
SK hynix passes Samsung as South Korea's most valuable company — memory company surpasses valuation milestone on the back of HBM
• SK hynix's early and sustained investment in HBM has positioned it as the primary beneficiary of the AI infrastructure boom, allowing it to surpass Samsung in market value and emerge as the leading AI memory supplier. Meanwhile, Samsung has fallen behind in HBM market share and customer adoption, prompting a renewed focus on next-generation HBM development, yield improvement, and qualification with major AI customers to strengthen its position in the rapidly growing AI memory market.
Kioxia emerges as a major AI storage beneficiary
• Kioxia reportedly surpassed Toyota in market capitalization, highlighting investor optimism toward AI-driven storage demand. The company advanced its next-generation 332-layer BiCS10 NAND roadmap and continued evaluating future fab expansions. Kioxia increasingly relies on LTAs and long-term customer commitments to manage constrained NAND supply.
Sandisk strategically focuses on AI-driven capacity allocation and next-generation NAND roadmap
• The company is shifting capacity toward AI and enterprise storage markets while advancing next-generation BiCS NAND, high-capacity QLC SSDs, and long-term customer agreements to strengthen its position in the rapidly expanding AI infrastructure ecosystem. The prioritized high-capacity QLC NAND and enterprise SSD development, including its Stargate architecture and future ultra-large-capacity SSD roadmap, is targeting AI storage clusters and hyperscale datacenters.


To reinforce South Korea‘s leadership in the AI memory era, the government has launched a $520 billion public-private semiconductor investment program alongside Samsung and SK hynix, including four new fabs and dedicated HBM production facilities. The initiative highlights the industry's long-term confidence that AI-driven demand will continue to drive substantial memory capacity expansion and technology investment.
Micron–Anthropic strategic agreement to scale next-generation AI infrastructure
•Micron is securing long-term AI demand while evolving from a component supplier into a strategic AI infrastructure partner, leveraging HBM, DRAM, SSDs, and ecosystem investments to strengthen its position in the AI value chain. The collaboration with Anthropic provides both long-term demand visibility and an opportunity to co-develop AI-optimized memory and storage solutions for future large-scale AI deployments.

demand for legacy‑node SSDs is minimal within overall SSD shipments. Rapid expansion in AI‑related applications in 2026 is expected to drive sustained growth in demand for mainstream advanced‑node 3D‑type SSD products.
Flash Sufficiency: limited wafer starts and controlled output reductions initially constrain supply, while gradual capacity adjustments and demand normalization improve availability in 2H26.Despite improving supply conditions, the market is expected to remain relatively tight for most of 2026, preventing any meaningful price correction.
Price Trend: capacity reallocation toward higher-margin products tightens market availability and supports sustained price increases.Spot prices may remain slightly above contract prices during periods of short-term procurement urgency and inventory replenishment.
DRAM Sufficiency: capacity migration toward HBM and high-value AI memory products continues to constrain conventional DRAM output, limiting supply growth. Despite some quarter-to-quarter fluctuations, the overall supply environment remains tight.
Price Trend: DRAM prices are expected to remain on an upward trajectory throughout 2026. The wide gap between spot and contract price is mainly driven by urgent procurement demand, AI-related supply tightness, and limited spot market inventory, while contract prices are moderated by long-term agreements and quarterly negotiations, resulting in a more gradual increase.

Learn what an NVMe SSD is, how NVMe differs from SATA SSDs, and why NVMe storage is ideal for automation, industrial systems, and high-frequency data logging.

Learn what ECC DRAM is, how error correction works, and why it matters for data integrity, system stability, networking security, and mission-critical computing.

NVR 스토리지 선택 시 고려해야 할 핵심 사항: 안정적이고 고성능의 감시 시스템을 구축하기 위해 HDD, SSD, 감시용 SSD 및 AI NVR 스토리지 요구 사항을 비교합니다

온도는 모든 SSD에서 보이지 않는 제한 요소이며 SSD 온도 범위는 성능 사양이자 신뢰성 사양이기도 합니다. 지속적인 부하가 걸리면 NVMe 펌웨어는 종합 온도를 추적하고 가볍거나 강하게 스로틀링을 시작합니다. 그래서 쓰기 처리량은 크게 떨어지고 지연 시간은 급증합니다. 열은 NAND 내부의 전하 누설도 가속시켜 데이터 유지력 감소를 빠르게 만듭니다. 이 효과는 셀이 프로그램/지우기 마모로 노화될수록 더 심해집니다. 따라서 드라이브가 나중에 전원이 꺼진 상태로 남아 있을 때 내구성을 더 빨리 소모하고 데이터 보존 문제의 위험도 높아집니다.

SSD에서의 가비지 컬렉션은 컨트롤러가 조용히 수행하는 정리 작업으로, 여전히 유효한 페이지들을 새로운 공간으로 압축해 옮기고 이제 대부분 쓸모없어진 블록을 지우는 과정이며, NAND 플래시는 이미 프로그램된 페이지를 덮어쓸 수 없기 때문에 먼저 삭제가 필요하고 이 삭제는 페이지 단위가 아니라 블록 단위에서 이루어지므로 결국 업데이트는 “다른 곳에 먼저 쓰고 나중에 정리하는” 방식으로 이루어집니다.

모든 SSD 내부에는 OS가 인식할 수 있는 용량보다 더 많은 물리적 NAND가 존재합니다. 제조업체는 이 플래시의 일부를 사용자 용량으로 표시되지 않는 컨트롤러 전용 공간으로 예약합니다. 이와 같이 예약된 영역을 오버프로비저닝(OP)이라고 합니다. 예를 들어 TechTarget에 따르면 물리적 NAND 용량이 976GB인 SSD에서 호스트가 접근할 수 있는 용량은 800GB에 불과하며, 나머지 176GB는 컨트롤러 전용 오버프로비저닝 용량으로 남겨 둡니다. 또한 이 공간에는 FTL 매핑 테이블과 배드 블록 풀과 같은 내부 메타데이터가 저장됩니다. 따라서 SSD 오버프로비저닝이 무엇인지 묻는다는 것은 실제로 드라이브 내부에 있는 이러한 숨겨진 작업 공간을 의미합니다.