PCIe DMA for high-frequency trading and packet processing

Nadi DMA

25.6 ns, doorbell to completion. On UltraScale+ silicon.

Every nanosecond spent in the DMA is budget taken from your algorithm. Nadi's median on UltraScale+ silicon is 25.6 ns.

25.6 ns doorbell → completion · 10 cycles @ 390.625 MHz · Alveo U280, 64-byte frames

Download the product brief

Evaluate

Try Nadi DMA on your silicon

NDA available.


Comparison with AMD/Xilinx XDMA and QDMA

Nadi DMA
25.6 ns · 1× 25.6 ns · 10 cycles at 390.625 MHz
Median of ten million packets, 64-byte frames on an Alveo U280
Xilinx XDMA
264 ns · 10.3× slower 264 ns · 10.3× slower than Nadi
Payload and descriptor on card, one million samples, every sample the same
~1500 ns default · 59× slower ~1500 ns · 59× slower than Nadi
Same XDMA core, same card, no tuning
Xilinx QDMA
488 ns · 19.1× slower 488 ns · 19.1× slower than Nadi
Payload and descriptor on card, one million samples, every sample the same
~1100 ns default · 43× slower ~1100 ns · 43× slower than Nadi
Same QDMA core, same card, no tuning

doorbell to completion (ns)

Same device-side interval on this Alveo U280: doorbell accepted to completion queued, 64-byte payload and descriptor already on the card. Nadi is the median of ten million packets. XDMA and QDMA are one million samples, every sample the same. Dashed: same products, same card, no tuning.

Ultra Low-Latency DMA Engine

High-frequency trading

Nadi DMA completes a transaction in 25.6 ns, the median of ten million packets on AMD/Xilinx UltraScale+ silicon. By using Nadi DMA, you know you're getting the best performance out there.

Low-latency packet processing

On 64-byte frames, DMA latency is a large share of the per-packet budget. Lawful Interception and Low-Latency Packet processing require increasingly faster response times. With Nadi DMA, you know you're getting it.


What is included?

The IP Core is delivered as an encrypted IP.

A Linux VFIO userspace driver comes with it. For Networking applications, we include a 10/25G and 100G reference NIC with it's own Linux network device driver and Data-Plane Development Kit (DPDK) Driver.

Request an evaluation license to start integrating Nadi DMA into your design.


Frequently asked questions

What is Nadi DMA?

A PCIe DMA core for high-frequency trading and low-latency packet processing on AMD UltraScale+ FPGAs (PCIE4 and PCIE4C). Doorbell to completion is 25.6 ns, the median of ten million packets on an Alveo U280. Siliscale licenses it as encrypted IP from the Vivado catalog.

Who is it for?

Teams building high-frequency trading systems, and teams building low-latency packet pipelines, on an UltraScale+ card in a host.

What does 25.6 ns buy?

Tuned XDMA is 10.3× slower. The XDMA default is about 59× slower. Tuned QDMA is 19.1× slower. The QDMA default is about 43× slower. Same Alveo U280.

What software ships with it?

A Linux VFIO userspace driver. A 10/25G and 100G NIC reference, with a Linux network driver and a DPDK driver.

How is it licensed?

Evaluation is free and capped at about a million transactions. Production is a paid license for one project. Use the form on this page for either.


Get in touch

Related: Xilinx QDMA IP core tutorial · All IP cores