The module supports a bandwidth of 32GB/s and data transfer speeds of up to 32GT/s via a PCIe Gen5 x8 interface, providing the fast processing capabilities critical for AI workloads. By incorporating four 64GB CXL memory modules, a server configured with eight 128GB DRAM modules can increase its memory capacity by 30% and its bandwidth by 40%, meeting the demanding requirements of AI servers without needing additional DIMM slots. This optimizes the hardware architecture and reduces system complexity. Furthermore, the CXL memory module enables memory pooling, which optimizes the allocation of available memory among CPUs and components, significantly reducing the use of redundant memory and improving overall system efficiency.
Innodisk's CXL memory module uses the E3.S 2T form factor based on the EDSFF standard. This design allows for flexible memory expansion and facilitates module swapping between servers, ensuring seamless integration with minimal cost and complexity. CXL is an open standard promoted by leading companies in the industry and is expected to rapidly evolve into a comprehensive ecosystem, a critical aspect for cloud data center applications, network communications, and edge servers.
Innodisk plans to begin shipping this advanced CXL memory module in the first quarter of 2025.
This email address is being protected from spambots. You need JavaScript enabled to view it.
