MoSys Announces Optimized P4 Pipeline Support for Stellar Packet Classification Platform IP for FPGAs
Solution Designed to Easily Integrate with P4 Pipelines in 5G Wireless and Cloud Data Center Networks
SAN JOSE, CA / ACCESSWIRE -- May 17, 2021 -- MoSys, Inc. (NASDAQ:MOSY), is focused on Accelerating Data Intelligence and provides both semiconductor and IP solutions that enable fast, intelligent data access for Cloud, networking, security and communications systems. Today, MoSys announced that it will be adding support for optimized P4 (Programming Protocol-Independent Packet Processors) pipeline acceleration to its Stellar Packet Classification Platform IP family. The solution, which is expected to be available to early access customers in Q3, 2021, can be easily integrated into P4 pipelines available from leading FPGA vendors.
"MoSys has long been a supporter of P4, and now we are adding P4 support to our Stellar Packet Classification Platform family of virtual accelerators, designed as IP that can be loaded into FPGAs," noted Michael Miller, MoSys' chief technology officer. "With this new milestone of enabling P4 support embedded in an FPGA, complex five tuple Access Control List (ACL) rule databases can be extended and accelerated using a combination of internal FPGA SRAM and HBM memories, thus enabling large databases of rules."
The P4 open source community (p4.org) is named after the four Ps as in a July 2014 whitepaper entitled, "Programming Protocol-Independent Packet Processors" and has been growing in significance ever since. With more than 80 industrial contributors and 12 academic contributors, it helps provide a protocol-independent methodology that specifies how devices like intelligent SmartSwitches or SmartNICs should process packets. Now, with the MoSys Stellar Packet Classification Platform solutions, designers will be able to increase capacity and performance by accelerating P4 processing directly in an FPGA.
The MoSys Stellar Packet Classification Platform is an innovative virtual accelerator that is based on the MoSys Graph Memory Engine (GME), which performs ultra-high-speed embedded search and classification of packet headers using ACL or LPM rules, thereby allowing TCAM-like functions to be replaced by FPGA-accelerated algorithms.
For more info on P4 - join the P4 workshop happening this week May 18-20, 2021 - see 2021 P4 Workshop
The MoSys Stellar Packet Classification IP is being designed to support a wide range of silicon, SmartNICs and SmartSwitches. MoSys also supports selected Intel and Xilinx FPGA families and products such as Intel® Stratix®10 and Xilinx® UltraScale+™.
The Stellar Packet Classification Platform IP targets:
- Routing - supports Longest Prefix Match (LPM) IPv4/IPv6 routing, including support for P4 and virtual routes for Cloud Data Centers, 5G User Plane Functions (UPF), P4 based systems, Network Classification, Carrier-Grade NAT, Broadband Network Gateways (BNG), NFVi, Flow Steering, L3 Forwarding and Filtering, vRouter, Open vSwitch Offload, and Cloud Gateways.
- Security, Load Balancing and Traffic Analysis - supports very complex, 10+ tuple Access Control List (ACL) type lookups for Network Firewalls, Allow/Deny Lists, P4 based systems, Network Detection and Response (NDR), Anomaly Detection, Lawful Intercept, Anti-DDoS, L4 Load Balancing, Application Delivery Controllers (ADC), Application and Network Analysis, Network Telemetry, Test and Measurement, Network Packet Brokers, and other markets and use cases.
Additional Resources:
- Stellar Packet Classification Platform Family
- Anti-DDoS (ACL) Use Case
- Network Firewall (ACL) Use Case
- Data Center Routing (LPM) Use Case
- 5G Routing (LPM) Use Case
About MoSys, Inc.
MoSys, Inc. (NASDAQ:MOSY) is focused on Accelerating Data Intelligence and provides both semiconductor and IP solutions that enable fast, intelligent data access and decision making for a wide range of markets including cloud networking, security, 5G networks, SmartNIC, test and measurement, and video systems. MoSys's Quazar family of high-speed memories and the Blazar family of Accelerator Engines are memory integrated circuits with unmatched intelligence, performance and capacity that eliminate data access bottlenecks to deliver speed and intelligence in systems, including those scaling from 100G to multi-terabits per second. MoSys's Stellar family of Virtual Accelerator Engines includes software, FPGA RTL and RISC-based firmware that accelerate applications and are portable across a wide range of hardware configurations with or without MoSys silicon chips. More information is available at: MoSys.
|
Related News
- MoSys Partners with Silicom to Provide Stellar Packet Classification IP Optimized for Intel FPGA-Based SmartNICs and Infrastructure Processing Units
- Achronix FPGAs Add Support for Bluespec's Linux-capable RISC-V Soft Processors to Enable Scalable Processing
- Efinix Releases TinyML Platform for Highly Accelerated AI Workloads on Its Efficient FPGAs
- Lattice Certus-NX FPGAs Optimized for Automotive Applications
- Lattice Propel Accelerates Time-to-Market for Embedded Processor-based Designs on Latest Nexus Platform FPGAs
Breaking News
- Ubitium Debuts First Universal RISC-V Processor to Enable AI at No Additional Cost, as It Raises $3.7M
- TSMC drives A16, 3D process technology
- Frontgrade Gaisler Unveils GR716B, a New Standard in Space-Grade Microcontrollers
- Blueshift Memory launches BlueFive processor, accelerating computation by up to 50 times and saving up to 65% energy
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
Most Popular
- Cadence Unveils Arm-Based System Chiplet
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Esperanto Technologies and NEC Cooperate on Initiative to Advance Next Generation RISC-V Chips and Software Solutions for HPC
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- Arteris Selected by GigaDevice for Development in Next-Generation Automotive SoC With Enhanced FuSa Standards
E-mail This Article | Printer-Friendly Page |