You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This article describes how NTT DATA used the SCI score to help measure the carbon footprint of a large scale three-tier on premise software system
Description of problem
There are a number of large scale multi tier, on-premise software systems being developed and maintained by NTT DATA that run enterprise scale applications such as ecommerce. While there is a desire by all enterprises to migrate to the cloud at some point in the future, there will be a considerable portfolio of on-premise solutions that we need to live with for the next few years. It is important that the carbon footprint of these applications are measured and optimised, especially as they continue to scale.
How the use case solves the problem
NTT DATA has a mature set of tools and techniques to optimise performance of such software systems. We use industry standard benchmarks such as TPC-W which provide a benchmark for performance in terms of Web Interactions per second for a given load. We have implemented tooling across the entire stack including networking, hardware and software to measure the rate of emissions using SCI. We have measured it in term of co2 equivalent per web interaction
Main benefits of the solution
This has helped us to gather carbon emissions related metrics , utilising some of the mature processes that we regularly perform such as the performance optimisation. This in turn helps us understand the impact of scaling the systems on the total carbon footprint and take appropriate actions to help bring down the carbon footprint
What was the outcome, how were carbon emissions reduced
We now have a baseline for the carbon impact per unit of work , based on a blueprint infrastructure, that provides a baseline performance. We will now use this to implement patterns that decrease the carbon footprint with no negative impact on performance and also to be aware of the carbon impact on any performance improvement, scaling decision that we make on the software system
The text was updated successfully, but these errors were encountered:
@Henry-WattTime @GadhuNTTDATA
Love the idea, I can get the admin process started on my end and get the article into the pipeline. I am happy to help with editing or anything else you need help with.
Executive Summary
This article describes how NTT DATA used the SCI score to help measure the carbon footprint of a large scale three-tier on premise software system
Description of problem
There are a number of large scale multi tier, on-premise software systems being developed and maintained by NTT DATA that run enterprise scale applications such as ecommerce. While there is a desire by all enterprises to migrate to the cloud at some point in the future, there will be a considerable portfolio of on-premise solutions that we need to live with for the next few years. It is important that the carbon footprint of these applications are measured and optimised, especially as they continue to scale.
How the use case solves the problem
NTT DATA has a mature set of tools and techniques to optimise performance of such software systems. We use industry standard benchmarks such as TPC-W which provide a benchmark for performance in terms of Web Interactions per second for a given load. We have implemented tooling across the entire stack including networking, hardware and software to measure the rate of emissions using SCI. We have measured it in term of co2 equivalent per web interaction
Main benefits of the solution
This has helped us to gather carbon emissions related metrics , utilising some of the mature processes that we regularly perform such as the performance optimisation. This in turn helps us understand the impact of scaling the systems on the total carbon footprint and take appropriate actions to help bring down the carbon footprint
What was the outcome, how were carbon emissions reduced
We now have a baseline for the carbon impact per unit of work , based on a blueprint infrastructure, that provides a baseline performance. We will now use this to implement patterns that decrease the carbon footprint with no negative impact on performance and also to be aware of the carbon impact on any performance improvement, scaling decision that we make on the software system
The text was updated successfully, but these errors were encountered: