Search Results
Nov 26, 2025, 09:17 ET AI to Reshape the Global Technology Landscape in 2026, Says TrendForce
bottlenecks in transmission speed and power efficiency. To address these limitations, HBM and optical interconnect technologies are emerging as critical enablers of next-generation AI architectures.Current generations of HBM leverage 3D stacking and through-silicon via to significantly reduce the distance
More news about: TrendForce
Nov 26, 2025, 03:00 ET iQOO 15 Launches as a Comprehensive Flagship with Performance Beyond Next, Delivering a Revolutionary Gaming Experience
Samsung 2K M14 LEAD™ OLED Display makes its debut on iQOO 15 — ushering in a new era of display technology.With 2600 nits global peak brightness (HBM) and an astonishing 6000 nits local peak brightness, the display sets a new benchmark for clarity and brilliance. To ensure consistent visual quality
More news about: iQOO
Nov 26, 2025, 03:00 ET iQOO 15 Launches as a Comprehensive Flagship with Performance Beyond Next, Delivering a Revolutionary Gaming Experience
Samsung 2K M14 LEAD™ OLED Display makes its debut on iQOO 15 — ushering in a new era of display technology.With 2600 nits global peak brightness (HBM) and an astonishing 6000 nits local peak brightness, the display sets a new benchmark for clarity and brilliance. To ensure consistent visual quality
More news about: iQOO
Nov 23, 2025, 19:01 ET Harbour BioMed Advances Global Strategic Collaboration with AstraZeneca to Discover and Develop Next-Generation Biotherapeutics in Oncology
CAMBRIDGE, Mass., ROTTERDAM, Netherlands and SHANGHAI, Nov. 23, 2025 /PRNewswire/ -- Harbour BioMed ("HBM" or the "Company"; HKEX: 02142), a global biopharmaceutical company committed to the discovery and development of novel antibody therapeutics for immunology
More news about: Harbour BioMed
Nov 21, 2025, 14:27 ET Supermicro 於 Supercomputing 2025 展示 HPC 叢集與 AI 基建的未來
配備液冷的 NVIDIA GB300 NVL72：機櫃級解決方案，搭載 NVIDIA GB300 Grace™ Blackwell 超級晶片，單一機櫃可提供 72 個 NVIDIA Blackwell Ultra GPU 及 36 個 Grace CPU，每個 GPU 配備 279GB HBM3e 記憶體4U HGX B300 液冷伺服器機櫃（內置 CDU）1U NVIDIA GB200 NVL4 伺服器 (ARS-121GL-NB2B-LCC)：專為大規模 HPC 及 AI 訓練構建的高密度液冷運算節點。基於
More news about: 美超微電腦股份有限公司
Nov 21, 2025, 12:32 ET Supermicro ने Supercomputing 2025 में HPC क्लस्टरों और AI इन्फ्रास्ट्रक्चर के भविष्य प्रदर्शित किए
Grace™ Blackwell Superchips के साथ रैक-स्केल समाधान, जो प्रति रैक 72 NVIDIA Blackwell Ultra GPUs और 36 Grace CPUs प्रदान करने के साथ-साथ प्रति GPU 279GB HBM3e भी उपलब्ध कराता है।इन-रैक CDU के साथ 4U HGX B300 सर्वर लिक्विड कूल्ड रैक1U NVIDIA GB200 NVL4 सर्वर (ARS-121GL-NB2B-LCC)
More news about: Super Micro Computer, Inc.
Nov 21, 2025, 10:16 ET Semiconductor Manufacturing Equipment Market worth $344.36 billion by 2032 - Exclusive Report by MarketsandMarkets™
devices transition to 2.5D and 3D architectures, chiplets, heterogeneous integration, fan-out wafer-level packaging (FOWLP), and high-bandwidth memory (HBM) become increasingly complex. These next-generation architectures require highly precise die bonding, thermal management, underfill, bumping, and wafer-level
More news about: MarketsandMarkets
Nov 20, 2025, 21:59 ET Supermicro ने AMD Instinct™ MI355X GPUs की विशेषता वाले प्रदर्शन और दक्षता-संचालित एयर-कूल्ड AI समाधानों के अपने पोर्टफ़ोलियो का विस्तार किया
का विस्तार कर रहा है। ये सिस्टम मौजूदा उद्योग में स्टैन्डर्ड OCP Accelerator Module (OAM) का लाभ उठाते हैं। ये एक्सेलरेटड GPU सर्वर प्रति GPU 288GB HBM3e, 8TB/s बैंडविड्थ, और 1000W TDP से 1400W TDP तक की वृद्धि प्रदान करते हैं, जो एयर-कूल्ड 8U MI350X सिस्टम की तुलना में दोहरे-अंकों तक का अधिक प्रदर्शन
More news about: Super Micro Computer, Inc.
