Browse Prior Art Database

A Method for Autonomous Serviceability Software Product Development Disclosure Number: IPCOM000237814D
Publication Date: 2014-Jul-14
Document File: 3 page(s) / 39K

Publishing Venue

The Prior Art Database


A method for self-problem detection and potentially self-healing in a software product is described by the article. The innovative aspect of this method is the fact that it is embedded throughout the development lifecycle of the product and enables continuous feedback loop into development once the product is deployed and installed at the consumer site (customer).

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

A Method for Autonomous Serviceability Software Product Development

Problem Statement:

Currently serviceability in software development mainly consists of logging error messages along the error paths in addition to what is called first failure data capture (FFDC) info that is automatically captured during failures to give some indication about problem determination.

The details and level of logging of error and trace messages is usually controlled through configuration setup. Upon failures, product documentation and support teams provide direction on how to further setup detailed traces in order to refine the logs and reach problem determination. This traditional process is time consuming and costly to both the software producer/provider and the software consumer as it usually requires many iterations going back and forth until the problem is determined. A resolution is usually provided through manual re-configuration or through code fixes delivered to the software consumer.

Summary of the Solution:

Currently software product serviceability is mostly static in nature. It can be characterized as being after the fact process that is lengthy, frustrating and costly while going back and forth between the software consumer and the software provider. This process is static in that it relies solely on manually stepping up trace and logging levels until the problem is determined. It is lengthy and costly because it may take a while going back and forth between the consumer and the provider of the software before the right trace and logging specification yields the correct problem determination. This costs resources - money and time.

The problem stems from the fact that serviceability of software is currently not a well embedded part of the software development lifecycle from conception to design, development and testing.

In this new method the attempt to close this gap is made by making software serviceability part of all the development lifecycle from its early stage all the way to deployment.

In addition to the usual product setup and configuration artifacts, products adopting this new method will ship with artifacts related to autonomous problem determination and possibly autonomous serviceability. Details follow in the next section.

This new method seeks to eliminate the extraneous processes going back and forth between the software consumer and provider by way of having problem determination embedded in the product runtime. A new metadata artifact will ship with the product in order to govern self-serviceability (explained in the details section)


Page 02 of 3

Solution Details:

In this new serviceability oriented software development serviceability is completely integrated in the design and development cycle.