The Asynchronous Logic - link collection

[jimjim2k]

Hi

The complexity of today's VLSI systems requires new design methods based on asynchronous techniques, concurrency, high-level synthesis, and verification. The research of the Asynchronous VLSI group at Caltech focuses on methods and tools for the design of high-performance and low-energy asynchronous digital circuits. Digital circuits are called "asynchronous" when they don't use clocks.

Check The following URls for starting point on "The Asynchronous Logic"

1. h**p://www.cs.man.ac.uk/async/
2. h**p://www.cs.man.ac.uk/async/tools/index.html (Tools)
3. h**p://www.cs.man.ac.uk/async/background/index.html
4. h**p://www.async.caltech.edu/

* -> t

tnx

 

[heat]

Hey,

thankx, that is really interesting stuff

 

[roli]

Designing Asynchrounous Digital circuits is the forefront challenge in Digital VLSI Design - and is the future for low-power, very high-speed organic based technologies.
Nice links !

 

[matteo128]

Thanks for the pointers.

Asynchronous logic is an interesting but definitely not new topic.

There was lot of research done on this topic in sixties, even a textbook or two were written at the begining of the seventies and then the subject went out of fashion only to reappear again 5 or 6 years again.

There are several fundamental unsolved problems there that limit the applicability to very specific problems. This reminds me of neural logic or fuzzy systems. They originaly appeared forties and sixties respectively. After a period of research they dissapered only to come back as a fashionable topic 5 0r 10 years ago. And again they didn't become a panacea as it was expected but only solution to very specific problems.

Matteo

 

[ulderico]

Thanks for the pointers

 

[Froed]

Does anyone have links to High-Speed-Digital Design ( With Verilog or VHDL ) ???

 

[rfsystem]

I have studied at 95 the realisation of a 16bit ALU with asynchron logic. Found for example the medium carry time for the ADD operation is only log2. The result is that it is not complicated to built up asynchron functional blocks but to link them together in a system architecture is not supported by the today tools. So on the circuit level there are solutions but the system architects need rethinking and also the test methologies.

 

[Guest]

Nice information !!
Thank you !!

 

[mc_eddy]

really nice...

 

[poorking]

the idea is really grate. But there is still a long way to go.

 

[LatDrIvE]

Re: The Asynchronous Logic

hello all,
i have seen the post and because i am involving with the asynchronous logic i would like to say that it is very promising but since now still be far away from the production of a full product. the only company that produce a totally asynchronous micro is Philips ( spin off " handshake solutions"), by using the tangram language.

Intel, Atmel, Caltech have shown their interest about asynchronous desing but eventhow the industry's downfall does not left any big thoughts for seriously asynchronous research yes. The industry is looking for GALS (globaly asynchronous localy synhronous) mostly than pure asynchronous design.

The major advantage about asynchronous is the low EMI and today's researh is mostly in smartcards security. The Dual-rail encoding ( one bit represented as "bit bit' "ie. logic 1 is "01") is providing efficient security but the switching frequency is high so that the power. These was some thought anf facts about asynchronous logic design.

thank you...