Yannick Tremblais is an IT innovation manager with Groupe Rocher. Their objective is to reconnect people to nature. The four pillars of their mission are Green IT, Eco-design of IS, Responsible IT usage, and IT for Green.
Julien Brun is head of COE API and EDA at L’Oreal. They have a program called L’Oreal for the future, which works towards sustainability. They focus on Sustainable Tech by design, Tech for sustainability, and inclusive Tech. This article by Yannick and Julien discusses sustainable API Green Score.
There is no standard in the market to measure the impact of APIs or related to API eco-design. Environmental impact is a shared concern, so we have decided to share our work with the community. These guidelines and rules of calculation should be shared and validated by many to be recognized as a standout.
API Thinking
It is a French collective created in 2020. It is dedicated to APIs and is more for the banking and insurance sector. It is a free collective. It has a large network of more than 80 different organizations with a lot of diversity. It is a group working on sustainable API and is not linked to any enterprise or commercial suite.
The group has 5 to 8 team members working to define goals to evaluate APIs.
API Green Score is a toolkit to help API users, designers, and owners to ask themselves questions about the digital impact of their API. API Green Score is a toolkit to help API users, designers, and owners to ask themselves questions about the digital impact of their API. This tool is based on seven domains to create relevant and realistic metrics for stakeholders. This toolkit considers eco-design and eco-consumption of APIs. It evaluates the API on a scale of A to E, with A being “Excellent” and E being “Very Poor.” The evaluation method and results are shared with all concerned stakeholders.
The Seven Domains
- API lifecycle – It is important to understand the references and details of what the APIs do. Without this, it is difficult to evaluate and optimize the APIs. Some points to be considered here are
-
- Decommission unused APIs
-
- Deploy APIs near the consumer.
-
- Reduce the number of API versions.
-
- Unify API catalog
-
- Create consumer referential
-
- Identify APIs for single usage.
- Data Exchange – The objective here is to reduce the impact of the action messages. If you have no tools to evaluate the payload size, it’s very difficult to improve the data exchange.
- Data – Here, we are concerned about how data is collected. Queries need to be optimized to get relevant data; a cache can be used to reduce API calls.
- Architecture – Promote event-driven architecture. It may not always be possible to have event-driven architecture but try to use Async API, Webhook, or pagination.
- Tools – You can use tools like Google. You can evaluate the energy consumption of an API and define a basis for rating.
- Infrastructure – Use adaptive infrastructure. Use as few cloud suppliers as possible between consumer and supplier. Be near a data center. Study the energy impact of each language, e.g., Python, C++, etc.
- Communication – Share your criteria and findings with peers and as many people as possible. The more you share, the more knowledge you gain, and your results improve.
API Green Score creates a card for each domain listed above. It has a description, governance details, KPIs per API, Tools to measure, Impact EcoScore, and an example.
The API Green Score Grid has a list of rules categorized into different categories based on the domains. It talks about what the rules mean for the API, how the APIs follow the rules, and the impact of the rule. APIs are then rated against the rules, and you get your green score. It also tells you ways to improve your green score.
