Dynatrace Cost & Carbon Optimization certified for accuracy and transparency

Globally, organizations are increasingly prioritizing sustainability. Board-level mandates and executive-level sponsorship have become the norm. Motivations include regulatory pressures, favorable investment potential, cost savings, competitive advantage, talent acquisition, and moral imperative. While a dynamic political climate can disrupt how these drivers are prioritized, the goal remains important, and momentum remains strong.

IT carbon footprint is top of mind for observability professionals in many industries. The explosion of AI models shines a new spotlight on the issue, with a recent study showing that using AI to generate an image takes as much energy as a full smartphone charge. For IT, a natural focus should be reducing carbon emissions from electricity consumption from on-premises, hybrid, and multicloud computing.

The challenge along the path

Well-understood within IT are the coarse reduction levers used to reduce emissions; shifting workloads to the cloud and choosing green energy sources are two prime examples. To continue down the carbon reduction path, IT leaders must drive carbon optimization initiatives into the hands of IT operations teams, arming them with the tools needed to support analytics and optimization.

Cloud service providers (CSPs) share carbon footprint data with their customers, but the focus of these tools is on reporting and trending, effectively targeting sustainability officers and business leaders. Unfortunately, they lack the detailed IT context practitioners need to identify carbon reduction opportunities and analyze optimization solutions. While stand-alone tools attempt to complement CSP solutions, they face similar challenges:

• Data granularity and accuracy: Tools struggle with providing detailed and accurate data on carbon emissions. This is partly due to the complexity of instrumenting and analyzing emissions across diverse cloud and on-premises infrastructures.
• Real-time monitoring: The periodic reports from cloud service providers lack real-time monitoring and actionable insights, limiting IT teams’ ability to make immediate adjustments to reduce carbon footprints.
• Integration with existing systems and processes: Integration with existing IT infrastructure, observability solutions, and workflows often requires significant investment and customization.
• Quantifying impact: Quantifying the effectiveness of various emission reduction methods can be challenging, making it difficult for IT leaders to prioritize actions and justify sustainability investments.

The Dynatrace carbon optimization solution

Introduced in 2023, Carbon Impact helps practitioners identify and evaluate meaningful opportunities to optimize their cloud and on-premises infrastructures. The Carbon Impact app directly supports our customers’ sustainability efforts through granular real-time emissions reporting and analytics, translating host utilization metrics into their CO2 equivalent (CO2e). By leveraging existing OneAgent® instrumentation, customers can get started in minutes with no new instrumentation hurdles.

Today, Carbon Impact has a new name: Cost & Carbon Optimization. The name change reflects the app’s expansion into cloud cost analytics, driven by the understanding that optimization approaches for both cost and carbon mostly overlap. You’ll be able to read more about our approach to cloud cost optimization in an upcoming blog post.

Industry certification for Dynatrace Cost & Carbon Optimization

To enhance the trust our customers and partners have in our approach, we commissioned the Sustainable Digital Infrastructure Alliance (SDIA) to test and certify the Cost & Carbon Optimization app. The certification focuses on accuracy and transparency in calculating greenhouse gas (GHG) emissions for AWS, Azure, GCP, and on-premises host instances.

The certification results are now publicly available.

“The SDIA has certified Dynatrace Cost & Carbon Optimization app, ensuring its accuracy and adherence to the environmental management principles outlined in ISO 14004.

“Dynatrace Cost & Carbon Optimization is a reliable estimation system to calculate the operational GHG emissions of IT infrastructure both in cloud and on-premises environments. The calculations and methodology used are in line with the best available scientific approach, as well as with relevant reporting requirements. Further, Dynatrace meets the requirements on transparency, allowing customers to have insight into the way these metrics are calculated, as well as recreating the measurements themselves for independent verification.”

Actions resulting from the evaluation

The certification process surfaced a few recommendations for improving the app. These are the outcomes:

• We replaced GCP’s emissions estimations with more accurate data from Ember, a global not-for-profit clean energy think tank.
• We implemented a wasted energy metric in the app to enhance practitioner actionability.
• We are updating product documentation to include underlying static assumptions.

Transparency breeds confidence

In the spirit of transparency, we’re publishing details of the data sources and assumptions the app makes. Some highlights:

• Data conversion: To derive CO2e and wasted energy metrics, we convert OneAgent data using external sources:
  • Energy emission data is sourced from the European Energy Agency and the Cloud Carbon Footprint tool to determine emission factors specific to cloud data center locations.
  • Thermal design power (TDP) values are derived from AMD and Intel to calculate CPU power consumption.
  • Power usage effectiveness (PUE) is derived from data provided by the cloud providers and data center operators.
• Static assumptions: To complete the calculations, we apply these assumptions:
  • Memory power calculations assume a static draw value of 3 W for each 8 GB memory module.
  • Network traffic power calculations rely on static power estimations for both public and private networks. These estimates are converted using the emission factor for the data center location. Static assumptions are:
    • Local network traffic uses 0.12 W per GB.
    • Public network traffic uses 1.0 W per GB.
  • CPU calculations apply these assumptions:
    • A virtual CPU (vCPU) on any cloud host equals one thread of a physical CPU core, with two threads per core.
    • A CPU operating at 100% utilization consumes power equal to its TDP.
  • Storage calculations assume that one terabyte consumes 1.2 Wh.
  • Cloud storage is replicated twice, which doubles the energy consumption per terabyte.
  • Annual country-level emission factors from the European Environmental Agency are used as input to the calculations.

Connecting carbon reduction to cloud cost optimization through the FinOps Framework

As we merge cost and carbon analytics into a single solution, we expect to reduce redundant or conflicting optimization efforts. Approaches to optimizing IT carbon emissions overlap significantly with those applied to cloud cost management (CCM). Optimizing resource utilization and reducing waste benefit each of these goals. The FinOps Foundation includes sustainability in its framework, stating, “FinOps and cloud sustainability mutually support each other through a similar approach to conscientious and responsible technology usage that enables workload efficiency. If you’re doing one of these – you’re amplifying the other.”