OTC: NLST

Further, other volatile memory solutions are being developed to accommodate the industry’s increasing need for bandwidth, such as the development of technologies pioneered by Netlist, like DDR5 Dual Inline Memory Module (“DIMM”) intelligent on-module power management, and high-capacity High Bandwidth Memory (“HBM”). We expect continued industry focus on moving logic and management functionality away from host systems and onto – or into – memory itself.

Product Technologies

Our portfolio of proprietary technologies and design techniques includes:

Distributed Buffer Architecture

We invented the distributed buffer architecture that enables the buffering of data signals using multiple data buffer devices distributed between a memory module’s edge connector and its installed DRAM. The result was shorter data paths, improved signal integrity, and reduced latency compared to the industry-standard design for DDR3 load-reduced dual in-line memory module ("LRDIMM"). The memory industry has widely adopted our distributed architecture for DDR4 LRDIMM. Our HyperCloud product was our first LRDIMM product built on this innovative and proprietary distributed buffer architecture.

Localized Module-Based Power Management Architecture

Early in our company’s history of research and development efforts, we developed innovative solutions for improving memory performance by increasing the precision of voltage regulation. We rethought the way power management is handled for each memory module within a system, by moving power management capabilities away from the host system and onto the modules themselves. The result was improved

granularity of system power loading, which could translate to improved efficiency and higher data transfer speed. Relocating power management functionality onto modules allows multiple distinct devices on a single module to receive power and operate harmoniously. While many in the memory industry have now adopted our approach for their DDR5-based DIMMs, our NVvault products have incorporated localized power management for many years.

Design Expertise

We have designed special algorithms that can be implemented in stand-alone integrated circuits or integrated into other functional blocks in application-specific integrated circuits (“ASICs”). We utilized these algorithms in our HybriDIMM product to incorporate load reduction functionality. We also incorporated these algorithms in our NVvault product line, which is known in the industry as NVDIMM-N.

Proprietary PCB Designs

We utilize advanced techniques to optimize electronic signal strength and integrity within a printed circuit board (“PCB”). These techniques include the use of 10-layer or 12-layer boards, matching conductive trace lengths, a minimized number of conductive connectors, or vias, and precise load balancing to, among other benefits, help reduce noise and crosstalk between adjacent traces. In addition, our proprietary designs for the precise placement of intra-substrate components allow us to assemble memory subsystems with significantly smaller physical size, enabling original equipment manufacturers (“OEMs”) to develop products with smaller footprints for their customers.

Very Low-Profile Designs

We believe we were the first company to create memory subsystems in a form factor of less than one inch in height. Our innovative very low profile (“VLP”) DIMMs provide developers of server blades, storage bridge bay applications, telecommunications servers, switches and routers with a wide range of high performance memory options where efficient use of motherboard space is critical. Our technology has allowed us to decrease the system board space required for memory, and improve thermal performance and operating speeds, by enabling our customers to use alternative methods of component layout.

2" High-Profile Designs

We have designed expanded-size memory systems in a form factor of two inches in height. Our larger 2” tall DIMM designs and associated technologies can provide developers with solutions requiring greatly expanded DIMM capacities while still satisfying requirements for high bandwidth functionality.

Thermal Management Designs

We design our memory subsystems to ensure effective heat dissipation. We use thermal simulation and data to obtain thermal profiles of the memory subsystem during the design phase, allowing us to rearrange components to enhance thermal characteristics and, if necessary, replace components that do not meet specifications. We also develop and use proprietary heat spreaders to enhance the thermal management characteristics of our memory subsystems.

Compute Express Link Technology (“CXL”)

We are investing in new technologies like memory expansion utilizing Compute Express Link (“CXL”), which improves server capacity and bandwidth beyond what traditional memory channels can provide. Traditional memory channel bandwidth and capacities can no longer keep up with host central processing unit needs. CXL provides a range of features designed to cater to the evolving needs of high-performance data center computing and AI. As this space matures, we believe that CXL technology will allow larger memory pools to

be placed further away from the host central processing unit and still be seamlessly utilized with the traditional channel memory.

Products

Our commercially available memory subsystem products and other products that we sell include:

Component and Other Product Resales

Due to our relationships with memory channel customers, in addition to our own products, we resell certain component products that we purchase for the purpose of resale. We have purchased certain of these products under the terms of a product supply agreement with SK hynix, Inc., a South Korean memory semiconductor supplier (“SK hynix”). We have also sourced resale products from other suppliers. Additionally, we sell excess component inventory to distributors and other users of memory integrated circuits. For information regarding our concentrations and customers, see Note 10—Major Customers, Suppliers and Products of the Notes to Consolidated Financial Statements in