In his technology predictions for 2022, Werner Vogels, VP and CTO of Amazon, reminded the industry that sustainability is no longer a secondary consideration — nor something someone else is responsible for. “As developers, we are trained to think about how to optimize our architectures for factors like security, performance, reliability, and cost,” he wrote. “In 2022, you can add sustainability to that list. What we will begin to see in the coming years is developers taking an active role in building sustainability-conscious architectures that take into account not just the problems they are solving, but the planet as well.”
Vogels’s vision makes it clear that there is a difference between how we measure the efficiency of databases and, perhaps, how we should.
In this article, we’ll look at several steps organizations can take to reduce the carbon footprint of their databases. It all starts with the cloud.
Move off premises
Migrating workloads to a public cloud creates an instant return on sustainability efforts, with an average 84% immediate reduction in carbon emissions as a result. This reduction results from a combination of factors, including the fact that cloud data centers use less physical space and resources than those on-premises; the use of more efficient servers and higher server usage; and the fact that public cloud providers are themselves committed to sustainable infrastructure and a reduced carbon footprint, which enhances and supports their customers’ efforts as well.
For example, the three major public cloud providers have made the following commitments:
Amazon Web Services (AWS) is committed to being powered with 100% renewable energy by 2025 and has pledged to be carbon neutral by 2040 across all Amazon properties.
Microsoft Azure has also pledged to be completely powered by renewable energy by 2025 and is committed to being water-positive and zero-waste by 2030.
Google Cloud already boasts the use of 100% renewable energy in all regions, with a goal to use only carbon-free energy by 2030.
Businesses can amplify that carbon reduction by selecting a database platform that aligns with their energy goals. MongoDB is currently in 80-plus regions, with 12 of those regions being powered by renewable energy as well as being low-carbon and/or zero-waste. Furthermore, MongoDB Atlas Serverless instances are optimized to use the underlying infrastructure — RAM, CPU, disks, network bandwidth — to its maximum before consuming additional cloud resources. Currently, 60% of organizations globally have yet to adopt any serverless technology; however, pushing to serverless not only helps to reduce carbon footprint but also can provide a competitive advantage.
Consolidate your platform
Databases require management on many fronts, from transactional data and relational data to search, analytics, graph use cases, time series, and more. Each of these facets is frequently handled in its own separate database, necessitating management, monitoring, and resources. Not only is this a technical pain point, but it can also affect the environmental goals of an organization and its carbon footprint overall.
To start, fewer workloads result in a lower carbon emission total. Thus, simply paring down the number of databases used is an easy way to reduce the consumption of resources and associated carbon emissions. MongoDB Atlas brings all of those requirements into a single, consolidated database platform that solves for disparate and siloed databases and reduces resources and consumption.
Strategize around prevention
The greenest energy is the energy that is never used in the first place. Toward that end, organizations can work to shift their architectural strategy to be more mindful of carbon and energy considerations, and adjustments can be made to reduce usage without affecting performance.
For example, about 40% of instances are at least one size larger than needed. Businesses can reduce such excess consumption by avoiding over-provisioned databases. This change can be done easily through automatic scaling with MongoDB Atlas cluster and storage auto-scaling, as well as with Atlas Triggers, which allows users to schedule scaling events.
Another consideration is the consumption of non-production environments. Currently, more than 44% of total compute resources are being used in non-production environments like dev, testing, staging, and pre-production. This consumption typically does not occur outside of work hours, and it’s easy to pause these resources outside of peak hours in Atlas, either on-demand or with an Atlas Trigger.
Continue to iterate
These major adjustments to the structure of your database architecture represent impactful steps toward creating a sustainable database, but there are many, more granular ways to improve database technology and better align with environmental goals. Implementing smaller changes will chip away at carbon emissions and create significant impact over time.
Such easy-to-execute suggestions include the following:
Consider the impact of data transfer over longer distances and find new regions to house data.
Determine the proper location of data and leverage the flexibility to deploy in any region, which grants more control over replication of data.
Use Atlas Global Clusters to ensure that data stays within a defined geography, thereby minimizing the infrastructure required to deploy a globally distributed application.
Employ Atlas Data Federation to minimize the need to replicate data between cloud object storage and Atlas by allowing users to query both sources from a single interface.
Use analytic nodes that are deployed in more environmentally sustainable regions.
Leverage Atlas’s many dashboards and alerts to be proactively notified when system resources are rising beyond acceptable thresholds.
Introduce an “eco mode” in your apps, giving users the ability to agree to potentially lower response times in exchange for a positive environmental impact.
Avoid peak times for predictable workloads; schedule these during a time that will not drive over-consumption.
Use data tiering; users can automate data lifecycles using Atlas Online Archive and reduce exhaustion of resources.
The time to act is now. Studies estimate that between 2% and 4% of global emissions come from the information and communications technology industry — about the same as from aviation. Much of that impact is concentrated in data centers, which consume 10 times to 50 times the energy per square foot compared to a typical office building.
With every day that businesses do not improve their carbon footprint, the environmental problem gets worse. For example, IDC found that the number of data centers worldwide has grown from 500,000 in 2012 to more than 8 million today. And data itself is exploding, with 2.5 quintillion new bytes of data created daily. Because of this, the amount of energy used by data centers will continue to double every four years, meaning data centers have the fastest-growing carbon footprint of any area within the technology industry sector.
Organizations need a proper database framework to combat these negative environmental trends without creating additional and unnecessary carbon emissions that add to the global warming crisis. As a developer data platform, MongoDB provides businesses with the database architecture to reduce consumption of resources, and MongoDB Atlas offers additional features to enhance carbon reduction efforts.
Businesses are a strong lever for change in the world. When it comes to carbon reduction, MongoDB is committed to creating sustainable and lean data architectures, both internally and for customers.
Check out these three stories of companies that have taken incredible strides to curtail their carbon impact. Business leaders can all be inspired by and learn from each other to make changes and collectively benefit the Earth and future generations.
Learn about this and more from MongoDB Solutions Architect, Snehal Bhatia’s, presentation on Designing a Sustainable Architecture from MongoDB World 2022..