- Google is moving thousands of internal workloads from x86 processors to Arm processors
- The company built an AI tool called CogniPort to automate migration fixes
- Google engineers spent months fixing test failures related to x86 infrastructure
Google has embarked on a hugely ambitious project to migrate all of its internal workloads from x86 processors to Arm processors, a process that involves one of the largest hardware transitions ever attempted by a global technology company.
This effort aims to enable its systems to run efficiently on x86 processors and its custom Axion silicon.
With around 30,000 apps already converted, Google continues to rely heavily on automation to manage the huge code base involved in the process.
Porting Workloads Warehouse-Wide
In a blog post describing the project, Google developer relations engineer Parthasarathy Ranganathan and developer relations engineer Wolff Dobson noted that the migration began with some of the company’s most critical systems, including F1, Spanner, and Bigtable.
Initially, the teams relied on conventional software development practices with dedicated engineers and weekly coordination meetings.
Although major architectural hurdles were expected, modern compilers and debugging tools have helped reduce many of the anticipated problems.
However, a lot of time was still spent fine-tuning thousands of tests closely tied to Google’s existing x86 infrastructure.
Engineers also faced challenges updating existing build and release systems, managing production deployments, and ensuring stability in mission-critical environments.
To speed up the transition, Google has developed a new AI tool called “CogniPort”.
The system works by analyzing build and testing errors and then attempting to fix them automatically, especially in cases where an Arm-specific library or binary fails to compile.
CogniPort showed a success rate of around 30%, with better performance in handling test fixes, data processing inconsistencies, and conditional platform code.
While not perfect, the tool represents a key step in enabling warehouse-wide automation and reducing the human workload required for such conversions.
The long-term motivation behind Google’s decision lies in performance and efficiency: its Axion-powered Arm servers are said to offer up to 65% better value for money and can be up to 60% more energy efficient than comparable x86 instances.
This change could lead to fewer x86 processors in Google’s vast data infrastructure, potentially transforming the makeup of its internal compute clusters.
For now, major apps like YouTube, Gmail, and BigQuery already work on x86 and Arm systems.
As Google migrates the remaining 70,000 packages, doubts persist about whether AI tools can handle such scale without adding new maintenance challenges across its entire systems.
Via The register
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds. Make sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp Also.
