Industry's First Verification IP for PCIe 7.0
In our recent blog, Synopsys Accelerates Trillion Parameter HPC & AI Supercomputing Chip Designs with Industry's First PCIe 7.0 IP Solution, Synopsys announced the first comprehensive PCIe Express® Gen 7 (PCIe 7.0) IP Verification IP (VIP) solution to support the speed and low latency required for Artificial Intelligence (AI) applications in high performance computing designs.
PCI Express Evolution
Since its launch in 2003, PCI Express has undergone continuous advancements in technology, specification, and transfer speed, demonstrating its success as a standardization initiative and data transfer protocol. PCIe 5.0 powered cloud computing resources with 32G transfer speeds and CXL coherency, while PCIe 6.0 doubled performance to 64G transfer rates using Flow Control Units (FLITS) and PAM4 modulation for effective, low latency communication and coherency. PCIe 7.0's load-store capabilities and up to 512 GB/s of bandwidth for secure data transfers make it possible to connect multiple accelerators and process large, complex AI and machine learning models efficiently.
To read the full article, click here
Related Semiconductor IP
- Band-Gap Voltage Reference with dual 2µA Current Source - X-FAB XT018
- 250nA-88μA Current Reference - X-FAB XT018-0.18μm BCD-on-SOI CMOS
- UCIe D2D Adapter & PHY Integrated IP
- Low Dropout (LDO) Regulator
- 16-Bit xSPI PSRAM PHY
Related Blogs
- Cadence Showcases World's First 128GT/s PCIe 7.0 IP Over Optics
- Progressing on Track: PCIe 7.0 Specification, Version 0.7 Now Available for Member Review
- Industry's First Verification IP for Display Port Automotive Extensions (DP AE)
- Skymizer Reduces Verification Cycles for AI Accelerator IP Development by 33% with Synopsys HAPS Prototyping
Latest Blogs
- AI in Design Verification: Where It Works and Where It Doesn’t
- PCIe 7.0 fundamentals: Baseline ordering rules
- Ensuring reliability in Advanced IC design
- A Closer Look at proteanTecs Health and Performance Management Solutions Portfolio
- Enabling Memory Choice for Modern AI Systems: Tenstorrent and Rambus Deliver Flexible, Power-Efficient Solutions