Re: Old programmers do NOT fade away

2020-12-30 Thread Martin Gainty
Hey Erick you should've been PMC with the immense amount of work you put into this project i would be happy to pickup some of the potential slack maybe the bad-apple-report ? BUT MY SCHEDULE CAN BE RANDOMLY FILLED WITH playing with grandkids Helping out eldest daughter..(something about father

Re: Old programmers do fade away

2020-12-30 Thread Alexandre Rafalovitch
Erick, I kept hoping to meet again at a future conference and have an extended version of the talk we had the first time we met. It was very valuable but I felt I only got a glimpse of what was possible. Perhaps, one day, I can travel near your actual "nest" and buy you a beer or two and listen

Re: Old programmers do fade away

2020-12-30 Thread Christian Moen
Take care, Erick. Thanks for all your contributions. Lots of people will miss you. I wish you all the best. On Wed, Dec 30, 2020 at 3:09 PM Erick Erickson wrote: > 40 years is enough. OK, it's only been 39 1/2 years. Dear Lord, has it > really been that long? Programming's been fun, I've

Re: influence of memory access patterns on dot-product performance

2020-12-30 Thread Michael Sokolov
Oh JMH is a good idea, I'll try that, thanks On Wed, Dec 30, 2020 at 10:31 AM Robert Muir wrote: > > Can you boil this down to a microbenchmark (e.g. JMH) so you can look > at assembly? > > Maybe with on-heap, the dot product is getting vectorized, but with > off-heap/unsafe it is not. > e.g.

Re: Old programmers do fade away

2020-12-30 Thread Michael Sokolov
Woah! That plan sounds like fun! I might have to join you, but not yet :) On the topic of squirrels, you must have seen this https://www.youtube.com/watch?v=hFZFjoX2cGg, but I share it again because it always deserves a second watch. Also, my own personal attempt at squirrel-proofing didn't go so

Re: influence of memory access patterns on dot-product performance

2020-12-30 Thread Robert Muir
Can you boil this down to a microbenchmark (e.g. JMH) so you can look at assembly? Maybe with on-heap, the dot product is getting vectorized, but with off-heap/unsafe it is not. e.g. something like this recent bug: https://bugs.openjdk.java.net/browse/JDK-8257531 You could re-run your bench on a

RE: Old programmers do fade away

2020-12-30 Thread ufuk yılmaz
I can only wish to be as productive as you 35 years from now, you have been an inspiration to me even though I only know you through your answers to hard questions here and presentations on youtube  Hope you will enjoy your free time even more from now, and find a way to protect those

Re: Old programmers do fade away

2020-12-30 Thread Bram Van Dam
Your presence on the user mailing list will be sorely missed. You've been an invaluable source of good advice for years. I owe you at least $bignum beers. Best of luck with the squirrel problem (they are rather tasty) and any other fun projects! Take care, and thank you, - Bram

Re: Old programmers do fade away

2020-12-30 Thread Atri Sharma
It has been a privilege to watch you work your magic and reference the great work that you have done in this project. Thank you for setting the bar for us. I am happy to support the BadApples report if you would like On Wed, 30 Dec 2020, 19:46 Erick Erickson, wrote: > 40 years is enough. OK,

Re: Old programmers do fade away

2020-12-30 Thread Ilan Ginzburg
Hey Eric, Sad and happy to read your message. You've been a clear voice in the Lucene Solr community and I was always AMAZED how willing you are to help and explain, over and over again when needed. That's the sad part. The happy part is that those squirrels do need to learn and the electric

Old programmers do fade away

2020-12-30 Thread Erick Erickson
40 years is enough. OK, it's only been 39 1/2 years. Dear Lord, has it really been that long? Programming's been fun, I've gotten to solve puzzles every day. The art and science of programming has changed over that time. Let me tell you about the joys of debugging with a Z80 stack emulator that

influence of memory access patterns on dot-product performance

2020-12-30 Thread Michael Sokolov
Hi, I've been working on improving performance of vector KNN search, and found some behavior that seemed surprising to me, showing huge differences in some cases comparing on-heap memory access with the way we access data today via IndexInput. I'd love to get some other eyes on this to help me