Nov 20, 2025, 17:33 ET Supermicro amplia seu portfólio de soluções de IA com resfriamento a ar, voltadas para desempenho e eficiência, e apresenta as novas GPUs AMD Instinct™ MI355X.
ar. Esses sistemas utilizam o módulo acelerador OCP (OAM), padrão da indústria já existente. Esses servidores com GPU acelerada contam com 288 GB de HBM3e por GPU, largura de banda de 8 TB/s e um aumento no TDP de 1000 W para 1400 W, aumentando em mais de 10% o desempenho em comparação com o sistema
More news about: Super Micro Computer, Inc.
Nov 20, 2025, 17:16 ET Supermicro breidt haar portfolio prestatie- en efficiëntiegedreven luchtgekoelde AI-oplossingen uit met AMD Instinct™ MI355X GPU's
servers. Deze systemen maken gebruik van de bestaande OAM (OCP Accelerator Module), een standaard in de sector. Deze versnelde GPU-servers bieden 288GB HBM3e per GPU, 8TB/s bandbreedte en met een boost van 1000W TDP naar 1400W TDP, die dubbelcijferige prestatieverbeteringen opleveren in vergelijking met
More news about: Super Micro Computer, Inc.
Nov 20, 2025, 06:37 ET 슈퍼마이크로, AMD 인스팅트 MI355X GPU 탑재 공냉식 AI 솔루션 출시
수 있게 됐다"고 덧붙였다. 슈퍼마이크로는 이번 10U 공냉식 서버 출시를 통해 수냉식 및 공냉식 기반 고성능 제품군을 다시 한번 확장했다. 해당 솔루션은 업계 표준 OCP 가속기 모듈(OAM)을 활용해 GPU당 288GB의 HBM3e 메모리와 8TB/s 대역폭을 제공한다. 또한 TDP가 1000W에서 1400W로 증가함에 따라 기존 8U MI350X 공냉식 시스템 대비 최대 두 자릿수의 성능 향상이 이루어져, 데이터를 더욱 빠르게 처리할 수 있다. 이번 제품군 확장으로 고객은 공냉식과 수냉식
More news about: Super Micro Computer, Inc.
Nov 20, 2025, 06:19 ET Supermicro（スーパーマイクロ）、AMD Instinct™ MI355X GPUを搭載した高性能・高効率な空冷AIソリューションを拡充
https://www.supermicro.com/ja/accelerators/amd Supermicroは、液冷および空冷の高性能製品ラインをさらに拡充し、新たに10U 空冷サーバーを発表しました。このサーバーは、業界標準のOCP Accelerator Module（OAM）を採用し、GPU あたり 288GB の HBM3e メモリーと 8TB/s の帯域幅を備えています。さらにTDPを1000Wから1400Wに引き上げることで、従来の空冷8U MI350X システムと比べ、二桁台の性能向上を実現し、データ処理をより高速化します。MI355X GPU搭載の10U サーバーがラインナップに加わったことで、空冷・液冷の両方のインフラにおいて、ラックあたりのパフォーマンスを大幅に向上させることが可能になりました。
More news about: Super Micro Computer, Inc.
Nov 20, 2025, 02:00 ET Supermicro擴大氣冷式效能與效率型AI解決方案產品組合可支援AMD Instinct™ MI355X GPU
Supermicro再次擴大液冷與氣冷式高效能產品線，並推出全新10U氣冷伺服器。這些系統採用現有的業界標準型OCP Accelerator Module（OAM）架構，並提供288GB HBM3e記憶體/GPU、8TB/s頻寬，而系統的熱設計功耗（TDP）也從1000W提升至1400W。與8U氣冷式MI350X系統相比，這些全新10U氣冷伺服器能達到兩位數的效能倍數提升，使客戶能更快速地處理資料。Supermicro MI355X GPU伺服器產品線的10U新機型，也可助力客戶在大規模氣冷或液冷基礎設施內，實現更高的每機架效能（Performance Per Rack）。
More news about: 美超微電腦股份有限公司
Nov 19, 2025, 17:34 ET WEKA 憑藉 NeuralMesh 上的 Augmented Memory Grid 衝破 AI 記憶體屏障
Token，支援更多並發用戶，並為長文本工作負載解鎖全新的服務模式。OCI 的裸機基礎設施配合高性能 RDMA 網絡及 GPUDirect Storage 功能，使其成為大規模加速推論的獨特平台。」 現今的推論系統面臨一項根本限制：GPU 高頻寬記憶體 (HBM) 速度極快但容量有限，而系統 DRAM 雖空間較大但頻寬不足。一旦這兩層記憶體飽和，鍵值緩存數據便會被清除，迫使 GPU 重新運算已處理過的 Token，造成運算週期、電力與時間的浪費。 WEKA 的 Augmented Memory Grid 透過在
More news about: WEKA
Nov 19, 2025, 16:05 ET Supermicro Expands Its Portfolio of Performance and Efficiency Driven Air-Cooled AI Solutions Featuring AMD Instinct™ MI355X GPUs
10U air-cooled servers. These systems leverage the existing industry standard OCP Accelerator Module (OAM). These accelerated GPU servers offer 288GB HBM3e per GPU, 8TB/s bandwidth, and with a boost from 1000W TDP to 1400W TDP, pushing up to double-digit more performance compared to the air-cooled 8U MI350X
More news about: Super Micro Computer, Inc.
