Rambus has announced the availability of its PCI Express® (PCIe®) 6.0 Interface Subsystem comprised of PHY and controller IP. The Rambus PCIe Express 6.0 PHY also supports the latest version of the Compute Express Link™ (CXL™) specification, version 3.0.
“The rapid advancement of AI/ML and data-intensive workloads is driving the continued evolution of data center architectures requiring ever higher levels of performance,” said Scott Houghton, general manager of Interface IP at Rambus. “The Rambus PCIe 6.0 Interface Subsystem supports the performance requirements of next-generation data centers with premier latency, power, area and security.”
The Rambus PCIe 6.0 Interface Subsystem delivers data rates of up to 64 Gigatransfers per second (GT/s) and has been fully optimized to meet the needs of advanced heterogenous computing architectures. Within the subsystem, the PCIe controller features an Integrity and Data Encryption (IDE) engine dedicated to protecting the PCIe links and the valuable data transferred over them. On the PHY side, full support for CXL 3.0 is available to enable chip-level solutions for cache-coherent memory sharing, expansion and pooling.
“PCIe is ubiquitous in the data center and CXL will become increasingly important as companies pursue ever-escalating speeds and bandwidths to support higher levels of performance in next-generation applications,” said Shane Rau, research vice president, Computing Semiconductors at IDC. “As a growing number of chip companies emerge to support new data center architectures, access to high-performance interface IP solutions will be key to enabling the ecosystem.”
Key features of the Rambus PCIe 6.0 Interface Subsystem include:
- Support for the PCIe 6.0 specification including 64 GT/s data rate and PAM4 signaling
- Implementio of low-latency Forward Error Correction (FEC) for link robustness
- Support for fixed-sized FLITs that enable high-bandwidth efficiency
- Backward compatible to PCIe 5.0, 4.0 and 3.0/3.1
- State-of-the-art security with an IDE engine (controller)
- Support for CXL 3.0 for new use models that optimize memory resources (PHY)