logo
ProjectCompleted And Deployed

Amazon Web Services Randomized Network Graph (RNG) Data Center Architecture

One of 10,000+ projects Taiyo tracks in United States.

Source: Crypto Briefing

Amazon Web Services (AWS) has replaced the traditional fat-tree network architecture in its data centers with a new design called Randomized Network Graph (RNG). This architecture uses quasi-random graphs to create a flat network topology, allowing data to take multiple paths between points, improving efficiency and reducing costs. The design includes a custom routing protocol called Spraypoint an

Status
Prospective
Sector
Power Generation (CCGT), Data Center
Location
United States of America
Updated
22 Jul 2026
Source
Crypto Briefing

ConstructChat

Ask ConstructChat about Amazon Web Services Randomized Network Graph (RNG) Data Center Architecture

Pick a question to edit it, or write your own.

First answer free · no account

I am a

Project Information FAQ

Project Information

3 Q
The project “Amazon Web Services Randomized Network Graph (RNG) Data Center Architecture” is an infrastructure initiative in the Power Generation (CCGT) sector, located in United States. Taiyo aggregates data on it from Crypto Briefing.

Participants

Not available for this record.

Status

Original status

completed and deployed

Taiyo status

Prospective

Taiyo last update

2026-07-22

Contact

Not available for this record.

Description

Description

Amazon Web Services (AWS) has replaced the traditional fat-tree network architecture in its data centers with a new design called Randomized Network Graph (RNG). This architecture uses quasi-random graphs to create a flat network topology, allowing data to take multiple paths between points, improving efficiency and reducing costs. The design includes a custom routing protocol called Spraypoint and passive optical devices called ShuffleBoxes, which reduce power consumption. The RNG architecture is now the default for the majority of AWS workloads, resulting in cost savings between 9% and 45% and improved energy efficiency, including a 40% reduction in energy use and a 33% boost in data center efficiency.

Original sector

Technology

Location

Country

United States of America (USA)

Source

Source

Crypto Briefing

URL

More Details

Record type

Project

Sector

Power Generation (CCGT), Data Center

Article published date

2026-05-31

More on this project

Research

Free account

View all details

You are reading the public part. A free account opens the rest:

  • Every detail on this project
  • News and media coverage
  • Technical research
View all details

Free account · no card · lands you back on this project