Markus Kett

In-Memory Computing - The Big Picture

What if your database is the real bottleneck? See how a database-less architecture can be 1000x faster and cut cloud costs by up to 99%.

In-Memory Computing - The Big Picture
#1about 3 minutes

The critical need for performance in modern applications

Latency is a significant cost for businesses, making high-performance, in-memory computing essential for modern applications.

#2about 2 minutes

Understanding the fundamental speed of in-memory operations

In-memory operations are orders of magnitude faster, measured in microseconds, compared to database access which is measured in milliseconds.

#3about 3 minutes

The core problem of object-relational impedance mismatch

Object-oriented programming languages are inherently incompatible with relational database models, leading to complex and slow data mapping.

#4about 3 minutes

Why NoSQL and mapping layers don't solve the bottleneck

Even with NoSQL databases, the need for data conversion and mapping layers like ORMs persists, creating a significant performance bottleneck.

#5about 3 minutes

Using distributed caches to reduce database load

A distributed cache cluster sits between the application and the database to store frequently accessed data in memory, reducing database load.

#6about 2 minutes

Differentiating in-memory data grids from distributed caches

In-memory data grids extend distributed caches by adding computational capabilities, allowing for distributed processing across the cluster.

#7about 3 minutes

The architecture and limitations of in-memory databases

In-memory databases run the DBMS in memory but often on a separate cluster, which still introduces network latency and requires data mapping.

#8about 4 minutes

A new paradigm: Database-less processing and system prevalence

The system prevalence architecture keeps the entire application state as an object graph in memory, leveraging native language APIs for ultra-fast queries.

#9about 3 minutes

Simplifying architecture and costs with Eclipse Store

Eclipse Store provides a persistence engine that stores the in-memory object graph directly to cloud blob storage, eliminating database clusters and reducing costs.

Related jobs
Jobs that call for the skills explored in this talk.

Featured Partners

Related Videos

Build ultra-fast In-Memory Database Apps and Microservices with Java 44:05

Build ultra-fast In-Memory Database Apps and Microservices with Java

Markus Kett

Database Magic behind 40 Million operations/s24:22

Database Magic behind 40 Million operations/s

Jürgen Pilz

Building Real-Time AI/ML Agents with Distributed Data using Apache Cassandra and Astra DB28:20

Building Real-Time AI/ML Agents with Distributed Data using Apache Cassandra and Astra DB

Dieter Flick

How building an industry DBMS differs from building a research one30:34

How building an industry DBMS differs from building a research one

Markus Dreseler

Databaseless Data Processing - High-Performance for Cloud-Native Apps and AI30:21

Databaseless Data Processing - High-Performance for Cloud-Native Apps and AI

Markus Kett

Scaling: from 0 to 20 million users28:12

Scaling: from 0 to 20 million users

Josip Stuhli

Swapping Low Latency Data Storage Under High Load22:32

Swapping Low Latency Data Storage Under High Load

George Asafev

Advanced Caching Patterns used by 2000 microservices44:37

Advanced Caching Patterns used by 2000 microservices

Natan Silnitsky

From learning to earning

Jobs that call for the skills explored in this talk.

Cloud Data Spaces


Municipality of Bilbao, Spain

€40-60K
Azure
Python
Node.js
Amazon Web Services (AWS)