Liquid-Markets-Solutions (LMS) and BittWare are addressing a weak point in the FSI market: the humble NIC card. Once a simple purchase for getting adequate low-latency performance, today’s NIC cards aren’t keeping up. At 825 Mb/s, latency on today’s NIC cards begin to degrade to several microseconds—quietly eating away at the edge your algorithms had in the market.
Enter ÜberNIC Nano: the latest addition to the ÜberNIC family of SmartNICs. It starts with the impressive latency of AMD UltraScale+ FPGAs and stays there—right through to multiple Gb/s. It does this with similar ease and cost to the NICs you’re using now.
Powerful Performance in a Single-Slot NIC
The Silicon Advantage
The internal latency of your NIC is a critical performance component. That’s why the ÜberNIC Nano is built on AMD’s UltraScale+ family of FPGAs—proven to provide the lowest possible latency.
But that’s not to say all NICs will perform equally. The real difference is how the network stack is designed, specifically how much is handled by the CPU. That’s where the ÜberNIC family gives the ultimate advantage: the entire network stack is implemented in hardware, ensuring the lowest possible latency.
Even today’s most popular NICs favored in FSI use a combination of hardware/software that results in poor latency within the scope of today’s market feeds. See our example using a 1-day NYSE stream below.
Fact: today’s market feeds regularly pass 850 Mb/s—the point at which even the best competitive NIC’s network stack begin to dramatically lose sufficient latency performance.
Enter ÜberNIC, the first commercially-available FPGA NIC with the entire network stack in accelerated hardware, allowing for consistent low latency—ready for today’s markets and the future.
Though the focus can be on low latency, another big advantage with LMS ÜberNIC is precision time. Specifically, this family of cards offer time synchronization with a variety of external devices and standards, including White Rabbit.
Powerful Performance in a Single-Slot NIC. Shown as Nano+ with optional Timing Kit.
Raw performance improvement that’s consistent through the critical peak market feed moments when you need it most!
More fiber pairs per card means getting more out of each server. Powerful enough for ancillary services like PCAP and timestamping on the same card.
Built-in support for time synchronization between cards and the CPU.
By moving the complete network stack into FPGA hardware, the ÜberNIC Nano achieves unmatched performance.
Single-slot card supports up to 8x 10/25G fiber pairs
Tap into the world’s top choice for FSI ultra-low latency performance, the AMD Virtex UltraScale+ family of FPGAs. Featuring incredible transceiver performance, LMS and BittWare built ÜberNIC on an excellent foundation. There’s no lack in I/O features either, with up to 32GB DDR4, 16 lanes of 10/25G with hard MAC, and PCIe Gen3 x16.
At its core, ÜberNIC Nano is a low-latency focused NIC. With the combination of full hardware acceleration and low-latency PCIe transfers, you get unmatched performance—even during peak market times.
* Available on ÜberNIC Nano+ only.
BittWare brings enterprise-class products to market using the latest technology such as AMD UltraScale+ FPGAs. Today’s high-speed interfaces require significant care in the design and manufacturing process to make a reliable, high-quality product at scale.
Pictured is our manufacturing facility at ISI, a Molex company, in California.
Only FPGA-based solutions with bundled hardware network allows for all these extensible functions.
Ready to learn more about ÜberNIC? Reach out to BittWare to learn more!
BittWare On-Demand Webinar Enterprise-class FPGA Servers: The TeraBox Approach FPGA-based cards are maturing into critical devices for data centers and edge computing. However, there’s a
Article Homomorphic Encryption Acceleration FPGA acceleration enables this unique solution that allows compute on encrypted data without decrypting or sharing keys Traditional Encryption Limits Encrypting
We examine our reference design for sustained 100 Gb/s capture to host DDR4 over a PCIe bus. Read the white paper, then request the App Note for even more detail!
BittWare Webinar Arkville PCIe Gen4 Data Mover Using Intel® Agilex™ FPGAs Webinar The Arkville IP from Atomic Rules was recently updated to support Intel Agilex