News & Analysis
Comment
DrQuine
The ability of defective chips to heal on the production line and throughout the ...
Research project looks to repairing working chips
Nicolas Mokhoff
3/15/2011 10:36 AM EDT
MANHASSET, NY -- Crisp consortium researchers are demonstrating a self-testing and self-repairing chip at the DATE2011 conference in Grenoble this week.
The consortium consists of a team of four companies and two universities in The Netherlands, Germany, and Finland which is developing a gracefully degrading chip while the IC is still operationally functional.
A run-time resource manager self-repairs the chip and reconfigures the chip's tasks to fault-free parts in real time.
The research team was able to ascertain how chips can test and repair themselves. The methodology combines a test for faulty components and connections on chips with a run-time resource manager which assigns tasks and communication channels to known-good components and pathways.
This allows many-core chips with some faulty cores to pass production test, since they will function for the full 100 percent without any compromise to reliability, according to the Crisp consortium.
"Chips can function with faulty components by developing architectures of multiple cores that can degrade while they keep functioning,' according to Hans Kerkhoff, Associate Professor, CTIT, University of Twente.
A run-time resource manager dynamically determines which core does which task and swaps tasks from failing cores to let the chip repair itself, said Kerkhoff.
The resource manager continuously determines the chip's optimum Quality of Service on fault-free components, according to Bart Vermeulen, Senior Principal Scientist at NXP, one of the consortium’s company partners.
The resource manager dynamically assigns new tasks to free resources during the entire chip lifetime.
Crisp which stands for Cutting edge Reconfigurable ICs for Stream Processing is a European Union project that researches optimal utilization, efficient programming and dependability of reconfigurable many-cores for streaming applications.
The project consortium consists of Recore Systems (project leader), University of Twente, Atmel Automotive, Thales Netherlands, Tampere University of Technology, and NXP Semiconductors.
Details of the program can be found here.
DATE 2011, the Design Automation & Test in Europe conference, runs this week in Grenoble, France.
The consortium consists of a team of four companies and two universities in The Netherlands, Germany, and Finland which is developing a gracefully degrading chip while the IC is still operationally functional.
A run-time resource manager self-repairs the chip and reconfigures the chip's tasks to fault-free parts in real time.
The research team was able to ascertain how chips can test and repair themselves. The methodology combines a test for faulty components and connections on chips with a run-time resource manager which assigns tasks and communication channels to known-good components and pathways.
This allows many-core chips with some faulty cores to pass production test, since they will function for the full 100 percent without any compromise to reliability, according to the Crisp consortium.
"Chips can function with faulty components by developing architectures of multiple cores that can degrade while they keep functioning,' according to Hans Kerkhoff, Associate Professor, CTIT, University of Twente.
A run-time resource manager dynamically determines which core does which task and swaps tasks from failing cores to let the chip repair itself, said Kerkhoff.
The resource manager continuously determines the chip's optimum Quality of Service on fault-free components, according to Bart Vermeulen, Senior Principal Scientist at NXP, one of the consortium’s company partners.
The resource manager dynamically assigns new tasks to free resources during the entire chip lifetime.
Crisp which stands for Cutting edge Reconfigurable ICs for Stream Processing is a European Union project that researches optimal utilization, efficient programming and dependability of reconfigurable many-cores for streaming applications.
The project consortium consists of Recore Systems (project leader), University of Twente, Atmel Automotive, Thales Netherlands, Tampere University of Technology, and NXP Semiconductors.
Details of the program can be found here.
DATE 2011, the Design Automation & Test in Europe conference, runs this week in Grenoble, France.
|
|
|
|
|
Navigate to related information
DrQuine
3/19/2011 12:19 PM EDT
The ability of defective chips to heal on the production line and throughout the entire chip lifetime is a positive advance to improve yield and reliability. Unlike placing chips in sockets (with higher failure rates than the chips themselves) for maintainability, this is a solution that will improve overall system reliability. The remaining question is how to set quality acceptance criteria at the factory? How much production line repair is tolerable and what represents a failing chip that won't be viable in operation?
Sign in to Reply