18534559. TUNING LARGE DATA INFRASTRUCTURES simplified abstract (Microsoft Technology Licensing, LLC)

TUNING LARGE DATA INFRASTRUCTURES

Organization Name

Microsoft Technology Licensing, LLC

Inventor(s)

Yiwen Zhu of Sunnyvale CA (US)

Subramaniam Venkatraman Krishnan of Santa Clara CA (US)

Konstantinos Karanasos of San Francisco CA (US)

Carlo Curino of Woodinville WA (US)

Isha Tarte of Woodinville WA (US)

Sudhir Darbha of Redmond WA (US)

TUNING LARGE DATA INFRASTRUCTURES - A simplified explanation of the abstract

This abstract first appeared for US patent application 18534559 titled 'TUNING LARGE DATA INFRASTRUCTURES

Simplified Explanation

An automated tuning service is used to automatically tune, or modify, the operational parameters of a large-scale cloud infrastructure. The tuning service leverages domain knowledge and data science to capture the dynamic behavior of the cluster in machine learning models, which are used for automated optimization procedures and strategic decision-making.

• The tuning service automatically adjusts operational parameters of a cloud infrastructure based on real-time performance data.
• Machine learning models are used to optimize parameter tuning and inform administrators in engineering decisions.
• Observational models and testing in production are combined to configure operational parameters for various applications.

Potential Applications

The technology can be applied in:

• Cloud computing
• Data centers
• Network infrastructure

Problems Solved

The technology addresses issues such as:

• Manual tuning of operational parameters
• Inefficient resource allocation
• Lack of real-time performance optimization

Benefits

The benefits of this technology include:

• Improved performance of cloud infrastructure
• Automated optimization procedures
• Strategic decision-making support

Potential Commercial Applications

The technology can be commercially applied in:

• Cloud service providers
• IT consulting firms
• Data analytics companies

Possible Prior Art

One possible prior art for this technology could be automated performance tuning tools used in data centers or network management systems.

How does the tuning service handle security considerations in parameter optimization?

The article does not address how the tuning service ensures that security is not compromised during the automated parameter optimization process. This is an important aspect to consider, especially in sensitive cloud environments.

What are the potential limitations of using machine learning models for parameter tuning in a cloud infrastructure?

The article does not discuss any potential drawbacks or limitations of relying on machine learning models for parameter tuning. It would be beneficial to understand any challenges or constraints that may arise from this approach.

Original Abstract Submitted

An automated tuning service is used to automatically tune, or modify, the operational parameters of a large-scale cloud infrastructure. The tuning service performs automated and fully data/model-driven configuration based from learning various real-time performance of the cloud infrastructure. Such performance is identified through monitoring various telemetric data of the cloud infrastructure. The tuning service leverages a mix of domain knowledge and principled data-science to capture the essence of our cluster dynamic behavior in a collection of descriptive machine learning (ML) models. The ML models power automated optimization procedures for parameter tuning, and inform administrators in most tactical and strategical engineering/capacity decisions (such as hardware and datacenter design, software investments, etc.). Rich “observational” models (models collected without modifying the system) are combined with judicious use of “fighting” (testing in production), allowing the tuning service to automatically configure operational parameters of a large cloud infrastructure for a broad range of applications.