Browse Prior Art Database

System for Real Time Performance Tuning and Recommendations

IP.com Disclosure Number: IPCOM000248671D
Publication Date: 2016-Dec-24
Document File: 2 page(s) / 92K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system to analyze the current system settings based on the environment to provide recommendations and feedback for optimal performance for on-premise and cloud based machines and services. This includes on-premise products, cloud based systems (data center), or mobile devices.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 50% of the total text.

1

System for Real Time Performance Tuning and Recommendations

Deploying software across multiple operating systems with different hardware configurations requires optimal tuning for software performance based on hardware configurations, middleware software versions, and tuning for end users systems. Setup and configuration of systems is based on default values and recommended documentation and often requires additional tuning by an expert support team. Configuration using non-default tuning best practices can be expensive and inconvenient for the client, and is usually not an automated process. Because the tuning can be applied only after performance issues have been encountered and identified, impacts on production environments occur and costs for maintaining the system increase.

The novel contribution is a system to analyze the current system settings based on the environment to provide recommendations and feedback for optimal performance for on-premise and cloud based machines and services. This includes on-premise products, cloud based systems (data center), or mobile devices.

The system intelligently, dynamically, and automatically maintains optimal system performance and increases customer satisfaction, reduces overhead for clients, and provides real time feedback for developers. Users can subscribe to the default setup, use existing platforms to develop a specific interface, or add the applicable information of interest for performance tuning. Natural language processing (NLP) systems can interpret documentation about the product.

The new system builds a corpus of information that reflects data collected from customer scenarios, from quality assurance (QA) environments, product documentation, and other historical data such as customer support issues opened against the product. Other included data can be event-based data, such as application logs, server logs, and database logs. In addition, information based on the historical data, big data (e.g., instant messaging/online information), and risk management. This solution provides more accurate and systematic tuning for each customer environment, not only the system resource tuning (e.g., central processing unit/random access memory (CPU/RAM), but also high availability (HA)/load balancer configuration/application server tuning/database configuration/statistic/index/spam-ware recommendations (mobile). Natural language processing understands product documentation such as the readme, the product manual, release notes, technotes, and documents related to the software that are published with the software and on external websites/forums. Natural language processing includes customer support tickets and escalations, blogs/forums, and other streams of support for a product.

The novel system allows an agent to contact the NLP service to upload current environment settings for each machine with an installed agent. The cognitive analysis service processes product documentation and quality assuranc...