb_toggle title=”How does the X201 support GPU utilization efficiency in AI labs?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n GPU clusters often stall when storage cannot sustain streaming throughput. With <\/span>up to 14.5 GB\/s read<\/b>, high queue depth optimization, and extremely low latency, the X201 keeps GPUs fully saturated. This minimizes idle cycles, accelerates epoch completion, and shortens overall training time\u2014improving compute ROI for AI research centers and enterprise ML teams.<\/span><\/p>\n [\/et_pb_toggle][et_pb_toggle title=”How does the D201 help reduce data-center power and cooling requirements?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n The D201\u2019s <\/span>high density per drive<\/b>, power-efficient Gen5 controller, and compact E1.S thermals reduce the number of servers required to store petabyte-scale data. Fewer nodes mean lower cooling demand, reduced PSU consumption, and simplified operational management. For hyperscalers and cloud service providers, this translates directly into lower TCO and more energy-efficient scaling.<\/span><\/p>\n [\/et_pb_toggle][et_pb_toggle title=”What Phison technologies enhance AI readiness across enterprise and education?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n Phison\u2019s <\/span>Pascari<\/b> portfolio provides enterprise-grade SSDs for training, inference, and large-scale analytics. Complementing this, <\/span>aiDAPTIV+<\/b> extends GPU memory on iGPU-based PCs, enabling AI agents and inference workloads to run locally with up to <\/span>25\u00d7 faster performance<\/b> for common applications. This brings AI capability to businesses, universities, and developers without requiring discrete GPUs.<\/span><\/p>\n [\/et_pb_toggle][et_pb_toggle title=”How do Pascari SSDs future-proof infrastructure for next-generation workloads?” _builder_version=”4.27.4″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n Modern workloads demand not only higher performance but higher capacity per node. With X201 and D201 offering <\/span>Gen5 bandwidth<\/b>, <\/span>up to 30.72 TB<\/b> (X201) and <\/span>15.36 TB<\/b> (D201), plus robust endurance options, organizations can scale AI, analytics, and cloud services without hardware refresh bottlenecks. The portfolio\u2019s controller-level innovation ensures longevity, predictable QoS, and alignment with emerging AI\/ML architectures.<\/span><\/p>\n [\/et_pb_toggle][\/et_pb_column][\/et_pb_row][\/et_pb_section]<\/p>\n","protected":false},"excerpt":{"rendered":" Supercomputing 2025 (SC25) was bigger than ever, with more than 16,500 attendees and a record 559 exhibitors, including Phison. The team showcased a range of new solutions and technologies, and gained attention in the press for some standout milestones.\u00a0\u00a0 \u00a0Here\u2019s a closer look at what made SC25 such a successful event for Phison.\u00a0\u00a0 \u00a0 Phison […]<\/p>\n","protected":false},"author":76,"featured_media":87877,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_et_pb_use_builder":"on","_et_pb_old_content":"","_et_gb_content_width":"","inline_featured_image":false,"footnotes":""},"categories":[23,3],"tags":[22],"class_list":["post-87874","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-all-posts","category-enterprise","tag-long-content"],"acf":[],"_links":{"self":[{"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/posts\/87874","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/users\/76"}],"replies":[{"embeddable":true,"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/comments?post=87874"}],"version-history":[{"count":8,"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/posts\/87874\/revisions"}],"predecessor-version":[{"id":87894,"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/posts\/87874\/revisions\/87894"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/media\/87877"}],"wp:attachment":[{"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/media?parent=87874"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/categories?post=87874"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/phisonblog.com\/zh\/wp-json\/wp\/v2\/tags?post=87874"}],"curies":[{"name":"\u53ef\u6e7f\u6027\u7c89\u5242","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}\u00a0<\/span><\/h3>\n
Phison Pascari X201 and Pascari D201 enterprise SSDs make their debut<\/h3>\n
Phison teams up with VDURA for the fastest data transfer ever recorded<\/h3>\n
Do more AI with less GPU, thanks to Phison\u2019s aiDAPTIV+ technology<\/h3>\n
\n
![\"\"](\"https:\/\/phisonblog.com\/wp-content\/uploads\/2025\/11\/Microsite-Pascari-ASI_Banner_BLACK_01-1024x213.png\")
<\/a><\/p>\nA breakout year and a glimpse of what\u2019s next<\/h3>\n
Frequently Asked Questions (FAQ) :<\/strong><\/h3>\n
<\/span> Choose the <\/span>D201<\/b> for <\/span>capacity-optimized deployments<\/b> at cloud, CDN, or archival scale\u2014where density, predictable QoS, and cost-efficient scaling matter most. Many organizations deploy both within the same architecture: X201 for compute-adjacent storage and D201 for scalable data lakes.<\/span><\/p>\n