Видео

Awesome! Watched this right after Dave Chiluk's conference reference-this video perfectly complements it. It really helps paint a clear picture of what CPU throttling is and the key takeaways from it.

I don't understand why, in the masking intro slide (at 22:42), the author says that the following has no branches: for (int i = 0; i < N; i++) s += (a[i] < 50 ? a[i] : 0); That's a ternary operation, which branches between left and right expressions. What am I missing?

Hard to understand English and unpleasantly small text...

thanks so much for the tutorial

I personally use 9 performance principles - some of which are the same as the 5 outlined in this video. Some of the differences seem to be "CPU to data distance" and "data size" ... I cover the principles I use in my video "My 9 + 1 core performance optimization principles" : ruclips.net/video/ULlFWomaPVw/видео.html

This is cringe

Somebody should have taken a minute to optimize that terrible audio compression.

Adobe should optimise their software to have less source code in general based on the results of this. Write slimmer code and the performance isssue goes away.

Awesome 🇷🇺

The "Closure" part of the optimization is basically how Forth works: a code pointer and a piece of data for each "word".

Thanks for the video. Is there any new material on topic of "closure generation" interpreters ? Very hard to find material that is not bytecode based.

Thanks very appreciated. Especially the examples in C. Is this directky compatible in Cython ?

I don't understand why these architecture specific instructions are not recognized directly by gcc on O3.

Anybody really cares about latency what to use a RPC and streaming? If I can tolerate 18ms I did not see why I can not tolerate 100ms.

great video. Thank very much for your lightening example and insightful explanation!

Great talk. Thank you Sean.

great talk

This was amazingly useful! Loved it. Thanks for the great work!

Amazing and full of knowledge talk.

50:00 Sean might be referring to “Eytzinger binary search”

21:45 ruclips.net/p/PLGvfHSgImk4Y1thqJLpcSscMTwgN-XM9l

31:43 Great talk. More about optimization remarks and viewing them can be found in Ofek’s talks such as ruclips.net/video/6HbyacS5eZQ/видео.html

This guy is seriously a god.

What happened to Concord.io?

Audio is too low. The (section) flags are neat. As it's a performance presentation, it would have been neat to include a running tally as go through the optimizations. You do cover it in the conclusion, of course. Good job!

thank you laurence very interesting. i have a trading app written in java that takes a good few hours to speed up. this has given me the incentive to find out exactly when jit compilation is happening.

this was brilliant. thank you very much. next stop ScyllaDB

I recognise Suchakra's slide at 4:00 :)

Link in the desc is broken link

"Developers love Kafka api" - are you serious? After a year working with Kafka, I know some configuration choices and design problems I still shudder at

He is missing baseline jits. Very fast jit compilers. They are about 10x faster than the fastest interpreter

Benchmarking these approaches provide a very valuable insight for anyone considering these approaches! Thanks for providing all these learnings. 😎

Why I never see throughput benchmark for this XD.

Excellent sneak peak of query based compilers.

The discussion in the last 10 minutes is full of insights! Thanks!

Good talk. 👍

Any tutorial on this?

Here is the link to the GitHub project page: github.com/Genivia/ugrep

he stated that intra-datacenter network latencies nowadays between machines was on par with inter-NUMA latencies. that’s just categorically untrue. 100s of nanoseconds vs 100s of microseconds. that’s 3 orders of magnitude.

Fantastic talk! Thanks for the wisdom!

Great job guys, and thanks Nitin to share this discussion and your vision. I really liked the concept and topic which fits o app such Linkedin. Nevertheless, there were some missing points like: - Evaluation of used prognostic models (Accuracy for prediction concepts expressed by error analysis) - Compare ML (Xboost) Vs DL (DNN) or offline NQS - Type & quality of NNs architectures were used for predictive models (info about the best model and its hyperparameters) - feature importance mechanism was used for feature engineering just for Xboost since in Deep Learning it's done by NN itself. - Is it possible to access the dataset you used like open-source?