AI/ML cluster fabric fidelity scores on collective communication emulations

DESPRE COMPANIE

Keysight Technologies is a leading technology company that helps enterprises, service providers and governments accelerate innovation to connect and secure the world. Keysight's solutions optimize networks and bring electronic products to market faster and at a lower cost with offerings from design simulation, to prototype validation, to manufacturing test, to optimization in networks and cloud environments. Customers span the worldwide communications ecosystem, aerospace and defense, automotive, energy, semiconductor, and general electronics end markets.

Site-ul companiei

DE LA ACEEAȘI COMPANIE

Toate stagiile de la Keysight Technologies Romania

AI/ML cluster fabric fidelity scores on collective communication emulations

Stagiu plătit la Keysight Technologies Romania · 26/06/2024

Categorii:

– Artificial Intelligence

Oraș:

București

Aptitudini necesare:

machine learning artificial intelligence python c++ git

Our goal is to benchmark AI/ML cluster fabric with realistic workloads which typically require investments into compute systems with GPUs and RDMA NICs that are costly and time-consuming to build and operate.

To achive this goal, we're going to create an automated system that is able find the highest collective communication emulation fidelity we can achieve using our software and hardware traffic generators. This system is meant to run continuously and provide insights into our collective communication product performance, highest fidelity it can achieve and the exact parametrization used to achieve it.

What you will gain: • Learn about how Machine Learning libraries work: topologies, algorithms and networking. • Take a deep dive into implementing algorithms using either python or c++ • Learn about ROCEv2 protocol and NIC specific features and how to create an automated system that is capable of searching for a given objective