![Tantra's Mantra with Prakash Sangam artwork](https://is4-ssl.mzstatic.com/image/thumb/Podcasts123/v4/6c/59/21/6c592137-6f0d-2365-178d-b04c33fa7ac9/mza_13834770549643395869.jpg/100x100bb.jpg)
Open RAN Layer1 - Generic Compute vs. Accelerator - Marvell's View
Tantra's Mantra with Prakash Sangam
English - September 12, 2022 03:04 - 26 minutes - 23.9 MBTechnology Business Management amazon ftc iot microsoft apple cellular cloud compute edge google Homepage Download Google Podcasts Overcast Castro Pocket Casts RSS feed
Open RAN Layer-1 workload that performs and manages most of the radio functions is one of to critical blocks. It is highly complicated and requires real-time processing. The complexity only increases with advanced features such as massive MIMO, carrier aggregation, URLLC, etc. The debate in the industry is whether general compute or optimized accelerators are the most cost and power-efficient solutions. In this episode, I speak with Joel Brand of Marvell Semiconductor on his views on this debate.
Relates articles to the subject:
http://bit.ly/TA-Insights (click on vRAN/Open RAN tab)