SanDisk Investor Presentations :
SanDisk’s growth is surging as agentic AI and real-time LLM inference shift memory demands toward high-capacity NAND flash and enterprise SSDs.
To support long context windows and persistent KV caching, SanDisk secured $93.9 billion in minimum contracted revenue commitments, targeting mid-to-high teens annual revenue growth through 2030.
Drivers of Agentic AI Growth
Storage Demand:
Agentic workflows require continuous, large-scale memory access and rapid retrieval, driving server eSSDs to absorb over half of global NAND bit shipments.
KV Cache Offloading:
Offloading persistent Key-Value caches to flash memory has become a critical architecture for handling long context reasoning.
High Bandwidth Flash (HBF):
Innovations like SanDisk's HBF and BiCS10 QLC NAND expand storage capabilities closer to compute, targeting an enterprise flash market projected to scale aggressively toward 2030.
Financial and Strategic Outlook
Business Models (NBM):
Long-term contracts with major cloud and AI infrastructure players cover roughly half of fiscal 2027 bit production and two-thirds of fiscal 2028 production at floor pricing, reducing historical cyclicality.
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
- buythedip·09:12That 93.9B floor sounds strong, but storage contracts love getting renegotiated when pricing rolls over. The real test is whether AI demand stays rigid enough to protect NAND marginsLikeReport
- JulianAlerander·09:12KV cache offload is the part I care about most here. If flash moves from cheap capacity to inference infrastructure, NAND ASP and margin both rerate fast.LikeReport
