Real-Time Server Monitoring and CNN Inference on FPGA

Neutrinos are subatomic particles, very similar to an electron, but without any electrical charge and 
a very negligible rest mass. They are the most abundant and perhaps the most mysterious matter 
particles in the universe!  

Detecting and tracking neutrinos poses a different challenge both in terms of hardware and 
algorithms. Since they have very little interaction with matter, they are incredibly difficult to detect. 
So, every bit of information produced by the neutrino interactions is required to be stored and 
analyzed at an incredible fast rate.  

Present day State-of-the-Art Convolutional Neural Networks (CNNs) have shown an un-precedented 
accuracy on many modern AI applications. They have become the de-facto standard for a wide range 
of tasks ranging from computer vision to machine translation due to their high accuracy and 
robustness. 

Research has proved that CNNs also hold the potential to excel in High Energy Physics (HEP) 
applications as compared to traditional methods in identifying particle interactions in sampling 
calorimeters or Time Projection Chambers. However, these algorithms are both computationally 
and memory extensive which limits them from running on normal CPU for real-time and power 
constrained applications. Thus, customized hardware implementation of machine learning 
algorithms can be a promising solution for higher performance and improved throughput.  

Another important aspect is to constantly monitor the server which hosts the FPGAs for the CNN 
inference and sent alert messages in crucial situations. 

This project involved the development of a lightweight, easy to use solution for the server 
monitoring so that all the important metrices can be visualized at a glance in real-time. Secondly, to 
compare the CNN inference between the server and the FPGA in terms of accuracy, power 
consumption and latency. The project also focuses on dealing with the numerical overflow problem 
due to fixed point computations in FPGA during CNN inference.