A TCP/IP Stack for High-Performance Chip-to-Chip

April 3, 2025Application Notes, NetworkingAurora Protocol, AXI4-Stream, High-speed Data, Long-Distance Distributed System, Network Acceleration, NPAP, TCP/IP

A TCP/IP Stack for High-Performance Chip-to-Chip

TB20250403

Stream processing enables parallel data handling by creating pipelines of Stream Processing Elements (SPE) between a data source (typically a sensor) and a data sink (typically a decision maker), synchronized through back pressure. For multiple FPGA systems, the so-called AURORA protocol is frequently used in industrial, medical and scientific applications. However, TCP/IP over multi-Gigabit Ethernet can offer a more reliable and resource-efficient solution for distributed systems, in particular when FPGA devices are farther apart.

This Technical Brief shares our experiment data, comparing the reliability and resource efficiency of Aurora protocol and TCP/IP, an MLE NPAP Full Accelerator running TCP/IP inside the programmable logic.

Table of Contents

A TCP/IP Stack for High-Performance Chip-to-Chip
- Stream Processing Pipelines with TCP
- References

Analyzing Network Impairment and Signal Integrity in High-Speed TCP/UDP/IP Ethernet Networks

March 27, 2025Application Notes, NetworkingHigh-speed Data, Integrated Bit Error Rate Testing (iBERT), Latency Optimization, Netperf, Network Acceleration, Network Impairment Emulation, NPAP, RTL Simulation, TCP/IP, TCP/UDP/IP

Analyzing Network Impairment and Signal Integrity
in High-Speed TCP/UDP/IP Ethernet Networks

TB20250327

Summary

MLE has been supporting customers in implementing distributed systems. MLE NPAP, the TCP/UDP/IP Stack originating from Fraunhofer HHI, has been one of the underlying technologies to connect FPGAs with each other as well as with servers and sensors reliably.

Customer implementations range from systems using a purpose-built LAN, or systems exposed to high levels of electro-magnetic inference (EMI), or rotating nodes using slip rings, or wireless.

Engineering teams often focus on delivering high throughput (close to linerate) and very low latency (i.e. a small Bandwidth-Delay-Product). Network impairment can quickly derail those engineering efforts. Therefore, MLE NPAP has built-in means to diagnose the effects of real-life network impairment.

This Technical Brief expands on analyzing network impairments and explains how these can be emulated when using Field-Programmable Gate Arrays (FPGAs), leading to technical insights allowing you to deliver better systems, faster.

Table of Contents

Analyzing Network Impairment and Signal Integrityin High-Speed TCP/UDP/IP Ethernet Networks

High-Speed Camera Recorder Powered by MLE’s Fast FPGA RAID (FFRAID)

March 25, 2025Application Notes, Storage400Gbps, Data Recording, Fast FPGA RAID, FFRAID, High-speed Data

High-Speed Camera Recorder Powered by MLE’s Fast FPGA RAID (FFRAID)

Success Story

Excelitas PCO is a pioneer in cameras and optoelectronics, offering distinct platforms for high-speed cameras for a wide range of industrial, scientific, car safety, and aerospace applications.

Challenge

The new pco.dimax series offers a unique platform of high-speed streaming cameras that enable clear images with a recording speed of over 2000 fps at a full resolution of 3.6 MPixel. This massive raw data rate comes with the challenges to record this image data gapless and lossless!

MLE helped Excelitas PCO develop the high-speed data acquisition and recording systems by implementing FFRAID — a Fast FPGA RAID — to offload CPU-intensive tasks and achieve up to 200 Gbps sustained recording speeds.

Table of Contents

High-Speed Camera Recorder Powered by MLE’s Fast FPGA RAID (FFRAID)

FPGA-Based Systems With MicroTCA

February 5, 2025Application Notes, Methodologies400Gbps, Agilex 5, AMD, AMD Versal, Data Recording, Fast FPGA RAID, FFRAID, High-speed Data, MicroTCA, mTCA, PCIe 4.0, PCIe 5.0, SoM, Trenz

FPGA-Based Systems With MicroTCA

TB20250116

Many FPGA-based systems projects are rather low volume and, therefore, favor off-the-shelf hardware over the costs and risks of custom hardware (PCB) design.
MicroTCA or mTCA or µTCA is a 20+ year old standard from PICMG for implementing embedded systems using off-the-shelf hardware. Most often MicroTCA is used for test & measurement or for scientific applications when application specific IOs and signals need to be processed where FPGAs provide the flexibility and processing performance.
However, as we will demonstrate in this Technical Brief, MicroTCA carries a huge baggage, and has little to no advantages over other approaches, namely FPGA based System-on-Modules or the “good old” PC architecture.

Table of Contents

FPGA-Based Systems With MicroTCA

FFRAID – A Fast FPGA RAID

January 30, 2025Application Notes, Storage400Gbps, AMD, AMD Alveo U55C, AMD U50, AMD V80, Data Decimation, Data Recording, High-speed Data, Indexing Metadata, Linux Soft-RAID, NVMe SSD RAID0, NVMe Streaming, PCIe 4.0, PCIe 5.0, RAID1, Security, TCG OPAL

FFRAID – A Fast FPGA RAID for High-Speed Data Acquisition

TB20250130

In past MLE Technical Briefs (“Sustained, High-Speed Data Recording with NVMe SSDs” and “NVMe Streamer for High-Speed FPGA Data Acquisition & Recording” and customer success stories we have presented FPGA-based solutions for high-speed data acquisition. This Technical Brief is a sequel describing how FPGA-based data acquisition can be pushed to new limits using MLE FFRAID, a Fast FPGA RAID for NVMe SSDs.

Table of Contents

FFRAID – A Fast FPGA RAID for High-Speed Data Acquisition

Shift-Left Your FPGA Design Projects

January 22, 2024Application Notes, Methodologies4K Image Processing, High-speed Data, Network Acceleration, NPAP, SmartNIC, Time-Sensitive Networking (TSN), Trenz

Shift-Left Your FPGA Design Projects

TB20231221

Summary

FPGA Full System Stacks comprising off-the-shelf FPGA System-on-Modules (SoM) plus pre-validated FPGA IP Cores and subsystems can greatly accelerate the time-to-market of your FPGA design project. Advantages of FPGA Full System Stacks include:

FPGA developers can rely on a tested and verified subsystem implementation. The concept of re-use increases design productivity while sharing the FPGA subsystem development costs and risks over many users.
Pre-validated FPGA IP-Cores and subsystems make clever use of the different FPGA resources to realize a cost/performance optimized domain-specific architecture.
Software is included in the form of kernel space device drivers, user-space programmer APIs, and sometimes even complete OS images, all nicely tuned for guaranteeing the overall system’s reliability and performance.

FPGA Full System Stacks from MLE are integrated with select FPGA SoMs from Trenz Electronics and are focused on applications such as:

Realiable, Low-Latency, High-Throughput Network Transports
High-Speed Data Acquisition
Augmented Stereo Computer Vision
High-Speed Data Record & Replay

We describe a design methodology using FPGA Full System Stacks and share our experiences from real customer designs.