- Xilinx User Community Forums
- :
- PLD Blog
- :
- SEU, Not Again!
Announcements
--- Win Prizes for Kudos! ---
From now on, every time you provide a good solution or post valuable information on Xilinx User Community Forums, you will be awarded Special Kudos by the Forum Administrators. Every six months, the top two Kudos winners will receive a very special prize from Xilinx!
So, for a chance to claim your well-earned reward, please continue to reply to questions and post elite messages on our Forums. Your contributions are always appreciated!
SEU, Not Again!
I know, I know, I know: Soft error upsets (SEU) are old news. Why do I continue to bring this subject up? Only because Xilinx FPGA devices are getting better, and better.
It Has Been Six Years
Lots and lots of counting (of upsets): 150 nm for 172.21 Gigabit-Years (Gb-Yr), 130 nm for 98.88 Gb-Yr, 90 nm for 63.8 Gb-Yr, 65 nm for 41.77 Gb-Yr. To what end?
The original purpose was to find out what our soft error rate really was, in real life. Quickly, the program became the following: How do we improve this at the next technology node? How do we control alpha-emitting contamination? How do we modify our circuit design? What new features are needed by the customer?
What is the Result, So Far?
Configuration Memory (per JEDEC89A)
150 nm: 396 FIT/Mb
130 nm: 375 FIT/Mb
90 nm: 240 FIT/Mb (V4)
65 nm: 138 FIT/Mb (V5)
The goal was to keep the soft failure rate of the largest device in our product line constant, or even decrease it. How are we doing?
Configuration Memory Size of Largest Family Member
150 nm: 20.9 Mb
130 nm: 21.9 Mb
90 nm: 40.8 Mb (V4)
65 nm: 63.9 Mb (V5)
So, how is the overall soft failure rate doing for the largest part?
150 nm: 396 * 20.9 = 8,276
130 nm: 375 * 21.9 = 8,213
90 nm: 240 * 40.8 = 9,792
65 nm: 138 * 63.9 = 8,694
Roughly doubling the capabilities of the largest device each generation, and the failure rate for approximately ten times the capability from 150 nm to 65 nm, the result is only 5% higher.
One side note: We still have NO soft logic errors. Can’t say that about 65 nm ASIC or ASSP!
If you only needed the capabilities of the largest part in 150 nm, the same functionality part in 65 nm is far more reliable, as it has more robust and reliable bits, and even fewer bits.
But, Here Is the Sad Part
We are alone. All alone. No one else has this track record. No one else can show these results. No one has succeeded in even keeping the failure rate constant. That is why it is so quiet: Be concerned, be very very concerned.
Match our published results against anyone. I really don’t care: ASIC, ASSP, uP, FPGA. You will find (after innumerable delays, NDA’s, and marketing ‘spin’) that Xilinx is alone in actually creating, selling, and supporting devices and proving that the soft error problems are actually getting less and less.
The newest devices offer built-in detection, and correction features for both configuration and user block RAM. There is a software tool to determine the de-rating factor for your design: Typically, only 1 in every 10 bit flips causes a functional failure (some designs have a de-rating as high as only 1 in 40).
The system architect may now take advantage of these new features to achieve the reliability and availability they require, by making straight-forward engineering choices: The resulting reliability and availability can all be predicted before the system hits the field.
Austin
-
kcmman
- Ken Chapman holds a first class BSc Degree with honours in Electronic and Electrical Engineering from the University of Surrey. Before obtaining his degree, he spent 4 years working in production environments, making precision instruments and working his way through all levels of a small electronics company. He spent 4 years at Racal Radar Defense Systems combining detailed digital design with all aspects of system integration. Ken joined the UK division of Xilinx in 1991, and was instrumental in developing innovative methods of implementing DSP functions in the Xilinx devices. He has filed several patents while at Xilinx, including the 'MULT_AND' gate seen in each Virtex™ and Spartan-II™ device that has made multipliers and other arithmetic functions smaller and faster.
-
peter.a
- Peter Alfke joined Xilinx in 1988 as director of applications engineering. He currently serves as Distinguished Engineer in the Advanced Products Group. He graduated in electronic engineering from the Technical University in Hannover, Germany in 1957. He went on to work in telecom and computer design with LM Ericsson and Litton Industries before moving to California in 1968. He has spent forty years in Applications Engineering with Fairchild, Zilog, AMD, and now Xilinx. He holds more than thirty patents, has authored many application notes, and given worldwide seminars on digital integrated circuits. He is active in the newsgroup comp.arch.fpga.
-
austin.lesea
- Austin graduated from UC Berkeley in 1974 and 1975 with his BS EECS in Electromagnetic (E&M) Theory and MS EECS in Communications and Information Theory. He has worked in the telecommunications field for 20 years designing optical, microwave, and copper- based transmission systems. He developed SONET/SDH GPS-based Timing Systems for 12 of those years. For the last ten years at Xilinx, Austin was in the IC Design department for the Virtex product line. His new role is working for Xilinx Research Labs, where he is looking beyond the present technology issues.
- Package Generated Alpha Particles: They are a Pro...
- Helping the Customer Drive to ‘Zero Defects’
- Targeted Platform Design
- It Is All About Quality
-
The ‘Programma
ble Imperative ’ -
Reliabilit
y - A True Winning Number Generator
- Spring Is Here!
- Happy New Year: Moore, More Moore, and More than ...
- SEU, Not Again!
-
austin.lesea
on:
SEU, Not Again!
- larthe on: Estimating Average Switching Rates
-
fpgasdr
on:
FIFOs, A Tutorial Descriptio
n -
gszakacs
on:
Terminatio
n is Never Easy -
austin.lesea
on:
Power to the PCB!
- tocktock on: Finding Stuff: Part 2
- seamusbleu on: Using Self Scanning Memory for State Machines
- brouhaha on: Finding Stuff
-
kcmman
on:
That Dangerous Asynchronous Reset! - q on: Quadrature Decoding with Resolution Options





