Engineering Bakar
Engineering Bakar
  • 9
  • 6 817
Understanding HyperLogLog: Space-Efficient Cardinality Estimation - Engineering Bakar
Join us in this video as we explore and discuss the intriguing concept of HyperLogLog. We kick things off by understandign the basic idea, history of different algorithms LogLog, SuperLogLog and HyperLogLog and concisely explain how HyperLogLog works and how to implement it effectively.
HyperLogLog is a powerful probabilistic data structure widely used for cardinality estimation in databases, big data systems, and distributed analytics platforms. Through our discussion, we aim to simplify this advanced concept and make it accessible to anyone interested in efficient data processing and scalable algorithms.
#HyperLogLog #CardinalityEstimation #BigData #ProbabilisticDataStructures #DistributedSystems #DataStorage #TechTalk #DataStructures #Algorithms #HashFunctions #TechTalks #ComputerEngineering #EfficientDataStructures
Переглядів: 35

Відео

Leader Election In Distributed Systems
Переглядів 2435 місяців тому
Welcome to our deep dive into Leader Election in Distributed Systems! 🚀 In this video, we'll explore the critical process of leader election, a fundamental concept in distributed computing. You'll learn: - What is Leader Election? - Practical Examples and Use Cases in real-world applications. - How Apache ZooKeeper Handles Leader Election. 📺 Don't forget to like, subscribe, and hit the bell ico...
Distributed Locking and Synchronisation
Переглядів 2,5 тис.Рік тому
In this video, we dive deep into the world of "Distributed Locking and Synchronization". 💡 What You'll Learn: - The crucial role of distributed locking and synchronization in maintaining data integrity and preventing chaos. - A breakdown of the core principles of locking and synchronization in computer science. - Explore real-world examples of how these techniques are used to manage shared reso...
Learning about Lamport clocks
Переглядів 157Рік тому
In this video, Baljeet, who recently deep-dived a bit into digital clocks, talks about Lamport Clock. Lamport clock is one of the basic Digital clocks which uses the Happened-Before relationship to increment the timestamps maintained by the processes. We can use Lamport clocks for the basic ordering of events across a distributed system. Today majority of the systems use an advanced version of ...
Mastering Length Extension Attacks: Exploiting Hash Functions
Переглядів 2,5 тис.Рік тому
In this video, we delve into the captivating realm of cryptographic vulnerabilities, focusing specifically on the Length Extension Attack. If your system relies on block-based hash algorithms like SHA-1 and MD5 for message authentication between two entities, it might be at a higher risk than you've anticipated. Join Vivek as he unravels the mechanics of the Length Extension Attack, shedding li...
How Merkle Trees Ensure Data Integrity - Engineering Bakar
Переглядів 356Рік тому
Join us in this video as we delve into and discuss the concept of Merkle Trees. Merkle Trees are fundamental to blockchain technology, distributed systems, and data verification processes. Through our discussion, we aim to explain the concept and make it more comprehensible for everyone interested in data integrity and cryptographic hash functions. The notes link: link.excalidraw.com/l/2wOAEgaj...
Understanding Bloom Filters: Space-Efficient Data Structures - Engineering Bakar
Переглядів 400Рік тому
Join us in this video as we delve into and discuss the fascinating concept of Bloom Filters. We begin by examining a document that lucidly and succinctly explains how Bloom Filters function and how they can be implemented effectively. brilliant.org/wiki/bloom-filter Bloom Filters are a fundamental component in databases, cache memories, and distributed data stores like Cassandra. Through our di...
Consistent hashing - Engineering Bakar
Переглядів 255Рік тому
Join us in this video as we explore and discuss the concept of Consistent Hashing. We start by examining a document that clearly and concisely explains how consistent hashing works and how it can be implemented effectively. web.stanford.edu/class/cs168/l/l1.pdf Consistent hashing is a key component in Content Delivery Networks (CDNs) and distributed data stores like Dynamo. Through our discussi...
Twitter System Design - Engineering Bakar
Переглядів 429Рік тому
In our first video, we will deep dive into the design of Twitter. Hope you enjoy it. Welcome to Engineering Bakar, where we're all about pulling back the curtain on tech and software engineering. We're a bunch of tech lovers who can't get enough of diving into how things work, from software design to the latest tech trends. We dig into cool stuff like how big platforms are built and run. Our ch...

КОМЕНТАРІ

  • @HariKrishnan-ff4hf
    @HariKrishnan-ff4hf Місяць тому

    Redis is more AP system ,short lock expiry duration,less overhead because of inmemory design and Zookeeper is more CP system, long lock expiry,more overhead because of consensus design

    • @Engineering-Bakar
      @Engineering-Bakar Місяць тому

      Lock expiry in with application in both cases. Zookeeper, etcd are designed for usecases where the amount of data is limited and they perform really well for small datasets. in general CP systems provides gurantees required for distributed locking where strictly one leader is required.

  • @HariKrishnan-ff4hf
    @HariKrishnan-ff4hf 2 місяці тому

    very well explained.

  • @fdddd2023
    @fdddd2023 5 місяців тому

    14:07 what if we don't use redis in HM mode - then we use sync replication so we getting strong consistency and don't need zookeeper, no?

    • @Engineering-Bakar
      @Engineering-Bakar 5 місяців тому

      Redis doesn't have native support for sync replication. An application will need to use WAIT command to check replication status additionally that still doesnt make redis as CP system as master node has accepted the write and failover can still cause data loss

  • @AnkithReddyAdudodla
    @AnkithReddyAdudodla 9 місяців тому

    OMG, that's great explanation!

  • @SecureInspect
    @SecureInspect Рік тому

    Great Content

  • @raghavgupta1998
    @raghavgupta1998 Рік тому

    Superb. The topics picked are very different. Great work. Thanks for this.

  • @raghavgupta1998
    @raghavgupta1998 Рік тому

    Thanks a lot sunnt for this session. It was nice as an overview for me. But i could not understand the security part, how is it gettinf afftected by h3_h4.

  • @raghavgupta1998
    @raghavgupta1998 Рік тому

    Could not understand. Need more context and detailed explanation.

  • @raghavgupta1998
    @raghavgupta1998 Рік тому

    Great session Sunny. Its our pleasre to be taught by you.

  • @akshshetty3698
    @akshshetty3698 Рік тому

    Excellent session! Thanks for keeping it simple, precise and easy to understand. Looking forward for more such topics.Keen to know your technical background too, hope you will respond with an intro clip in the channel.Keep up the good work👏

  • @ashishkumar-lm6cv
    @ashishkumar-lm6cv Рік тому

    Great Explanation 👏🏻. One doubt that I have is How zookeeper achieves consistency even being distributed. Can you elaborate more on zookeeper internals?

    • @Engineering-Bakar
      @Engineering-Bakar Рік тому

      Zookeeper provides sequential consistency and atomicity, we will do a deep dive session on zookeeper internals which will cover this, stay tuned.

  • @sumittanej
    @sumittanej Рік тому

    A good Bakar show! no doubt this is indeed a good discussion however this becomes too overwhelming deep diving on entity diagram if you guys could identify your target audience and later come little bit prepared and make it minimalistic or at high level design. you could also do few videos focused more at component, services level, talking more on system scalability, concurrent users, reliability and resiliency this would add more value. Cheer! Hope to see more good ones! ideas 1. eCommerce 2. tiny URl 3. money sharing app 4. prepaid card payment ecosystem for mall 5. Uber or any app 6. Google app