🖥️
Sunil Notebook
Interview Preparation
  • 📒Notebook
    • What is this about ?
  • System Design
    • 💡Key Concepts
      • 🌐Scalability
      • 🌐Latency Vs Throughput
      • 🌐Databases
      • 🌐CAP Theorem
      • 🌐ACID Transactions
      • 🌐Rate limiting
      • 🌐API Design
      • 🌐Strong Vs eventual consistency
      • 🌐Distributed tracing
      • 🌐Synchronous Vs asynchronous Communication
      • 🌐Batch Processing Vs Stream Processing
      • 🌐Fault Tolerance
    • 💎Building Blocks
      • 🔹Message
      • 🔹Cache
      • 🔹Load Balancer Vs API Gateway
    • 🖥️Introduction to system design
    • ⏱️Step By Step Guide
    • ♨️Emerging Technologies in System Design
    • ☑️System design component checklist
      • 🔷Azure
      • 🔶AWS
      • ♦️Google Cloud
    • 🧊LinkedIn feed Design
    • 🏏Scalable Emoji Broadcasting System - Hotstar
    • 💲UPI Payment System Design
    • 📈Stock Broker System Design - Groww
    • 🧑‍🤝‍🧑Designing Instagram's Collaborative Content Creation - Close Friends Only
    • 🌳Vending Machines - Over the air Systems
    • Reference Links
  • DSA
    • Topics
      • Introduction
      • Algorithm analysis
        • Asymptotic Notation
        • Memory
      • Sorting
        • Selection Sort
        • Insertion Sort
        • Merge Sort
        • Quick Sort
        • Quick'3 Sort
        • Shell Sort
        • Shuffle sort
        • Heap Sort
        • Arrays.sort()
        • Key Points
        • Problems
          • Reorder Log files
      • Stacks and Queues
        • Stack Implementations
        • Queue Implementations
        • Priority Queues
        • Problems
          • Dijkstra's two-stack algorithm
      • Binary Search Tree
        • Left Leaning Red Black Tree
          • Java Implementations
        • 2-3 Tree
          • Search Operation - 2-3 Tree
          • Insert Operation - 2-3 Tree
        • Geometric Applications of BST
      • B-Tree
      • Graphs
        • Undirected Graphs
        • Directed Graphs
        • Topological Sort
      • Union Find
        • Dynamic Connectivity
        • Quick Find - Eager Approach
        • Quick Find - Lazy Approach
        • Defects
        • Weighted Quick Union
        • Quick Union + path comparison
        • Amortized Analysis
      • Convex Hull
      • Binary Heaps and Priority Queue
      • Hash Table vs Binary Search Trees
  • Concurrency and Multithreading
    • Introduction
    • Visibility Problem
    • Interview Questions
    • References
      • System design
  • Design Patterns
    • ℹ️Introduction
    • 💠Classification of patterns
    • 1️⃣Structural Design Patterns
      • Adapter Design Pattern
      • Bridge Design Pattern
      • Composite Design Pattern
      • Decorator Design Pattern
      • Facade Design Pattern
      • Flyweight Design Pattern
      • Private Class Data Design Pattern
      • Proxy Design Pattern
    • 2️⃣Behavioral Design Patterns
      • Chain Of Responsibility
      • Command Design Pattern
      • Interpreter Design Pattern
      • Iterator Design Pattern
      • Mediator Design Pattern
      • Memento Design Pattern
      • Null Object Design Pattern
      • Observer Design Pattern
      • State Design Pattern
      • Strategy Design Pattern
      • Template Design Pattern
    • 3️⃣Creational Design Patterns
      • Abstract Factory Design Pattern
      • Builder Design Pattern
      • Factory Method Design Pattern
      • Object Pool Design Pattern
      • Prototype Design Pattern
      • Singleton Design Pattern
    • Java Pass by Value or Pass by Reference
  • Designing Data-Intensive Applications - O'Reilly
    • Read Me
    • 1️⃣Reliable, Scalable, and Maintainable Applications
      • Reliability
      • Scalability
      • Maintainability
      • References
    • 2️⃣Data Models and Query Languages
      • Read me
      • References
    • Miscellaneous
  • Preparation Manual
    • Disclaimer
    • What is it all about?
    • About a bunch of links
    • Before you start preparing
    • Algorithms and Coding
    • Concurrency and Multithreading
    • Programming Language and Fundementals
    • Best Practices and Experience
  • Web Applications
    • Typescript Guidelines
  • Research Papers
    • Research Papers
      • Real-Time Data Infrastructure at Uber
      • Scaling Memcache at Facebook
  • Interview Questions
    • Important links for preparation
    • Google Interview Questions
      • L4
        • Phone Interview Questions
      • L3
        • Interview Questions
      • Phone Screen Questions
  • Miscellaneous
    • 90 Days Preparation Schedule
    • My Preparation for Tech Giants
    • Top Product Based Companies
  • Links
    • Github
    • LinkedIn
Powered by GitBook
On this page

Was this helpful?

Edit on GitHub
  1. Research Papers
  2. Research Papers

Scaling Memcache at Facebook

  • Enhancements to Memcached:

    • Lease Mechanism: To mitigate the "thundering herd" problem, a lease mechanism is used. When a cache miss occurs, a lease token is assigned to one server, preventing multiple servers from simultaneously querying the backend database.

    • Adaptive Slab Allocator: This helps in better memory management by reducing fragmentation and optimizing memory allocation for different sizes of data​.

  • Handling Consistency and Stale Data:

    • McSqueal and Mcrouter: These tools are used to ensure data consistency across different regions by broadcasting invalidations whenever the data is updated in the database​.

    • Replication and Invalidation: SQL statements modifying data include the required cache keys, ensuring that updates are propagated to all relevant caches​.

  • Scaling Strategies:

    • Regional Clusters: Splitting servers into regional clusters reduces latency and improves reliability by keeping data closer to users and mitigating the effects of large-scale outages​.

    • Different Pools for Different Data Types: Assigning different pools to different types of data based on access patterns and memory requirements prevents negative interactions and optimizes performance​.

  • Load Balancing and Failure Handling:

    • Gutter Servers: These are used as fallback when primary memcached servers fail, preventing cascading failures that can overload the system​​.

    • Distributed Read Requests: Read requests are distributed across multiple replicas to balance the load and improve performance​.

  • Performance Optimization:

    • Eviction Policies: Utilizing Least Recently Used (LRU) eviction policies within each slab class ensures efficient memory use by discarding the least used items first​.

    • High Availability: The design incorporates mechanisms to handle server failures gracefully, ensuring high availability and reliability of the caching system​​.

PreviousReal-Time Data Infrastructure at UberNextImportant links for preparation

Last updated 11 months ago

Was this helpful?

Ref:

https://www.usenix.org/conference/nsdi13/technical-sessions/presentation/nishtala
Page cover image