Category: Data Center Energy
- Carbon footprint – laptops vs servers, Intel vs ARM ()
The carbon footprint of a laptop is mostly in the manufacturing. Servers are different, and ARM chips are starting to change the game.
- Sustainable computing – where to focus? ()
Progress in renewable energy has pushed IT carbon footprints down, revealing how much more work is needed in manufacturing.
- Zoom, video conferencing, energy, and emissions ()
It is obvious that Zoom is better for the environment than flying but what about vs driving an EV or public transport? The problem is we can’t be sure.
- Fighting over who has the greenest public cloud ()
Public cloud price wars have ended, and a new battle over who is the greenest has begun.
- Barriers for startups tackling data center energy consumption ()
There is scope to start something tackling the problem of data center energy consumption. However, there are significant challenges: how to measure the problem, the uncertainty around the current situation, and the limited number of growth buyers.
- Assessing the suitability of the Greenhouse Gas Protocol for calculation of emissions from public cloud computing workloads ()
A copy of my paper published in the open-access Journal of Cloud Computing.
- Hiding greenhouse gas emissions in the cloud ()
The scientific journal, Nature Climate Change, has published a paper I’ve written about how moving to the cloud can result in greenhouse gas emissions being hidden.
- How much water do data centers use? ()
US data centers are expected to use 660 billion litres of water in 2020. In 2018, Google consumed 15.79bn litres and Microsoft 3.61bn litres, primarily for their data centers. What is this used for and how does it impact the environment?
- Energy efficiency, edge compute and serverless ()
We want to encourage more efficient use of cloud resources but designing incentives is hard. That changes with serverless and edge compute. Now energy usage is linked to price.
- Carbon emissions aware cloud compute ()
The first example of carbon aware compute is uniquely suited to hyperscale cloud environments. It becomes even more interesting with carbon aware load balancing and serverless.