EDA: 7 Grand Challenges

The concept of a Grand Challenge is an established one in engineering, so here in 2012 what are the Grand Challenges in EDA? Let’s go out on a limb and name a few candidates:

No.1) Low power: This is the critical problem here in the era of mobile everything. If you can’t guarantee low power for your device, it’s going to go dark way too soon and be way too hot in the meanwhile. Great challenges remain in perfecting the tools to make this all happen.

No.2) Formal verification: There just has to be a way to guarantee that what we meant to design, has been designed and then manufactured. Isn’t that the goal of formal verification, and isn’t it true that we’re not quite there yet?

No.3) 3D-ICs: In the last several years, this one’s gotten a lot of attention, but it appears that there’s still a lot of work to do – at least on the logic side of the equation. Clearly more tools are needed.

No.4) Reconfigurable hardware: For some people, this one is the holy grail. If we can move from hard-wired solutions to cost-efficient and flexable reprogrammable solutions, we solve a host of problems including the need to know now what will sell 3 years from now, after the ASIC we’re designing today finally comes off the line. And why is this an EDA challenge?  ASICs have been the bread & butter of EDA.

No.5) Expressing system-level specs: If we were better at describing what we want that black box we call the design flow to do for us at higher levels of abstraction, we might be happier with the overall results, especially if …

No.6) System-level tools: were a reality. Perhaps they already are a reality from some people’s POV, but not everybody seems to be on board with this one.

No.7) Hardware/Software co-design: The idea seems so completely intuitive, but as with many things in engineering – so much more easily said than done. Especially when there are still less-than-perfect communications between the hardware team and the software team. More perfect tools needed here as well.

Okay, that’s it: 7 Grand Challenges in EDA. The list may be incomplete and might be debated by many, but certainly one thing is clear: There’s a lot of work to be done and, as a result, a lot of opportunity within EDA.

**********************

Tags: 3D-ICs, co-design, EDA, formal verification, Grand Challenge, hardware/software co-design, low power, reconfigurable hardware, system-level tools

This entry was posted on Wednesday, July 11th, 2012 at 11:34 am. 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.