XELERA SILVA

Ultra-Low Latency
AI Inference Platform

Xelera Silva provides best-in-class throughput and latency for Gradient Boosting Trees and Neural Network inference.

decision treee acceleration software picture

Ultra-low latency with Machine Learning Acceleration

Artificial Intelligence (AI) models—powered by Gradient Boosting Trees and Neural Networks—are increasingly critical for real-time data analysis in domains such as algorithmic trading, recommender systems, biosciences, and cybersecurity applications including ransomware and DDoS detection.
However, deploying these models in latency or throughput sensitive environments poses significant performance challenges.

The Xelera Silva platform addresses these demands with ultra-low latency AI inference delivered through a unified API. Leveraging high-performance hardware accelerator cards, Silva eliminates common latency and throughput bottlenecks, providing a turnkey solution for seamless integration of AI-driven decision-making into production systems.

You must accept cookies, to view this video. Open cookie preferences.

Technical features

Gradient Boosting Trees
1 µs – 1.5 µs typical latency

Algorithms:
XGBoost, LightGBM,CatBoost
 
Feature Types:
Numerical (float32), Categorical 

Batch Size: 1

Devices:

AMD Alveo U50, U55C, V80
Napatech NT200A02 

Deployment:
On-premise 

Operating System:
Linux Ubuntu, Linux Rocky, CentOS

Neural Networks
1.8 µs – 5 µs typical latency

Algorithms:
LSTM, Linear Layers, ctivations(sigmoid, tanh, relu)
 
Feature Types:
float16, bfloat16 

Batch Size:

Devices:
AMD Alveo V80 


Deployment:
On-premise 

Operating System:
Linux Ubuntu, Linux Rocky, CentOS

Your benefits

Unmatched Speed

Inference time with a typical latency of 1 to 3microseconds, depending on the AI algorithm

Seamless Integration

Neural networks and boosted tree algorithms under aunified high-performance software API in C/C++ and Python

Bring Your Own Model

Train your own model with the standard frameworks onyour data and dynamically deploy on the accelerator card

Model Hot-Swap

Concurrent execution of multiple models on a singleaccelerator with instantaneous model hot-swapping

Use Cases

High-Frequency Trading
Software API Integration

The picture shows the Use Case for High-Frequency Trading: Software Tick-to-Trade
The picture shows the Use Case for High-Frequency Trading: Hardware Tick-to-Trade

High-Frequency Trading
IP Core

Get OUR DatasheetS now!

Xelera Silva Datasheets

Software api integrationIP COreProduct brief

Deliverables

Xelera Silva is a turnkey full-stack solution designed to jumpstart best-in-class AI Inference acceleration.

Software packages

DEB / RPM packages and FPGA bitstreams for AMD AlveoU50, U55C, V80 accelerator cards and Napatech NT200A02.

API Support

API: C/C++, Python

Host library to load model to the FPGA and run inference

Example design

Jumpstart the AI inference acceleration with the provided example design

Support and User Guide

Integration and full lifecycle maintenance support
Periodic software updates

Getting Started

Pricing and Support

We understand that integrating solutions does not only require exceptional functionality but also transparent pricing models and reliable support. As technology evolves, so do we. We are committed to continuous innovation, ensuring that our software remains at the forefront of machine learning acceleration. With regular updates and feature enhancements, you can trust that you're always leveraging the latest advancements in the field. Our commitment to innovation means that you can stay ahead of the competition and unlock new possibilities for your projects. Contact us today to learn more about our pricing plans and support services. Unlock the full potential of your projects.

Latest Product news

OUR Technology partner
AMD Logo
STAC Member Logo
AMD Logo
STAC Member Logo