Sign in Subscribe
Machine learning

Cloud Instance Sizing and Deployments Using Machine Learning

Strategic Insights

— 1 min read

The inventors Dinh, Hung, Linsey, David J., and Mohanty, Bijan Kumar have developed a cloud instance prediction platform that utilizes machine learning algorithms to predict the configuration of a cloud instance based on the features of an application. The platform aims to simplify the process of selecting the optimal virtual environment for cloud-based software deployments. By analyzing the features of an application, the platform can predict the required amount of computer resources, such as central processing unit utilization, memory utilization, and disk input-output utilization, to ensure efficient and cost-effective cloud instance deployments.

Key Takeaways:

  • The cloud instance prediction platform receives a request to predict the configuration of a cloud instance based on the features of an application, which includes one or more features such as code size, code language, complexity tier, interactivity determination, and execution time.
  • The platform analyzes the features using one or more machine learning algorithms, including neural networks, to predict the configuration of the cloud instance.
  • The platform can predict multiple targets, including the cloud platform, central processing unit utilization, memory utilization, and disk input-output utilization.
  • The neural network uses a classification technique for the first target and a regression technique for the remaining targets.
  • The platform can interface with cloud platforms to collect runtime metrics and use them for training the machine learning algorithms.
  • The platform can be implemented as an apparatus with a processing device and a memory or as a non-transitory processor-readable storage medium with stored program code.
  • The platform can be used to predict the configuration of a cloud instance for micro-frontend and microservice applications.

Statistics:

  • The platform can predict up to 4 targets, including cloud platform, central processing unit utilization, memory utilization, and disk input-output utilization.
  • The platform uses a neural network with 2 hidden layers and a rectified linear unit activation function.
  • The platform includes 12 claims, with 2 claims related to the method, 6 claims related to the apparatus, and 4 claims related to the article of manufacture.
  • The patent application was filed on March 27, 2024, and made available online on October 2, 2025.

Sources:

  • Dinh, Hung; Linsey, David J.; Mohanty, Bijan Kumar. Cloud Instance Sizing And Deployments Using Machine Learning. U.S. Patent Application Number 20250310194, filed March 27, 2024 and posted October 2, 2025. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(20250310194)&db=US-PGPUB&type=ids

Read more