WEBVTT

1
00:00:03.850 --> 00:00:09.941
Infrastructure is bought to run software, and the amount is set by what it demands at its worst moment.

2
00:00:09.941 --> 00:00:13.468
When code is inefficient, the response is rarely to fix it.

3
00:00:13.468 --> 00:00:20.201
It is to add nodes or raise the instance size, and the cost lands on the cloud bill every month after.

4
00:00:20.201 --> 00:00:29.178
Servers in data centres are active a fraction of the time and idle for the rest, and as hardware efficiency gains plateau, growing software complexity cancels them out.

5
00:00:29.178 --> 00:00:36.872
For a CTO the case is financial before it is environmental: reduce the demand, and the infrastructure, its cost and its emissions shrink together.

6
00:00:38.472 --> 00:00:40.903
GreenCode approaches infrastructure from the software side.

7
00:00:40.903 --> 00:00:56.530
Rather than tuning a data centre around fixed demand, it reduces the workload and then right-sizes what runs it, in a loop: measure under load, attribute energy to specific code and components, optimise, and measure again until the improvement stops or a target is met.

8
00:00:58.130 --> 00:01:00.638
The infrastructure part starts with discovery.

9
00:01:00.638 --> 00:01:07.746
GreenCode reconstructs a running estate and its dependencies, at the scale of tens of thousands of services.

10
00:01:07.746 --> 00:01:21.960
That map is used to find three kinds of waste: energy hotspots, services that are partly or wholly unused and could be shut down or moved to serverless functions, and unnecessary communication between components.

11
00:01:21.960 --> 00:01:33.666
The application is then deployed to a reconstructed, instrumented version of its environment and stress-tested, so that code and infrastructure changes are benchmarked in the same controlled setting.

12
00:01:35.266 --> 00:02:14.541
The components sit in the infrastructure work: tools that mine a running deployment and reconstruct a representative environment through infrastructure as code; a toolkit that monitors the energy of the infrastructure and maps it back to application metrics; energy-aware resource management that scales services in line with demand and grid carbon intensity; a recommender that proposes infrastructure changes without loss of performance; predictive models of resource use for database services, a significant hotspot in most estates; and management of jobs across idle or redundant devices, so existing machines are used before new ones are bought.

13
00:02:16.141 --> 00:02:29.232
Two partner use cases illustrate it: a research computing cluster with resources often idle and no way to see where its energy goes, and a cloud-hosted system moving to a cloud-native architecture, measured on the provider's bill per customer.

14
00:02:29.232 --> 00:02:41.652
The outcome is an estate that is smaller, better understood and easier to rebuild: capacity bought against measured demand rather than a safety margin, unused services switched off, and the infrastructure definition reviewed and versioned like code.

15
00:02:41.652 --> 00:02:44.673
These tools are being validated on partner systems now.

16
00:02:44.673 --> 00:02:54.408
Reach out to us with the estate you suspect is bigger than it needs to be, and we will help you understand how much smaller GreenCode can make it.