Nov 19, 2025, 16:05 ET Supermicro Expands Its Portfolio of Performance and Efficiency Driven Air-Cooled AI Solutions Featuring AMD Instinct™ MI355X GPUs
10U air-cooled servers. These systems leverage the existing industry standard OCP Accelerator Module (OAM). These accelerated GPU servers offer 288GB HBM3e per GPU, 8TB/s bandwidth, and with a boost from 1000W TDP to 1400W TDP, pushing up to double-digit more performance compared to the air-cooled 8U MI350X
More news about: Super Micro Computer, Inc.
Nov 19, 2025, 12:03 ET Semiconductors in Server Market Set for Strong Growth as AI and Cloud Adoption Surge Through 2032 | Valuates Reports
technologies: Adoption of next-generation memory standards such as DDR5, HBM, and CXL is accelerating to support the massive data throughput required by AI servers. Memory technologies associated with high-speed DRAM and HBM are becoming critical components of server architectures.
More news about: Valuates Reports
Nov 19, 2025, 10:40 ET WEKA ने NeuralMesh पर Augmented Memory Grid के साथ AI मेमोरी बैरियर को तोड़ा
एक अद्वितीय प्लेटफ़ार्म बनाता है।" आज के इंफरेंस सिस्टमों को एक मूलभूत बाधा का सामना करना पड़ता है: GPU उच्च-बैंडविड्थ मेमोरी (HBM) असाधारण रूप से तेज है, लेकिन इसकी क्षमता सीमित है, जबकि सिस्टम DRAM अधिक स्थान प्रदान करता है, परन्तु बैंडविड्थ बहुत कम है। एक बार जब दोनों स्तर भर
More news about: WEKA
Nov 19, 2025, 10:00 ET Vieworks' TDI Line Scan Camera Awarded World Class Product of Korea
fully integrated imaging solutions that combine its industrial cameras with lenses and lighting systems, targeting high-value inspection markets such as HBM semiconductors and OLED displays. TDI technology remains a cornerstone of Vieworks' imaging innovation, enabling high-speed and
More news about: Vieworks Co., Ltd.
Nov 19, 2025, 10:00 ET Vieworks' TDI Line Scan Camera Awarded World Class Product of Korea
fully integrated imaging solutions that combine its industrial cameras with lenses and lighting systems, targeting high-value inspection markets such as HBM semiconductors and OLED displays. TDI technology remains a cornerstone of Vieworks' imaging innovation, enabling high-speed and
More news about: Vieworks Co., Ltd.
Nov 18, 2025, 19:30 ET WEKA Breaks The AI Memory Barrier With Augmented Memory Grid on NeuralMesh
platform for accelerating inference at scale." Today's inference systems face a fundamental constraint: GPU high-bandwidth memory (HBM) is extraordinarily fast but limited in capacity, while system DRAM offers more space but far less bandwidth. Once both tiers fill, key-value cache (KV
More news about: WEKA
Nov 18, 2025, 18:27 ET WEKA durchbricht die KI-Speicherbarriere mit Augmented Memory Grid auf NeuralMesh
in großem Maßstab." Heutige Inferenzsysteme stehen vor einer grundlegenden Einschränkung: GPU-Speicher mit hoher Bandbreite (HBM) ist außerordentlich schnell, aber in seiner Kapazität begrenzt, während System-DRAM mehr Speicherplatz, aber weitaus weniger Bandbreite bietet. Sobald
More news about: WEKA
Nov 18, 2025, 18:22 ET WEKA franchit la barrière de la mémoire de l'IA avec Augmented Memory Grid sur NeuralMesh
échelle. Les systèmes d'inférence actuels sont confrontés à une contrainte fondamentale : la mémoire à grande largeur de bande (HBM) du GPU est extraordinairement rapide, mais sa capacité est limitée, tandis que la DRAM du système offre plus d'espace, mais beaucoup moins de largeur
More news about: WEKA
Nov 18, 2025, 10:30 ET ASUS Unveils Comprehensive Storage Solutions at SC25
performance, the system enables ultra-fast AI inference alongside training. Built for efficiency and scalability, the system features industry-leading HBM capacity (288GB) and up to 8TB/s memory bandwidth, enabling rapid training and simulation of large AI models. while a direct GPU-to-GPU interconnect
More news about: ASUS Computer International
Nov 18, 2025, 10:06 ET WEKA Breaks The AI Memory Barrier With Augmented Memory Grid on NeuralMesh
platform for accelerating inference at scale." Today's inference systems face a fundamental constraint: GPU high-bandwidth memory (HBM) is extraordinarily fast but limited in capacity, while system DRAM offers more space but far less bandwidth. Once both tiers fill, key-value cache (KV
More news about: WEKA