Are you kidding me, you are killing it out here more than seasoned professional in youtube, keep em coming it's worth watching your vidoes, be it short or long (most of the senior folks will prefer longer formats as you get into more nitty gritty) but please keep em coming. A BIG YES TO PREMIUM CONTENT---- Awaiting :)
This length is perfect to watch when grabbing lunch or dinner. The longer ones are better suited when you're at your study table so this new format fills the gap. Thanks for doing this.
Great video. Could you release a short video like this discussing when to use SQL vs NoSQL database? I know most of your videos mention it doesn't really matter these days but some interviewers still ask me to discuss DB tradeoffs for SQL vs NoSQL. Thanks in advance!
Wanna say thank you for your videos and how helpful they've been. I had my L5 amazon system design interview the other day and I followed your guide. Interviewer was impressed with the level of detail and structured approach. Again, thanks a lot! Keep up the awesome work!
Thanks for getting this out! This is one of those topics where engineers think they know because they have memorized "pick two from CAP", but really don't understand.
Firstly, I love every bit of your content! It is hands down the best free resource available on internet for System Design! And hence I am worried about your "Premium Content", my concern being, like everywhere else, eventually the good stuff will land on the Premium tab. [Not saying it isn't worth it though!] Hoping you would stick to your thought: "Moreover, it's 100% free forever. While we could charge for this material we think it's far more important to level the playing field of system design interviews." Good luck anyway! Learnt a lot from you guys and hoping to continue doing so! Thanks!
I just finished reading your blog on dynamo DB (like 10 mins ago) and just when I laid down and opened RUclips to watch some videos before I sleep, there's a new video ahahah!. I just finished watching it, and stopped it as soon as you said "don't watch it further if you're a junior/mid level engineer (mid level in my case" Thank you for making such videos and topics. Tomorrow I'll read up Cassandra blog and the Food Delivery app System design(I forgott the same lol) Thanks again and goodnight :)) Watcjiny your Videos is helping me understanding missing gaps to fill up for my interviewes and giving me a LOT of confidence. Please don't stop🤞🏼🙏🏻
Great format for quick revisions. Although, maybe you could include more depth on the CAP models to be used during the HLD videos. Eg. Like when in a HLD question we want the causal consitency model as the user expects ... Great going.
Thank you so much for the awesome content! ❤❤❤Those deep dive concepts and real world examples are super helpful. #1 channel to recommend.💯💯💯 May I ask for Staff+ interviews, is the level of details & depth in your video above the hiring bar? or people should go even deeper? Similarly for other topics that you have shared before like geographical search, leader election, file loading to blob... does Staff + interview require deeper details than your video? Thanks
Staff+ is typically still evaluated at most companies as Staff. Many will have two system design interviews though. One general and one domain specific. In the domain specific more depth is certainly required, but it’s narrowly focused on that domain, whatever it is.
How did you get EXP in sys design? Like i would have never thought of having 'chunking' in an upload/storage system like drop box in a sys design. A future video on your journey from jr -> staff & resources you used along the way would be golden. As of now, i am still pondering what resources you used to gain vast experience. I myself use Alex Xu's sys design book and have read parts of designing data intensive application for system design. I feel like i am still lacking depth. Appreciate it.
For this you should start reading engineering blogs/videos from all these companies, like you mentioned dropbox. here is a talk from 2012 ruclips.net/video/PE4gwstWhmc/видео.html No book can give you real world experience unless they have actually implemented it hence these engineering blogs are gold-mine, keep reading exploring and actually implementing all these concepts, consistent hashing, chunked uploads(don't use library at first go break it on your own and assemble in the server)
@@jain007neeraj gotchu, that's a great resource which I have never seen it get pointed out as much as sys design books. I will start there. I appreciate the help and shall start there. Thank you kind stranger
Hey, at the end of this video I teased a premium offering that will have more content (the free content will stay free and continue to grow!). Online auction is done and part of that premium offering launching end of this month along with other popular problems like google docs, job scheduler, strava, yelp, etc.
@@hello_interviewThis sounds super interesting. For content as good as yours we will be more than happy to pay. Thank you for making system design so interesting.
Thanks for the great content. I have a separate question. In case we want to create a system that wants 10,000 nodes to work together to do webcrawling with minimum network calls and no centralized manager like zookeeper. Do you think p2p connection with DHT managed by Kademlia or Chord algorithm will work. Also can you give couple of pointers on whether to choose kademlia or chord ?
what about payment service for an ecommerce service where a payment service interacts with Payment Service Provider and further to payment gateways. Can we say it employ different consistency levels for different parts of the system.For example, within internal payment service it should be strong read consistent or high value/critical transactional data might require strong consistency, while less critical data could have eventual consistency.?
Can't we achieve a strong consistency with Cassandra within a single partition? Speaking of Ticketmaster, if we partition by event, all reads/writes that demanding a strong consistency will end up on a single partition, it's it?
