Everspin Demonstrates World’s First CXL-Connected MRAM at SNIA Developer Conference 2026

Customers can run their own workloads on the working system to measure its benefit for their applications.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

Working platform delivers memory-semantic access with writes up to 100x faster than NAND-based solid-state storage and is ready for customer workload testing

CHANDLER, Ariz.--(BUSINESS WIRE)-- Everspin Technologies, Inc. (NASDAQ: MRAM), a leading developer and manufacturer of MRAM persistent memory solutions, announced today the successful demonstration of the world’s first Compute Express Link (CXL)-connected MRAM platform at SNIA Developer Conference (SDC) 2026. The demo added a fast persistent memory tier between DRAM and NAND storage that can reduce the time processors and accelerators spend waiting for data and help AI systems run at higher utilization.

The working system connects Everspin PERSYST MRAM to CXL-enabled platforms as persistent memory with memory semantics, combining CXL’s standard high-speed connectivity with MRAM’s ability to retain data without power. It creates a fast persistent buffer and cache tier with cache-line-level, nanosecond-class access and writes up to 100x faster than NAND-based solid-state storage. Data remains available through a power interruption or system restart without relying on battery-backed DRAM or having to first flush data to an SSD.

The demo uses a Supermicro AS-1116CS-TN server with an AMD EPYC 9355 32-core processor, an AMD Alveo U250 FPGA platform with a Wolley CXL controller and Everspin PERSYST 1GB DDR4 MRAM UDIMMs. The configuration provides an open standards-based platform for demonstrating coherent access to persistent MRAM through CXL. It also enables memory resources to be pooled and shared across hosts, allowing persistent memory capacity to be allocated where workloads need it. For AI systems, that creates opportunities for applications such as checkpointing and persistent key-value (KV) cache.

“Everspin has been working with data center customers for 15 years, and we continue to innovate as AI changes the memory needs of systems,” said Sanjeev Aggarwal, president and CEO of Everspin Technologies. “CXL gives us a standard way to bring persistent MRAM to a much broader range of compute platforms. We now have a working system so customers can run their own workloads and measure the benefit for their specific applications.”

At SDC, Everspin and Wolley demonstrated read and write access to MRAM through the CXL interface under demanding test conditions.

“Wolley enables memory and storage connectivity through CXL, and this demonstration with Everspin shows how persistent MRAM can be brought into that environment with memory semantics,” said Dr. Bernard Shung, CEO of Wolley. “Together, we are demonstrating a practical way to connect persistent memory through CXL and make it available to today’s compute platforms.”

The SDC demonstration builds on Everspin’s work to extend MRAM into new memory architectures, including persistent caching, write buffering and metadata applications. To learn more about Everspin MRAM for data center applications, visit the Everspin Data Centers page.

About Everspin Technologies

Everspin Technologies, Inc., is the world’s leading provider of Magnetoresistive RAM (MRAM). Everspin MRAM delivers the industry’s most robust, highest-performance, non-volatile memory for Industrial IoT, Data Centers, and other mission-critical applications where data persistence is paramount. Headquartered in Chandler, Arizona, Everspin provides commercially available MRAM solutions to a large and diverse customer base. For more information, visit www.everspin.com

Cautionary Statement Regarding Forward-Looking Statements

This press release contains forward-looking statements regarding future events or results. Forward-looking statements are identified by words such as “will,” “expects” or similar expressions and include, but are not limited to, statements regarding Everspin’s anticipated business plans and business strategy. These forward-looking statements are subject to known and unknown risks, uncertainties, assumptions, and other factors that may cause actual results or outcomes to be materially different from any future results or outcomes expressed or implied by the forward-looking statements, including, without limitation, the risks set forth under the caption “Risk Factors” in Everspin’s Annual Report on Form 10-K for the year ended December 31, 2025 filed with the SEC on March 5, 2026, as well as in Everspin’s subsequent filings with the SEC. Any forward-looking statements made by Everspin in this press release speak only as of the date on which they are made, and subsequent events may cause these expectations to change. Everspin disclaims any obligations to update or alter these forward-looking statements in the future, whether as a result of new information, future events or otherwise, except as required by law.

View source version on businesswire.com: https://www.businesswire.com/news/home/20260929225389/en/

Media Contact:
Everspin Technologies
Stephanie Quinn
T: 480-316-8370
E: stephanie.quinn@everspin.com

Source: Everspin Technologies, Inc.

Key Terms

compute express link technical

Compute Express Link (CXL) is a high-speed connection standard that lets processors, memory modules, and specialized accelerator cards share data and memory more directly and quickly than traditional links. For investors, CXL matters because it can boost server performance, reduce latency, and enable more flexible data-center designs—changes that can increase demand for compatible chips, servers, and cloud infrastructure, similar to how a faster highway improves trade and productivity.

mram technical

MRAM is a type of computer memory that stores data using tiny magnetic bits, so it keeps information even when power is turned off. Think of it as a digital sticky note that stays attached without electricity but can be rewritten quickly and many times. Investors care because MRAM can lower device power use, simplify product design, and offer faster, longer-lasting storage, potentially changing costs and competitive positions across electronics and data-center industries.

dram technical

A dram is a small, traditional unit used to measure either mass or liquid volume in manufacturing and pharmaceuticals — roughly 1.77 grams for a weight dram or about 3.7 milliliters for a fluid dram, similar to a small teaspoon or a couple of drops. Investors care because dram-based measurements affect drug dosing, packaging sizes, labeling compliance and raw-material usage, which in turn influence production costs, inventory counts and regulatory risk.

fpga technical

A field-programmable gate array (FPGA) is a type of computer chip whose internal wiring can be changed after it is made, allowing engineers to program custom hardware functions without designing a new chip. For investors, FPGAs matter because that flexibility lets companies quickly adapt products to new software, standards, or customer needs—like a toolbox that can be rearranged to build different machines—so demand and pricing can shift with trends in data centers, telecommunications, AI, and specialized electronics.

ddr4 technical

DDR4 is a generation of high-speed computer memory used in personal computers, laptops and servers to temporarily hold the data a processor is actively working on — think of it as a short-term workspace that lets devices run programs smoothly. Its prevalence, pricing and compatibility affect demand and upgrade cycles for PCs, servers and components, influencing revenue, margins and product road maps for hardware and chip makers.