URL shortener system design | tinyurl system design | bitly system design

What if the Zookeeper goes down?

I'm still confused how zookeeper based approach solved two users calling the service for the same long url at the same time. Won't we generate the two different tinyurl?

Thank you so much! This video was very helpful.

Great lecture!

What if zoo keeper goes down. Doesn’t it makes it a single point of failure. .. ???

Best explanation on internet, Thanks a lot 🙂

while this videos is very informative why the hell are you mixing indian and british accents in a every damn sentence.. while the subject matter seems quite easy my real difficulty is understanding your fake accent.

Thanks for the Video. How counter range impacts the Short URL generation when we apply at the App server level.

such good quality content, Hats off

Really really awesome video. Thank you for your great efforts!

If we redeploy the server does it mean we get a new set of ranges from zookeeper? won't we exhaust the range very fast by that?

There is still a collision problem. two counter outputs can have same b62 value or md5 hash output (at least the first 7 characters)

You can save some storage by having the domain on the application properties from AS

Very good explanation, and solving by breaking problems to small level and solving is more good.

I see majority of problems were solved here but what about duplication, if the same long url is shortened by multiple users, the system is going to generate multiple short urls. This is redundant isnt it??

How can we solve this? Make a read once before we generate and write??

Isn't the ZooKeeper, a single point of failure?

can we run a replicate database (cassandra or dynamo db) for storing available range instead of using zookeeper.

I have one query, why we are going for hasing, simply get the unique id generated in the table and use that?

Very complex design

Great !!!!!!!!!!!!!!!!!!!!!!! Awesome !!!!!!!!!!!!!!!

Great job, Can you further explain why choosing no SQL over SQL for this design?

How is counter ensuring that we get a unique short url everytime ?

Thanks bud! Love your lectures.

What if the zookeeper fails?

While registering to zookeeper I would have added a unique string for different app service with the same range counter. I think that will also solve the mentioned issue.

Amazing explanation. But I've one question, what happens when all ranges in zookeper are exhausted? Is there a way to do some sort of compaction on ranges to reclaim them?

i have a simple question ..let us say i am creating the Short URL and will reach the Server 1 (range 1L-2L) .. i may get first counter is 100000 .. its success. next request comes how do make sure we will get the next number ie i am trying to create the short url by tmw and reach out the server 1 again .here how do get the next number since first number is already used .we supposed to get the 100002 ...

hi bro, can you keep updating vedio? can not wait your new shares

Great video! Just one thing I don't understand. How can we be sure that base62 encoding will always produce string with length seven?

Can't we use something like (Timestamp (Per Nano Second) + Redis Counter) => This will make sure that we have a unique id??

very much clear. Thank you. If long url record already exists in database then you will simply fetch short url right?

How do the databases scale out? If the table grows and grows lookup queries when you have cache misses will become a significant bottleneck. Replicas won’t help you here. Even if you sharded instances you’d still need a centralized table because you would need to lookup a specific instance in order to get the long URL given a short URL. Thoughts?

Didn't understand how counter works here and how it guarantees no collisions

great video!

We can do it in NoSQL like mongoDB - checking duplication using unique : true in mongoose which will be throwing us duplication error

It is true that base62 won`t produce collisions with counter? I don`t think so.

Why are you using the first 7 characters? Or did you arbitrarily decide that?

I wish you showed how to actually get the next available range of numbers from zookeeper

increible!!! gracias por compartir hermano!!! mis respetos, saludos desde Mexico.

Question about zookeeper here: Say zookeeper gives AppServer1 the counter range 1 to 10,000. Does AppServer1 constantly report to zookeeper what number it's currently up to? If not, then if AppServer1 goes down at counter 5,000, do we lose all ability to use values 5,001 to 10,000 for ever?

Also - great presentation I learnt a lot. Thank you so much for this!

Xin chào

What if Zookeeper goes down ?

why cant we prefix the counter with some character to uniquely identify the server, like A1, A2, ... B1, B2, .... C1, C2,...

nice explanation. many thanks. all the best. keep going !!!

this is a legit gold mine in terms of information. Really informative and you get to understand the details

can't we use, queue data structure instead of zookeeper to handle coordination?

I guess to ensure uniqueness across systems, we can append timestamp and system name to the random number