Thanks for the video. I believe one of the non-functional requirement listed was URL's not being predictable Do you think there is a workaround to tweak this design which would make it less predictable?
Isn't there a possibility of collision using this approach too? Example: for counter value 500000 Base62 hash is GbaifA3M, for counter value 500001 Base62 hash is GbaifA3N. Now if you take the first 7 characters then it ends up in a collision.
bhai achi he sari videos!!! bas ye redis zookeeper wagera ni h playlist me ek poora netflix pe ya koi uber type app ka system design video needed h aapse
@EngineeringDigest: Insted of zoo-keeper why we can't use MySQL servere here. Because we can use a base conversion algorithm (e.g., Base62) to convert the id to a short URL. This ensures uniqueness and compactness. You explained that, we will zoo-keeper to get the range of last record only. Any other specific advantage of using zoo-keeper?
I have recently worked on short URL. To generate a Random number I have taken the ID of user and time stamp to generate Short URL of 8 characters. So I don't think I need an extra condition to check if that url is already present in database as the combination of the URL consists of time stamp which will obviously differ in milliseconds. Let me know if I am correct. I am a fresher so if someone could correct me that will help me to understand things better.
Nice explanation but a few questions: 1) How will you manage availability of application servers? For eg if one server crashes then is how will it be replaced? Will zookeeper assign a new server with a new range? What happens when the crashed server recovers and it still has range left? 2) You mentioned read write replica for Mongodb. How will the read server duplicate data from write server? Will it be sync or async? We might need write through (sync) writes for high durability but it will increase the write time. So what's your thoughts on this?
For your second question, I think async would be better because we want high availability. Because write is very less compared to read consistency won't be a big issue. What do you think about this? Correct me if I am wrong
Your explanation was great for resource assumption and other things but I felt that the Design section was a bit compromised. Please for the next system Design video elaborate on the design bit more.
Agar Hashmap avoid hi karna tha to itne bakaiti ka fayda kaise hua? Last me kis url k liye konsa random number generate hua and uska base 62 kya hai ye bhi to kahi se map krke nikaloge hi na ? Hashmap to usme bhi lagega.
you did not tell how to convert a number to 7 character base62 string. say my counter is just at 2 how that will convert to a 7 char base62 string? its just a adhura gyan
This question was asked to me in an interview of a stratup and now I feel how wrong my answers were. Thanks for this video!!
This playlist is amazing.
Premium Content in less time.
Bhaiya aapne bhaut acche se smjhaya hai mujhe bhaut help milli project smjhane me
We want more such System design solved examples from you.. You explain very nicely.
I will try my best 😃
I have watched all 34 videos, well explained thanks
Can you please create any social media app system design video for getting basic fundamentals
kaafi simplistic and straight forward expalanation , good for quick revision perspective but lacks indepth explanation.
Much better than Paid courses
Best video till now I have seen ❤
Very nice explanation sir, Thank you!
thanks
Very well explained Sir
So much clean and clear explanation ..
Thank you 🙏
Nice explanation...thank you
Thanks for the video. I believe one of the non-functional requirement listed was URL's not being predictable
Do you think there is a workaround to tweak this design which would make it less predictable?
In which video can I find about partitioning techniques and consistent hashing?
Isn't there a possibility of collision using this approach too? Example: for counter value 500000 Base62 hash is GbaifA3M, for counter value 500001 Base62 hash is GbaifA3N. Now if you take the first 7 characters then it ends up in a collision.
bhai achi he sari videos!!! bas ye redis zookeeper wagera ni h playlist me ek poora netflix pe ya koi uber type app ka system design video needed h aapse
Bhai make more of these kind ❤❤
Very nice explanation ... thankyou
Glad you liked it
@EngineeringDigest: Insted of zoo-keeper why we can't use MySQL servere here. Because we can use a base conversion algorithm (e.g., Base62) to convert the id to a short URL. This ensures uniqueness and compactness.
You explained that, we will zoo-keeper to get the range of last record only.
Any other specific advantage of using zoo-keeper?
nice video..looking forward for more videos like this
Can u do a system design for paytm or any other payment application? I would really like to learn about that
Sure, will do it
super ❤
Nice explanation
Nice video, more please
We want more system design videos
Cant we generate ascii number corresponding to url instead of using zookeeper for counter generation?
Gagar mein sagar, tahe dil se shukriya !
60gb in total can be cached. TTL need not to be set.
60GB free nai milti
Lets say you don't want to use zookeeper. Db is postgresql how you can achieve it!?
Asked in interview
Why can't we use mongo db object id which is always unique
I have recently worked on short URL. To generate a Random number I have taken the ID of user and time stamp to generate Short URL of 8 characters. So I don't think I need an extra condition to check if that url is already present in database as the combination of the URL consists of time stamp which will obviously differ in milliseconds. Let me know if I am correct. I am a fresher so if someone could correct me that will help me to understand things better.
sir plz can you implement this in nodejs ? for deep understanding plz
great work..
Nice explanation but a few questions:
1) How will you manage availability of application servers? For eg if one server crashes then is how will it be replaced? Will zookeeper assign a new server with a new range? What happens when the crashed server recovers and it still has range left?
2) You mentioned read write replica for Mongodb. How will the read server duplicate data from write server? Will it be sync or async? We might need write through (sync) writes for high durability but it will increase the write time. So what's your thoughts on this?
For your second question, I think async would be better because we want high availability. Because write is very less compared to read consistency won't be a big issue. What do you think about this? Correct me if I am wrong
1.25 M * 500 bytes = ~0.6 GB
Bolo system design waale baba ji ki jaii🎉🎉
jai 😆
Your explanation was great for resource assumption and other things but I felt that the Design section was a bit compromised. Please for the next system Design video elaborate on the design bit more.
Thanks for feedback!
Agar Hashmap avoid hi karna tha to itne bakaiti ka fayda kaise hua? Last me kis url k liye konsa random number generate hua and uska base 62 kya hai ye bhi to kahi se map krke nikaloge hi na ?
Hashmap to usme bhi lagega.
Attention problems
0.58 gb approx to 1gb
Good
Why are we using mongo here?? Can someone provide better explaination?
aap haryana se h kya?
Fbls , nvr seen the so dept video
thanks :)
Don't forget to join our discord
plz share the slides
you did not tell how to convert a number to 7 character base62 string. say my counter is just at 2 how that will convert to a 7 char base62 string? its just a adhura gyan
Msst
Thanks