Xelera Silva provides best-in-class throughput and latency for Gradient Boosting Trees and Neural Network inference.
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.
Inference time with a typical latency of 1 to 3microseconds, depending on the AI algorithm
Neural networks and boosted tree algorithms under aunified high-performance software API in C/C++ and Python
Train your own model with the standard frameworks onyour data and dynamically deploy on the accelerator card
Concurrent execution of multiple models on a singleaccelerator with instantaneous model hot-swapping
High Frequency Traders use decision algorithms to automate trading instructions. The automated decisions are increasingly made by AI models. A low latency is key for these systems. Silva overcomes the latency disadvantage of Machine Learning algorithms: Inference of Gradient Boosting Trees and Neural Networks models is performed with a latency of a few microseconds. This enables our clients to make better, sophisticated trading decisions and win speed races. The turn-key accelerator connects to the software-based trading system and offloads the AI inference to a PCIe-attached hardware accelerator card.
The turn-key accelerator connects to the software-based trading system and offloads the Gradient Boosting Trees and Neural Networks inference to a PCIe-attached hardware accelerator card (ultra-low latency PCIe transfer included in round-trip latency).
*Median Inference Latency
Benchmark Model:
Test model: LightGBM Regression
Number of features: 128
Number of trees: 1000
Number of levels: 8
Batch size: 1
Test system:
Dell PowerEdge R750, Intel(R) Xeon(R) Gold 6346, Ubuntu 20.04.4 LTS
Accelerator:
AMD Alveo V80
*Median Inference Latency
Benchmark Model:
Type: LSTM
Number of inputs: 256
Number of hidden: 256
Number of layers: 3
Batch size: 1
Test system:
Dell PowerEdge R750, Intel(R) Xeon(R) Gold 6346, Ubuntu 20.04.4 LTS
Accelerator:
AMD Alveo V80
In addition to the turnkey version, Xelera Silva is also available as an IP core. The inline Machine Learning accelerator is inserted into the fast path of network-bound hardware accelerators and receives input from the card's network port. In this way, no data needs to be transferred via the PCIe bus and the corresponding latency for data transfer is eliminated. This product is relevant for customers with their own FPGA teams and offers the lowest latencies.
Developers of HFT hardware accelerators
integrate the IP Core into their FPGA design to benefit
from AI inference at lowest latency.
*Inference Latency
Benchmark Model:
Test model: LightGBM Regression
Number of features: 128
Number of trees: 1000
Number of levels: 8
Batch size: 1
FPGA:
AMD Alveo V80
* Inference Latency
Benchmark Model:
Type: LSTM
Number of inputs: 256
Number of hidden: 256
Number of layers: 3
Batch size: 1
FPGA:
AMD Alveo V80
Xelera Silva is a turnkey full-stack solution designed to jumpstart best-in-class AI Inference acceleration.
DEB / RPM packages and FPGA bitstreams for AMD AlveoU50, U55C, V80 accelerator cards and Napatech NT200A02.
API: C/C++, Python
Host library to load model to the FPGA and run inference
Jumpstart the AI inference acceleration with the provided example design
Integration and full lifecycle maintenance support
Periodic software updates
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.
Do you have any Questions?