Kind of. It depends on your design. Cassandra does not support multi-table or multi-row atomic transactions. So, for example, if you wanted to update an "available tickets" column on the event table and add a ticket row, you need some workarounds, which would be easier served by just using an RDBMS.
This video was good but again it fails to cover what other videos/material fail at. What is the proof of CAP's theorem? Lets say if I sacrifice partition tolerance then I should be able to have consistency and availability as per this. But if there is partition tolerance then I ll loose on both consistency and availability. Now consider a single node database. There wont be any partition so it is partition tolerant. This means I should be able to be consistent and available. But if the node is down then I loose both consistency and availability. Its more like if I am not available then I cannot be consistent as I have one node. This goes against the theorem. I might be wrong here but I was expecting to clear my misunderstandings here.
CAP theorem is about distributed data stores and doesn’t make sense in the context of single node systems. If you sacrifice partition tolerance you’re guaranteeing a reliable network - should be straightforward to see why you can guarantee consistency and availability in this setting.
Shoot, I forgot to say the obligatory, "don't forget to like and subscribe!" Still a RUclips n00b😉
Subscribe and don't forget to smash that bell icon 😅
@@sudosai😂😂
Are you kidding me, you are killing it out here more than seasoned professional in youtube, keep em coming it's worth watching your vidoes, be it short or long (most of the senior folks will prefer longer formats as you get into more nitty gritty) but please keep em coming.
A BIG YES TO PREMIUM CONTENT---- Awaiting :)
Not a youtube n00b at all, you're just better than the rest
we got you 😊
This length is perfect to watch when grabbing lunch or dinner. The longer ones are better suited when you're at your study table so this new format fills the gap. Thanks for doing this.
Appreciate the straight forward titles, thumbnails, and content.
no bs
As always, your videos are great and these short videos are good to catch during a break. Thanks a lot for such amazing content!
Great video. Could you release a short video like this discussing when to use SQL vs NoSQL database? I know most of your videos mention it doesn't really matter these days but some interviewers still ask me to discuss DB tradeoffs for SQL vs NoSQL. Thanks in advance!
Wanna say thank you for your videos and how helpful they've been. I had my L5 amazon system design interview the other day and I followed your guide. Interviewer was impressed with the level of detail and structured approach. Again, thanks a lot! Keep up the awesome work!
Love to hear that! Well done to you :)
Thanks for covering this topic. The video wasnt too long. It balanced depth while also covering a wide range of practical scenarios which was great.
Glad you found it useful!
Thanks for getting this out! This is one of those topics where engineers think they know because they have memorized "pick two from CAP", but really don't understand.
Totally
Firstly, I love every bit of your content! It is hands down the best free resource available on internet for System Design! And hence I am worried about your "Premium Content", my concern being, like everywhere else, eventually the good stuff will land on the Premium tab. [Not saying it isn't worth it though!]
Hoping you would stick to your thought:
"Moreover, it's 100% free forever. While we could charge for this material we think it's far more important to level the playing field of system design interviews."
Good luck anyway! Learnt a lot from you guys and hoping to continue doing so! Thanks!
Very nice. One of the best explanations I've met so far.
Right on!
Insane clarity with sufficient depth and detail. Love how you backed each one up with examples too
Glad you liked it!
I just finished reading your blog on dynamo DB (like 10 mins ago) and just when I laid down and opened RUclips to watch some videos before I sleep, there's a new video ahahah!.
I just finished watching it, and stopped it as soon as you said "don't watch it further if you're a junior/mid level engineer (mid level in my case"
Thank you for making such videos and topics. Tomorrow I'll read up Cassandra blog and the Food Delivery app System design(I forgott the same lol)
Thanks again and goodnight :))
Watcjiny your Videos is helping me understanding missing gaps to fill up for my interviewes and giving me a LOT of confidence.
Please don't stop🤞🏼🙏🏻
Perfect timing!
Great overview on CAP theorem. Straight forward and will definitely be helpful.
Very useful! Please continue making such informational content. Cheers!
Love your videos, very informative and insightful!
Can you please also share/link the Excali draw link in the video description for reference, thanks!
Oh yes of course! Thanks for the reminder, adding now. Also direct linking: link.excalidraw.com/l/56zGeHiLyKZ/8ntWRaa0Q6K
Great video!
Can you do a video on sharding and how to generally handle hot partitions?
Brother its very informational quality content that you put out there.
Happy to see one last video before my interview tomorrow :)
having my one tomo as well, wish you luck!!!!!!!
Good luck to you both! Go crush it 💪
Very clear and crisp. Thank you 🙏
Format is good - nice to get small deep dives
Amazing as usual! I prefer long videos to go deeper in details if the use case needs it.
Those won't stop, don't worry :)
I really appreciate the clear and concise explanation. It's easy to understand..
