CXL memory pooling breaks AI's second wall, memory, lifting effective DRAM utilization from about 50% to 85% with no new chips and enabling 3x concurrent LLM instances. Astera Labs is the purest, priciest play at about 31x forward sales, with CXL driving 179% Q4 revenue growth. Marvell is diluted but has the most durable balance sheet. Penguin Solutions is cheap and re-rating into AI-inference memory. Micron offers CXL as free optionality at about 9x forward, the cheapest in big-cap semis. x.com  … Show more

Memory is the part of the AI stack with pricing power now, as scarcity migrates from GPUs in 2023 to memory today. HBM is sold out, contract prices keep repricing higher, and Nvidia raised server prices citing memory costs; models cannot run without the memory to hold them. That supports Micron, SanDisk and Penguin Solutions. Micron's blowout print is not fully priced in if its trajectory and execution continue, leaving some alpha. x.com  x.com 

Astera Labs looks relatively expensive because similarly fast-growing AI infrastructure names trade at much lower multiples. Broadcom and Nvidia are at 14x, Credo at 15x and Lumentum at 18x; comparable peers look 50-80% cheaper. Astera is a great company and not necessarily expensive in absolute terms. Its roughly 2x multiple over Credo may persist, but the setup looks harder; with no valuation support, it could fall 50% if weakness continues. x.com  x.com  x.com  x.com 

Penguin Solutions wins whoever wins the accelerator or model, because it packages raw memory into custom modules and builds the clusters they go into. Integrated Memory, now 57% of sales, more than doubled to $275M as record net sales rose 48% and guidance was raised. Advanced Compute looks flat only because of a pivot from low-margin hyperscaler work; non-hyperscale sales grew 81%. Photonics is near-free optionality. Risks: lumpy cluster sales, thin cash conversion, high DRAM prices. x.com 

Astera Labs and Montage Technology win the memory redesign rather than the shortage itself. They don't sell memory but determine how much memory everyone else needs to buy, a different kind of moat from that of memory producers. x.com 

Astera Labs carries a huge premium for flawless execution against a CXL market that is small near term. Switch-based CXL memory pooling capturing $15-16B in 2026-2028 is low probability: Marvell guides its XConn switch business to only about $100M by fiscal 2028, PCIe included. Pooling software likely stays immature into 2028, per a Samsung white paper, and Nvidia is contesting the warm KV-cache tier. Astera does CXL best, but SK hynix and Micron offer better risk-adjusted returns. x.com  x.com 

The AI memory wall is a structural bottleneck only beginning to be priced in, favoring companies that move data closer to compute. Training wastes about 90% of energy moving data, HBM4 uses 3x DDR5's wafer capacity, and DRAM and NAND spot prices rose 80–90%. Corsair holds DRAM bought at 2025 prices; Penguin's CXL-based MemoryAI server trades near 10–12x forward earnings with near-100% EPS growth; Rambus has HBM4E controllers and 80%-plus gross margins; Sanmina's Viking makes US-built memory. … Show more

Penguin Solutions benefits as more AI tightens DRAM and lifts demand for its memory systems, with CXL as a high-beta moonshot. It reaffirmed FY26 revenue and EPS at the high end of guidance, and its CEO says memory and AI infrastructure demand keeps growing. CXL lets GPUs use DRAM pools about 20x cheaper than HBM and 10-100x larger, easing the KV-cache bottleneck in long-context inference. Penguin is the most compelling high-beta CXL bet, though CXL is still tiny for its business. x.com  x.com 

Penguin Solutions looks more like memory-cycle strength dressed up as an AI pivot story, and its price has already run ahead of that story. Its legacy AI factory business was concentrated in hyperscaler hardware, and the pivot to enterprise, neocloud and sovereign customers is a rebuild that the numbers do not show yet, with longer sales cycles and delayed bookings. The IP is good, and winning qualifications could change the view, but that needs a real shift first. x.com 

AI inference is hitting a memory limit, and CXL switching, which pools memory across servers, is where the value map changes. HBM is costly and limited per chip, and bigger models outgrow one server. Marvell is the bet on switching as the control point, via XConn and Structera S, now in production. Penguin Solutions makes CXL deployable in its KV cache server. Micron's CXL DRAM is a structural tailwind still tied to DRAM cycles, and Credo sits in the competitive transport layer. x.com  x.com  … Show more