FPGA Full System Stacks (FFSS)
“When Wall Street wanted to program computers without having computer programmers, they invented the spreadsheet!” – This was our spirit when we came up with the FPGA Full System Stack: Make building FPGA-based systems easier for engineers without expert knowledge in FPGA design.
Why An FPGA Full System Stack?
For compute-intensive, time-critical applications in Automotive, Aerospace/Defence, Industrial/Scientific/Medical, FPGAs are often the choice. However, programming FPGAs, particularly those System-on-Chip (SoC) FPGA with embedded CPUs, has long been considered complex and risky.
Integrated and pre-validated building blocks of FPGA hardware and software subsystems, which we call “FPGA Full System Stack” can greatly de-risk development and accelerate new product initiatives.
What makes up the FPGA Full System Stack?
Every FPGA Full System Stack comprises the following elements (top-down):
- Application Example Subsystems which are customizable by users based on your applications
- A Targeted Performance Subsystem, including one or more MLE IP-Cores
- A Board Support Package, with a Linux Base System and Board Part Files
- The FPGA device (such as the AMD VE2302, for example)
- A SoM (such as the TE0955, for example)
- A carrierboard (such as TEBF0955, for example)
Pre-integrating and testing this complex stack gives you, the implementer the advantages:
- You 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 IPs and Subsystems make clever use of different FPGA resources to realize a cost/performance optimized domain-specific architecture. No need for reading long datasheets or making timing closure.
- Operating system and application 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.
MLE FPGA Full Systems Stacks
MLE and Trenz Electronics have been planning for many different FFSS to come. Our initial focus is on “embedded” platforms for performance networking and for performance storage / data acquisition based on AMD Versal AI Edge. Below are some examples.
If you are interested in other boards, other FPGA devices or other functionality, please contact us!
Network Acceleration
![]() | TE0950 with Full Accelerated 25G TCP/UDP/IP The FFSS-TE0950-NPAP-25G is a customizable FPGA design platform with accelerated 25 GigE network connectivity. This FFSS combines MLE NPAP (the TCP/UDP/IP Network Protocol Acceleration Platform) with Trenz Electronic’s TE0950 Evaluation Board featuring the AMD Versal™ AI Edge VE2302. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. |
![]() | TE0950 with Full Accelerated 10G TCP/UDP/IP The FFSS-TE0950-NPAP-10G is a customizable FPGA design platform with accelerated 10 GigE network connectivity. This FFSS combines MLE NPAP (the TCP/UDP/IP Network Protocol Acceleration Platform) with Trenz Electronic’s TE0950 Evaluation Board featuring the AMD Versal™ AI Edge VE2302. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. |
![]() | TE0950 with 25G Linux Network Stack (Non-Accelerated) The FFSS-TE0950-Netdev-25G is a customizable FPGA design platform for Linux-based (non-accelerated) 25 GigE network connectivity. This FFSS combines an MLE optimized Linux network stack with Trenz Electronic’s TE0950 Evaluation Board featuring the AMD Versal™ AI Edge VE2302. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. |
![]() | TE0950 with 10G Linux Network Stack (Non-Accelerated) The FFSS-TE0950-Netdev-10G is a customizable FPGA design platform for Linux-based (non-accelerated) 10 GigE network connectivity. This FFSS combines an MLE optimized Linux network stack with Trenz Electronic’s TE0950 Evaluation Board featuring the AMD Versal™ AI Edge VE2302. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. |
![]() | TE0950 with 25G Configurable Multi-Queue FPGA NIC The FFSS-TE0950-MQNIC-25G is a customizable FPGA design platform with configurable Multi-Queue 25 GigE network connectivity. This FFSS combines an MLE optimized Linux network stack with Trenz Electronic’s TE0950 Evaluation Board featuring the AMD Versal™ AI Edge VE2302. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. |
![]() | TE0950 with 10G Configurable Multi-Queue FPGA NIC The FFSS-TE0950-MQNIC-10G is a customizable FPGA design platform with configurable Multi-Queue 10 GigE network connectivity. This FFSS combines an MLE optimized Linux network stack with Trenz Electronic’s TE0950 Evaluation Board featuring the AMD Versal™ AI Edge VE2302. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. |
Storage Acceleration
![]() | TE0950 with NVMePL x4 Bandwidth Accelerated NVMe data streaming with 4 lanes in PL, based on TE0950. |
![]() | TE0950 with NVMePL x1 Bandwidth Accelerated NVMe data streaming with 1 lane in PL, based on TE0950. |
![]() | TE0950 with NVMePS Data read/write onto a Linux-connected NVMe SSD using Linux file system, based on TE0950. |
FPGA Full System Stack Applications
FPGA Full System Stacks from MLE focus on applications such as:
- Reliable, Low-Latency, High-Throughput Network Transports
- High-Speed Data Acquisition
- Augmented Stereo Computer Vision
- High-Speed Data Record & Replay
How does FPGA Full System Stack help your FPGA design project?
Trenz and MLE have established close engineering collaboration and a track record of shipping integrated FPGA solutions based on Trenz hardware running MLE’s System Software Stacks with Compute, Video, Storage and/or Network Acceleration. This facilitates concurrent engineering of hardware and software while reducing engineering and commercial risks. The design methodology is:
Build PoC with StarterKits
StarterKits for system architecture exploration and validation via Proof-of-Concept implementations
Validation & Shipment
Ship products using StarterKits with System Software Stacks integrated
Chip-Down Board
Cost-down for volume ramp-up using customized, chip-down hardware
Documentation
- Network Acceleration FPGA Full System Stacks:
- FFSS-TE0950-NPAP-25G Datasheet
- FFSS-TE0950-NPAP-10G Datasheet
- FFSS-TE0950-Netdev-25G Datasheet
- FFSS-TE0950-Netdev-10G Datasheet
- FFSS-TE0950-MQNIC-25G Datasheet (available upon request)
- FFSS-TE0950-MQNIC-10G Datasheet (available upon request)
- Storage Acceleration FPGA Full System Stacks:
- FFSS-TE0950-NVMePL-x4 Datasheet (available upon request)
- FFSS-TE0950-NVMePL-x1 Datasheet (available upon request)
- FFSS-TE0950-NVMePS Datasheet (available upon request)
- MLE presents “From Software to Silicon: Accelerating Automotive In-Vehicle Network Protocols for Zonal Architectures” at the “Driving the Future Symposium”
- MLE Presents “In-Vehicle Network – Automotive Zone-based Architecture with Time Sensitive Network – Auto/TSN” at FPGA Conference 2025
- MLE Presents Automotive Rapid Prototyping System for SDV / Zonal Architectures
Trenz Electronics

Since 1992 Trenz Electronic has operated as a provider of development and production services for the electronics industry. Services include design-in support as well as turnkey designs which typically covers all steps from product specification, hard- and software design up to prototyping and production. Trenz specializes in the design of high-speed data acquisition, high-accuracy measurement and embedded digital signal processing systems based on FPGA and CPU architectures. Trenz development service is supplemented by FPGA- and MCU-based development boards and tools.









