Databricks Photon – Fun With Vectors | Distributed Systems Deep Dives With Ex-Google SWE
Databricks Photon – Fun With Vectors | Distributed Systems Deep Dives With Ex-Google SWE
https://people.eecs.berkeley.edu/~matei/papers/2022/sigmod_photon.pdf
The ladies keep dropping hints at me out here, telling me they’re restricted
source
Reviews
0 %
Can you add a link for the paper you are talking about
Hello, brother. I believe your videos pack a lot of detail and knowledge; however, you talk in a tone as if you're gliding over a lot of things without going in depth on any of them. Also, you make statements before you start the videos that suggest uncertainty in the information you are about to deliver, as though the information might be wrong, and if it is, then we should figure it out together. It's like you're hedging.
I don’t know whether you think you don’t know enough or if you feel you're not the best person to teach this, but I believe you know a lot based on all the content you put out and your position in life in general.
I don’t know what your aim is with the channel, but I would assume it’s some kind of go-to source for this type of information, right?
What I think could be improved are:
Maybe you should cut down on downplaying your tutorials.
Maybe you should speak with a bit more authority.
Maybe you should also slow down a bit; some videos might benefit from not being compressed so much , i mean the informations in them.
Also, if you could include this in your videos: why something is the way it is, and how it ties into the greater picture of things. It could help with clarification.
Please this is not specifically about this video; its just what i have noticed about your vids i have watch so far;
These are just some of my sincere thoughts; please don’t take it the wrong way. I just love your videos so much. I even completed the book Designing Data-Intensive Applications, which you suggested, and I believe you are doing great work here.Thank you and keep it up Big bro .❤
reall happy to see his face skin gets better
I swear I know this guy from my hometown McDonalds.
As a Java suffer I can explain a little bit about the trade-off, at that time this project has been started on Java 8 and for those who understand a little bit about Java Ecosystem the code is not the game change but how the JVM will improve it over the time. In short, whenever you compile your code Java code for the a .class it will be interpreted by the JVM on JRE, Just-in-time, or JIT, has the responsibility to over the time your program is running it will analyse and perform optimisations over the existent bytecode C1 and C2 compiler (that is why on that time a app running for days had a performance better than if I've just started for a couple of hours). Although In my defense, today the Java world changed you have several of options JIT, Ahead of Time Compilation, and most recently CRaC (Coordinated Restore at Checkpoint). BUT when low latency is required basically you will avoid everything that has an JVM and will write it on Rust or ZIG
Topics suggestion: Ray Data (e.g., Exoshuffle)
Luigi was a data engineer lmao
Jordan whats your twitter
Nice! Hahahaha.