EDACafe Industry Predictions for 2019 – Weebit Nano Ltd.

In 2019 the move towards intelligent memory systems will gain momentum together with the rapid growth of applications requiring memories which are faster and able to perform more advanced functions.

New applications, led by Artificial Intelligence, are leading the quest for new types of memories, post Von-Neumann. They require huge memories which are fast and need to also use a lot less power. But more than that – these memories need to be intelligent and able to do a lot more than just store, read and write data. They need to perform smart searches and pattern-matching internally. They need to be distributed among many compute elements and merged into them. These new architectures require new types of memories, including the fast-growing Emerging Non-Volatile Memories.

Neuromorphic Computing, Processing-In-Memory, In-Memory Compute, and other similar domains will get a major boost as they become major research targets, with a growing number of companies adopting them. Memories such as Resistive RAM (ReRAM), which are a natural fit for these domains thanks to its physical structure, will become more popular and start gaining market share.

On the other hand, the struggle with Moore’s law will become even more challenging as the time required to move from one processing node to the next continues to expand. One solution to this will be to find other ways to shrink the size of the designs, and specifically the memories in them which are now taking up most of the area. 2019 will be a year of struggle to find such solutions.

Coby Hanoch, Weebit Nano Ltd, CEO

Tags: Artificial Intelligence, Industry Predictions, Moore's Law, Neuromorphic Computing, Predictions, ReRAM

This entry was posted on Monday, January 7th, 2019 at 12:50 pm. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.