IN-MEMORY DATA STRUCTURE STORE
PROJECT NO - 61
TEAM MEMBERS -
1. NAVEEN PAUDEL - 211B192
2. NIKHIL SAHU - 211B195
3. PANSHUL KHURCHWAL - 211B202
Project Overview
What is it about? - A high performance in-memory data structure store used for optimizing data
retrieval speed for web applications .
Purpose - This project is designed to provide efficient, in-memory data store for applications requiring
fast access to data.
Key Features - Persistence options, Performance optimization, Replication, Scalable, LRU Eviction
Technology Stack used - JS, NodeJS, Sockets, Custom In-memory Data Structures
What is the problem ?
1. Latency and Speed Issues in Traditional Databases .
2. Handling High Throughput for Read/Write Operations .
3. Complex Data Manipulation in Real-Time Applications .
How do we solve it?
1. Storing data entirely in-memory, offering extremely low latency.
2. The architecture of the application will allow it to handle hundreds of thousands of requests per
second with minimal delay.
3. It supports complex atomic operations on data structures, enabling real-time manipulation of
data.
So, What is the trade-off?
1. It uses all its RAM to store the data, which ultimately means that RAM size needs to be large
and its expensive then disk storage.
2. It is in-memory so total dataset size is constrained by the available RAM on the server.
3. It is primarily single-threaded for simplicity and performance optimization which can be a
limitation for multi-core CPUs.
About Key Features
Persistence - Ability of a system to store data in a way that ensures it remains available and intact,
even after the system is restarted. It is a process of saving data to non-volatile storage .
Performance Optimization - I/O Multiplexing, Use of Single Threading, Expiry Keys, In-Memory
Storage, Pipelining, Pub/Sub Messaging, Clustering & Sharding .
Replication - It allows data from one server(master) to be copied to one or more servers(slaves).
This is essential for data redundancy, high availability and scalability.
Traditional Architecture
Modern Architecture(Single Layer)
Modern Architecture(Multi Layer)
Use Cases & Applications in the Industry
1. Store and manage user shopping carts for quick access during checkout.
2. Analyze and display live financial data, such as stock prices and trading volumes.
3. Maintain real time leaderboards for gaming, allowing players to see their rankings instantly.
4. Manage user sessions in web applications, enabling fast retrieval of session data and enhancing
user experience.
Timeline
Implementing a client and server
Building up replication
side communication channel .
procedure.
Building up persistence Building up streams and
procedure . transactions
THANK YOU !!!!!
