New Data Center AI Features in Cadence Reality DC 2024.1

We’ve been hinting at it for a while now, and it’s finally here. We’re excited to announce Cadence Reality DC 2024.1 is now available! This is our smartest release yet. This blog covers the important highlights regarding AI and automation we don’t want any of our customers to miss.

As always, if one of these new features stands out to you and you have further questions, please let our support team know. If you’re new to Cadence Reality DC and want to learn more, fill out our free trial form for a personalized software demo.

Automatically Identify Data Center Cooling Issues: Expert System

This is the smartest release we’ve ever launched. In 2024.1, Cadence data center software now includes a built-in expert system to help highlight potential issues and offer useful recommendations. We’ve literally named it “Expert System” because it’s that smart.

The Expert System assesses the state of the data center's digital twin model. It can then identify and describe any cooling problems as well as potential fixes. From here, the user can select the fixes they deem appropriate (either as one-off fixes or as bulk changes), and the Expert System will generate a new version of the model to test the impact of these changes.

Example of the Expert System discovering model issues

The current focus of the Expert System is to assess a solved model to identify cooling issues. Future extensions of this system will look at energy optimization, power, and general modeling errors, so be on the lookout for those in future releases!

Simplify Server Modeling: AI Model Features

Cadence data center software now includes more complete IT equipment behavior for simulation using machine learning. AI models are trained using Celsius EC Solver simulations, which can be augmented with any physical test data to generate intelligent models of servers.

These AI models offer several benefits in that they can:

• Capture complex fan control schemes
• Predict internal conditions (CPU, GPU, and hard disk temperatures for example)
• Capture the thermal response to different IT workloads (GPU, storage, etc.)
• Predict thermal throttling behavior due to elevated temperatures

The accuracy of our AI models is within 1% of the Celsius EC Solver simulation results.

Automate Optimization: Optimality Integration

Optimality can be launched from Cadence Reality DC to optimize a model and then merge the optimization scenarios back into the model. The process is simple. First, the baseline model is created in Cadence Reality DC. Next, the user will define the optimization parameters. Any property can be an input, and any result can be an output. The user can then launch Optimality from within Cadence Reality DC. Finally, Optimality will merge any optimization scenarios with its results back into the model, simplifying this editing process for the user.

Automate Data Center Insights for Reliable Decision-Making

Example of an insights dashboard included in Cadence data center software that supports decision-making

Cadence Reality DC provides a powerful toolset to optimize data center capacity and mitigate risk. With each new release, we strive to incorporate new functionality to help our customers design and operate more confidently. That’s exactly why this release includes important automations, so customers can spend less time on the manual processes and more time on the decisions that matter. 2024.1 equips data center designers and operators with the insights necessary to make informed, reliable decisions.

Learn more about Cadence Reality DC by visiting our product page or filling out our free trial form for a personalized software demo.