The UKI Google Cloud Next 22 event at Wembly Studium London was fantastic.
John Abel, Technical Director, Office of the CTO a Google Cloud great speaker, makes it unique in explaining Sustainable Development Principles for developers and Engineers.
He shifted the developer and engineer's daily tech practise from thinking about how they can help save the world to practising Google Cloud sustainable development principles.
During his presentation, he emphasised Google Cloud Sustainable Development feathers and developer cut off their additional device uses:
John primarily focuses on the following points.
- What is Sustainable Development Principles?
- Software Developer & Engineer Consideration
- Sustainable Development Lifecycle
- Conclusion Google Cloud
Google Cloud focus on Carbon Free Energy (CFE)
Use less energy
Take advantage of Al-optimized, efficient data centers and better utilization.
Google data centers are twice as energy efficient as a typical enterprise data center and compared to 5 years ago Google Cloud deliver 7x the compute power for the same amount of electricity.
Zero net operational emissions
Meet your public sustainability commitments and reduce the environmental impact of IT operations.
Google has neutralized all legacy emissions since 1998 and matched 100% of electricity use with renewable energy since 2017, including the electricity that powers your workloads.
Carbon-free energy by 2030
First company to set the goal to completely decarbonize electricity By 2030, Google Cloud aim to run on 24/7 carbon-free energy at all their data centers, around the clock.
Sustainable Development Principles
- Reducing Carbon Emissions
- Use Techniques and Tools to identify positive impact
- Aligned to development lifecycle Google Cloud
Developer Ergonomics
Everyone is aware that we use additional computer monitors or computers in our offices and homes on a daily basis. It is sometimes unnecessary. Even in most offices, the computer runs continuously, which is completely unnecessary. He explains how we can stop carbon emissions and summarises the following points.
- Laptop 20-50w vs desktop 80-150w
- Use it or snooze it
- Wide Screen vs Multi-Screen all down to CO2 emissions (lifetime CO2 can exceed 1000 kg of CO2 Emissions)
- Know your data, embodied carbon can be up to 80% of the lifetime carbon emissions
- Communication style balance chat, email and video or just speak in person
- Use Cloud Workstation to pick low CO2 region for developing Google Cloud
Plan Sustainably
• Check CFE, by regional carbon
https://cloud.google.com/sustainability/region-carbon
Restricting Resource Locations
Carbon free energy for Google Cloud regions
• Google Cloud Best Practice Reduce your Google Cloud carbon footprint
https://cloud.google.com/architecture/reduce-carbon-footprint
Design Sustainably
- Set sustainability budget, similar to SRE error budget
- Responsible Al Practice :
- Sparse model vs dense model
- Use Optimized Processors for ML like TPU
- Use Cloud Computing energy costs reduction by a factor 1.4-2
- Pick the cleanest energy regions reducing CO2 emission by factor 5-10
Building responsible Al for everyone
Google Cloud are working to develop artificial Inteligence responsibly in order to benefit people and society: Check the Al practices Google Cloud - https://oi.google/responsibilities/responsible
Develop Sustainably
- Consider Data Mesh/Data ASA Product to reduce copies of data
- Reduce custom code by upto 70% with Vertex AL Vertex Al
- Reduce idle resources by considering Cloud Run
- Build delay and scala maching managed M.Tools for any u se else Cloud Run
- Code execution style using tools (e.g. Pyflakes, trace.py or Pylint)
- Bad and placed ovaalne including a Phort Java NET duy on managed
- Balancing the use of lower level languages like C vs interpreter Languages like Python
Extra Credit:
Check the following C2C community member's post :
