Відео

Heisenberg laughing in the background😂😂😂😂

Nicely done! I knew nothing about websockets before this. Now I feel like I have a basic understanding. You could read some documentation and articles for 10-20 minutes or you could watch this 2 minute video. I'll go with the video.

Nice visual explanation!

Thank you

the challenge here is counting bets, hits and misses, querying all that could turn out very taxing so there has to be a smart implementation to achieve that

thank u!

Insightful

This guy has been doing javascript for over 85 years.

why is the set of all sets a paradox, i dont see the problem if it can contain itself

Summarized: Intro (00:00:00) Creating a URL shortener, similar to TinyURL. Users input a long URL and receive a shortened version. Requirements (00:00:10) Short URL expiration: Set to 100 years (effectively permanent). Custom URLs: Allowed, with a maximum length of 16 characters. Expected requests: 100 million shorten requests per month. Functional Requirements Shorten URLs. Redirect short URLs to their original long URLs. Shortened links should be seven characters long. Users can create custom short URLs up to 16 characters. Short links remain in the system for 100 years. Non-functional Requirements Continuous service uptime. Fast URL redirection, even during high traffic. REST API exposure for external application integration. Facts and Figures (00:01:42) Read-to-write ratio: Approximately 200:1 (200 lookups for every short URL created). Expected short URLs generated per month: 100 million. New short URLs created per second: 40. Expected reads or redirects per second: 8,000. Total objects in the system (assuming 100-year storage and 100 million new URLs per month): 120 billion. Storage (00:02:27) Each object is 500 bytes, requiring a total of 60 terabytes of storage. 700 million read requests per day, with 80% of requests for 20% of data. 20% of requests require 70 gigabytes of memory. 40 URLs generated per second. 120 billion short URLs generated in 100 years. 8,000 URL reads per second. Server Requirements (00:05:45) URL generation service must produce unique short URLs that do not point to the same long URL. URL encoding solutions like Base62 or key generation services can be used to achieve this. Security concerns exist with certain techniques, such as MD5, which is vulnerable to collision attacks. Best practices for URL encoding are constantly changing, so research is necessary for enterprise-level solutions. Best Practices (00:06:33) Keep short links as short as possible, ideally no more than seven characters. The application should be resilient to load spikes for both URL generation and redirections. Following a short link should be fast. NoSQL (00:09:07) A NoSQL database might be better suited for this system due to its huge storage size (60 terabytes) and high amount of reads and writes (8,000 URL reads per second, 40 URL writes per second). NoSQL databases are easier to scale compared to relational databases. NoSQL databases provide features like sharding and replication by default, making it easier to distribute data across multiple machines. Using a hashing strategy to generate short URLs is preferred over using a key generation service, as it eliminates the need for a dedicated database to store already generated random seven-character strings. Caching (00:10:48) To optimize performance, use a caching mechanism like Memcached or Redis to cache 20% of daily traffic, as 80% of clicks are on 20% of links. Implement a linked hash map to store URLs by hash, allowing for constant time lookup and preserving insertion order, and use a least recently used (LRU) policy to manage cache size and replace old entries. Place load balancers between clients and application servers, application servers and database servers, and application servers and cache servers to distribute requests and manage traffic. Use a round-robin strategy initially to distribute requests evenly among servers, then implement health checks to adjust load distribution based on server performance. Require custom URLs to be at least eight characters long to minimize conflicts with randomly generated URLs. Consider a dedicated database instance and web servers for custom URLs if they are a paid service.

Hey i like your system design videos can you make more of that if possible

Priceless video, please keep sharing! Appreciate it!

Simple and just deep enough for non coder to understand. Awsome !

Hands down top 2 system design vid on TinyURL on this site.

*Promo sm*

1 year experiences SDE , currenlty unemployed struggling for 3 months. Strugglin for a job. Have been learning DSA, MERN from past 4 months, system design and SQL is only subjects that are left now, after that will be ready to approach decent companies. Found no free good precise resource for system desigin but then came acorss your videos. I request to kindly create system design crash course playlist which precise to the point in biref information, helpful to crack SDE1 interviews.

The estimation of cache size would be good enough? I think we should estimate cache size based on shorten urls that are more getting heated then others url. So, If it's only 20% of shorten urls that are created with in a month, we can roughly expect that it would be ((500 bytes * 100 mil) / 100) * 20 = around 9 GB.

Hi In article with base64 conversion there is a problem to address. We have a counter that is shared across servers, so that counter is critical section and race condition might arrive as diff servers can read same value and generate same short-url. We must ensure mutual exclucion in that case. Correct me if i am wrong :)

15 mins of video and hours of value, great video Loved it👍 Would love to see more videos like this on designing systems

When talking about choosing between RDS vs NoSQL, IMO I was a little uncertain when it mentioned strong (RDS) vs eventual consistency (NoSQL). To RDS with a single instance, it might be confident to state that it can align with strong consistency, but when comes with replication nodes, RDS may also not guarantee consistency

I can't stress this enough: your content is super valuable.

Great one, mate! 👏A video on when to use cache, local storage and in memory would be appreciated! 😊

hornpub😂😂

Thank you so much for the quality content. Please make a series of system design! I feel so struggle and inefficient when designing complex system.

one of those videos where you get most from viewing time - very concise, effective, no side bs, covers edge cases etc 💜

I can't stand VSCode

Amazing video and explanation. Is it possible that you make a coding part where you actually build the component, it would be amazing. 👌👌

I think you should have expanded on how to compute that tinyurl. Its more relevant than explaining the lru which was very superficial

Amazing simplicity bro, keep rocking

Very well done content, mate. Keep on!

Why is your channel not being recommended more? This is great stuff!

Very Interesting.

Can you please share the blog post you mentioned at the end?

where does the 70GB cache storage come from? Given 60TB total storage, wouldn't 20% of 60TB be around 12 TB?

your channel reminds me of that scene from the matrix where they just upload free knowledge into Neo's brain...

man I come to each of your videos expecting about x100 more views than what they have - IF I EVER MAKE IT, I'M GOING TO REDESIGN THE UA-cam ALGORITHM TO ENACT JUSTICE ON YOUR VIDEOS BC THEY'RE CRIMINALLY UNDERRATED

dude your videos are actually brain p*rn, thank you

Love your videos. Please do more!

Definitely helpful. Love your videos

Simple and yet scalable design compare to others. Thanks a lot...

good example of how join works. btw, actually this is not how the twitter timeline works. In a very brief way. it works based on fan out service and followers feed will be updated after a new tweet is posted by a person.

Bro can you suggest me some course for system design or do you teach system design from scratch

Hey, have you looked it into redis, it looks like a good option for in memory but you can also reconstruct from disk in case of restart or failure

can people really do all this math in their head that quick?

Amazing

I’m a design systems product designer and this is one of the most insightful development videos I’ve seen! Thank you for making this 🙌

Awesome your approach to software. This is what a lot of us are looking for.... Going beyond coding, and observe real engineering. Subscribed!

What a nice piece of knowledge and thought delivery 👏👏

What I think now is websocket connections are between client and server to push real-time updates to UI. And, Pub/Sub architecture is adopted by kafka that's nothing to do with websockets at all. Websockets is a protocol for communication between client and server alike HTTP/HTTPs. PUB/SUB is event driven to get almost real-time updates.

Very nice