🫶🏻
Great format for quick revisions. Although, maybe you could include more depth on the CAP models to be used during the HLD videos. Eg. Like when in a HLD question we want the causal consitency model as the user expects ...
Great going.
just having my system design interview tomorrow, thank you I think you are on time xD
You got this!!
Thanks for design interview focused explanation. Like to see more such videos that explains concepts.
You got it!
Great content , perfect format
Thank you! Your content is the best in class!
Glad you like it!
Thank you so much for the awesome content! ❤❤❤Those deep dive concepts and real world examples are super helpful. #1 channel to recommend.💯💯💯
May I ask for Staff+ interviews, is the level of details & depth in your video above the hiring bar? or people should go even deeper?
Similarly for other topics that you have shared before like geographical search, leader election, file loading to blob... does Staff + interview require deeper details than your video?
Thanks
Staff+ is typically still evaluated at most companies as Staff. Many will have two system design interviews though. One general and one domain specific. In the domain specific more depth is certainly required, but it’s narrowly focused on that domain, whatever it is.
Great video, explains the concept with examples which helps in setting the context.
💙
Thanks about great content, Would be grateful if you do a detailed comparison SQL vs NO SQL ?
Your videos are awesome as always!
🫡
Great format, thank you!
You bet!
thanks for going into this!
Hi Trisha 👋 pleasure :)
great format!
How did you get EXP in sys design? Like i would have never thought of having 'chunking' in an upload/storage system like drop box in a sys design.
A future video on your journey from jr -> staff & resources you used along the way would be golden.
As of now, i am still pondering what resources you used to gain vast experience. I myself use Alex Xu's sys design book and have read parts of designing data intensive application for system design. I feel like i am still lacking depth.
Appreciate it.
For this you should start reading engineering blogs/videos from all these companies, like you mentioned dropbox. here is a talk from 2012 ruclips.net/video/PE4gwstWhmc/видео.html
No book can give you real world experience unless they have actually implemented it hence these engineering blogs are gold-mine, keep reading exploring and actually implementing all these concepts, consistent hashing, chunked uploads(don't use library at first go break it on your own and assemble in the server)
@@jain007neeraj gotchu, that's a great resource which I have never seen it get pointed out as much as sys design books. I will start there. I appreciate the help and shall start there. Thank you kind stranger
Nice explanation.
great video. makes perfect sense
Loving these meta no pun intended videos
i loled
Hey love your content, wanted to know if the auction system design blog/ video is coming out
Hey, at the end of this video I teased a premium offering that will have more content (the free content will stay free and continue to grow!). Online auction is done and part of that premium offering launching end of this month along with other popular problems like google docs, job scheduler, strava, yelp, etc.
@@hello_interviewThis sounds super interesting. For content as good as yours we will be more than happy to pay. Thank you for making system design so interesting.
like your style, clean and concise! looking forward to the premium content.
Top tier content
Amazing, quite informative.
🫶
Thanks for the great content. I have a separate question. In case we want to create a system that wants 10,000 nodes to work together to do webcrawling with minimum network calls and no centralized manager like zookeeper. Do you think p2p connection with DHT managed by Kademlia or Chord algorithm will work. Also can you give couple of pointers on whether to choose kademlia or chord ?
God! I wish you taught CS.
😛
gold mine
what about payment service for an ecommerce service where a payment service interacts with Payment Service Provider and further to payment gateways. Can we say it employ different consistency levels for different parts of the system.For example, within internal payment service it should be strong read consistent or high value/critical transactional data might require strong consistency, while less critical data could have eventual consistency.?
Can't we achieve a strong consistency with Cassandra within a single partition? Speaking of Ticketmaster, if we partition by event, all reads/writes that demanding a strong consistency will end up on a single partition, it's it?
Kind of. It depends on your design. Cassandra does not support multi-table or multi-row atomic transactions. So, for example, if you wanted to update an "available tickets" column on the event table and add a ticket row, you need some workarounds, which would be easier served by just using an RDBMS.
This video was good but again it fails to cover what other videos/material fail at. What is the proof of CAP's theorem? Lets say if I sacrifice partition tolerance then I should be able to have consistency and availability as per this. But if there is partition tolerance then I ll loose on both consistency and availability. Now consider a single node database. There wont be any partition so it is partition tolerant. This means I should be able to be consistent and available. But if the node is down then I loose both consistency and availability. Its more like if I am not available then I cannot be consistent as I have one node. This goes against the theorem. I might be wrong here but I was expecting to clear my misunderstandings here.
CAP theorem is about distributed data stores and doesn’t make sense in the context of single node systems. If you sacrifice partition tolerance you’re guaranteeing a reliable network - should be straightforward to see why you can guarantee consistency and availability in this setting.
Great video. Perfect length. Wish I had known about these for my last two interviews 🥲. Very helpful though
You'll crush the next one!