블로그
-
엔비디아의 역사적인 발자취
technologies/embedded/article/21178111/jon-peddie-research-nvidias-geforce-256-the-first-fully-integrated-gpu technologies/embedded/article/55268044/jon-peddie-research-electronics-history-nvidias-pascal-architecture-and-the-gtx
blog.naver.com · 2026.06.05
-
아마존, 알파벳, 마이크로소프트 모두 자체 AI 칩 개발 경쟁을 벌이고 있는 이유, 엔비디아에 주는 의미
이 글은 The Motley Fool의 기사 "Amazon, Alphabet, and Microsoft are all racing to design their own AI chips 아마존, 알파벳, 마이크로소프트는 각각 자체 인공지능 칩을 설계하고 데이터 센터에 적극 활용하고 있는데, 이는 엔비디아의 GPU 수요에 영향을 줄 수 있다는 우려가 나오고 있습니다.
blog.naver.com · 2026.06.09
-
Photonics War: TSMC First, or Samsung First?
A GPU performs calculations, while HBM stores the data needed for those calculations. The problem: current HBM–GPU connections rely on electrical I/O signals.
blog.naver.com · 2026.04.18
-
엔비디아 어디까지 갈 것인감
J's Chips Insight] 젠슨 황의 타이베이 선언: "GPU 회사는 끝났다, AI 인프라 회사가 왔다" - NVIDIA GTC Taipei 2026 키노트 전체 내용 심층 PlfUcxUrVKyNzKP1 NVIDIA GTC Taipei 2026 Keynote | Full Replay NVIDIA GTC Taipei 2026 keynote with Founder and
blog.naver.com · 2026.06.02
-
NVIDIA InfiniBand XDR
High Bandwidth for AI Data Centers Optech 1.6T and 800G OSFP224 transceivers provide the bandwidth required for GPU-to-switch, switch-to-switch and storage-to-compute connectivity in modern AI infrastructure.
blog.naver.com · 2026.05.14
-
맥북 네오 리뷰, 99만원 AI 노트북의 성능·게임·배터리 총정리
AI-driven hardware price increases, Apple’s affordable MacBook Neo delivers strong everyday performance and surprising gaming capability at a budget price. 2025년부터 2026년에 걸쳐 AI 수요로 인한 GPU 공급 부족이 반도체 가격 전반을 끌어올리면서
blog.naver.com · 2026.05.18
-
[VGA] GIGABYTE GV-R777OC-1GD (Rev. 2.0)
Powered by AMD Radeon HD 7770 GHz Edition GPU Integrated with the first 1GB GDDR5 memory and 128-bit interface Features Dual-link DVI-I / HDMI / mini DisplayPort * 2 Overclock to 1050 MHz Support CrossFireX™ and
blog.naver.com · 2026.04.26
-
800G-OSFP-DR8 (2xDR4)
High-Performance Optical CX8, H100 B200 AI Infrastructure This module is specifically designed to support advanced GPU and networking environments, including platforms based on NVIDIA ConnectX-8, NVIDIA H100, and NVIDIA
blog.naver.com · 2026.03.16
-
1.6T DR8 OSFP (MMS4A00-XM)
As AI clusters move toward larger GPU counts and higher east-west bandwidth, 1.6T optics are becoming AI/HPC connectivity—balancing performance, power efficiency, and interoperability.
blog.naver.com · 2026.02.04
-
ConnectX‑8 400G SuperNIC
offering a total of 800Gb/s network capacity, enabling high-bandwidth connectivity for large-scale GPU clusters and latency-sensitive workloads.
blog.naver.com · 2026.01.19
