XCENA Showcases CXL Memory Innovations for AI Inference at OCP Global Summit 2026

– XCENA and SK hynix are jointly optimizing RDMA-based shared-memory performance; their collaboration demo was featured at the SK hynix booth

– MX1 demonstrations show a KV cache shared across inference servers and near-data acceleration of RAG vector search

– CPO Harry Kim presents on near-data processing and, with Samsung Electronics, on CXL memory architecture for agentic AI; also joins an open-middleware panel with Marvell, Samsung, SK hynix and Gottbrath Technologies

– Distinguished Engineer Grant Mackey joins Microsoft, Seagate and Meta for an OCP Composable Memory Systems (CMS) Sub-Project technical update on addressing the memory wall for AI and general compute platforms

SAN JOSE, Calif., October 15, 2026--XCENA, a fabless semiconductor company developing next-generation memory infrastructure for the AI era, demonstrated its CXL-based memory solutions for AI inference demonstrations at OCP Global Summit 2026, addressing AI inference's growing memory requirements and data-movement burden. The event was held Oct. 12–15 at the San Jose McEnery Convention Center. The demonstrations address two of the biggest constraints on AI inference: growing memory requirements and the cost of moving data.

XCENA showed two AI inference use cases built on MX1, its Compute Express Link (CXL)-based computational memory solution.

The first demonstration showed multiple AI inference servers sharing a KV cache through a CXL memory pool. A KV cache stores context a model has already processed so it can be reused when generating responses. Keeping it in a shared pool, rather than duplicating it on every server, reduces redundant copies and data movement across servers.

The second used MX1's near-data processing (NDP) capability to accelerate vector search, a core step in retrieval-augmented generation (RAG). MX1 runs the search in memory, where the data resides, rather than repeatedly moving large datasets to the host processor.

XCENA is also collaborating with SK hynix to apply a CXL hybrid memory architecture to AI workloads. Using InfiniteMemory®, MX1's SSD-backed memory expansion capability, the two companies confirmed that the architecture can deliver significantly greater memory capacity scalability than conventional CMM-D while delivering higher performance than SSDs, and are now validating its applicability in real-world AI serving environments. At the summit, SK hynix unveiled "CXL Hybrid Memory" built on the MX1 chip and showcased an RDMA-based architecture using it.