Behrad Babaee

Leveraging Moore’s Law: Optimising Database Performance

Your database is wasting modern RAM as a simple cache. Learn how a new architecture delivers predictable microsecond latency and slashes server costs.

Leveraging Moore’s Law: Optimising Database Performance
#1about 4 minutes

The evolution of Moore's Law and its impact on software

Moore's Law drove CPU speed increases until 2005, after which the industry shifted focus from single-core performance to multi-core scalability.

#2about 2 minutes

Comparing server hardware from 2006 to 2024

Modern servers have vastly more RAM and significantly faster storage compared to 2006, fundamentally changing the ratio of memory to disk.

#3about 7 minutes

Traditional database architecture and its reliance on caching

Databases designed for limited RAM now use the extra memory in modern hardware as a cache, which sits on top of the original disk-based architecture.

#4about 6 minutes

The problems and unpredictability of database caching

Caching leads to inconsistent performance across environments and fails to improve overall application latency when multiple parallel queries are involved.

#5about 4 minutes

An alternative architecture with the index in RAM

A modern database design can leverage abundant RAM to hold the entire index in memory, enabling direct, fast access to data on SSDs without a cache.

#6about 4 minutes

Achieving speed and efficiency without caching

By using fewer resources like CPU cycles and disk I/O, an index-in-RAM architecture provides consistently fast performance and reduces infrastructure costs.

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

Matching moments

The architectural advantage of a SQL-native design
07:52 MIN

The architectural advantage of a SQL-native design

Fault Tolerance and Consistency at Scale: Harnessing the Power of Distributed SQL Databases

Answering questions on Cube's architecture and use cases
34:43 MIN

Answering questions on Cube's architecture and use cases

Making Data Warehouses fast. A developer's story.

The critical need for performance in modern applications
04:52 MIN

The critical need for performance in modern applications

In-Memory Computing - The Big Picture

The architecture and limitations of in-memory databases
20:28 MIN

The architecture and limitations of in-memory databases

In-Memory Computing - The Big Picture

Why GPU acceleration surpasses traditional CPU performance
02:12 MIN

Why GPU acceleration surpasses traditional CPU performance

Accelerating Python on GPUs

Using distributed caches to reduce database load
15:25 MIN

Using distributed caches to reduce database load

In-Memory Computing - The Big Picture

Understanding the fundamental speed of in-memory operations
07:27 MIN

Understanding the fundamental speed of in-memory operations

In-Memory Computing - The Big Picture

Optimizing cache efficiency with a dedicated sharded layer
12:55 MIN

Optimizing cache efficiency with a dedicated sharded layer

Scaling: from 0 to 20 million users

Featured Partners

Related Videos

See all videos
In-Memory Computing - The Big Picture30:36

In-Memory Computing - The Big Picture

Markus Kett

Database Magic behind 40 Million operations/s24:22

Database Magic behind 40 Million operations/s

Jürgen Pilz

Leveraging Real time data in FSIs41:51

Leveraging Real time data in FSIs

Tim Faulkes

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

Single Server, Global Reach: Running a Worldwide Marketplace on Bare Metal in a Cloud-Dominated World50:06

Single Server, Global Reach: Running a Worldwide Marketplace on Bare Metal in a Cloud-Dominated World

Jens Happe

Scaling Databases21:09

Scaling Databases

Tobias Petry

Things I learned while writing high-performance JavaScript applications27:27

Things I learned while writing high-performance JavaScript applications

Michele Riva

Scaling: from 0 to 20 million users28:12

Scaling: from 0 to 20 million users

Josip Stuhli

From learning to earning

Jobs that call for the skills explored in this talk.