George E. Stark

An examination of the effects of requirements changes on software maintenance releases

Requirements are the foundation of the software release process. They provide the basis for estimating costs and schedules, as well as developing design and testing specifications. When requirements have been agreed on by both clients and maintenance management, then adding to, deleting from, or modifying those existing requirements during the execution of the software maintenance process impacts the maintenance cost, schedule, and quality of the resulting product. The basic problem is not the changing in itself, but rather the inadequate approaches for dealing with changes in a way that minimizes and communicates the impact to all stakeholders. Using data collected from one organization on 44 software releases spanning seven products, this paper presents two quantitative techniques for dealing with requirements change in a maintenance environment. First, exploratory data analysis helps one to understand the sources, frequency, and types of changes being made. Second, a regression model helps managers communicate the cost and schedule effects of changing requirements to clients and other release stakeholders. These two techniques can help an organization provide a focus for management action during the software maintenance process. Copyright

Modeling a complex global service delivery system

Enterprises and IT service providers are increasingly challenged with improving the quality of service while reducing the cost of service delivery. Effectively balancing dynamic customer workload, strict service level constraints, and diverse service personnel skills challenges the most experienced management teams. In this paper we describe a modeling framework for analyzing complex service delivery systems. The interaction among various key factors are included in the model to allow decision-making around staffing skill levels, scheduling, and service level constraints in system design. We demonstrate the applicability of the proposed approach in a large IT services delivery environment.

Classifying server behavior and predicting impact of modernization actions

Today the decision of when to modernize which elements of the server HW/SW stack is often done manually based on simple business rules. In this paper we alleviate this problem by supporting the decision process with an automated approach based on incident tickets and server attributes data. As a first step we identify and rank servers with problematic behavior as candidates for modernization using a random forest classifier. Second, this predictive model is used to evaluate the impact of different modernization actions and suggest the most effective ones. We show that our chosen model yields high quality predictions and outperforms traditional linear regression models on a large set of real data.

Impact of HW and OS type and currency on server availability derived from problem ticket analysis

Technology refresh is an important component in data center management. The goal of this paper is to assess the impact of HW and OS currency on server availability based on a large set of incident tickets and server attributes data collected from several different IT environments. In order to achieve this we first identify the server failure incidents using a machine learning method for automatic ticket classification. Then we conduct the data analysis to inspect the impact of HW and OS type along with their currency on the rates of server failures. This can further be used to derive guidelines to support the technology refresh decisions in the data centers.

COACH: Cognitive analytics for change

This paper presents our initial efforts towards building a cognitive analytics framework for change management. We propose a novel predictive algorithm for change risk calculation based on historical change failures, server failures, change triggered incidents as well as expert user input. Our predictive algorithm provides significant improvement over traditional risk assessments in proactively capturing problematic changes when tested with real client account data.

On the adoption and impact of predictive analytics for server incident reduction

The Predictive Analytics for Server Incident Reduction (PASIR) solution developed at IBM has been broadly deployed to 130 IT environments since the beginning of 2014. The infrastructures of these IT environments, pertaining to various industries around the world, are serviced by IBM support groups. More specifically, incidents occurring on servers, including the descriptions of the problems, are reported into a ticket management system. These tickets are then resolved by the assigned support teams, which record in the system the resolution steps taken. PASIR, first classifies the incident tickets of an IT environment to identify high-impact incidents describing server unavailability and performance degradation issues by using ticket descriptions and resolutions. Second, the occurrence of these high-impact tickets is correlated with server properties and utilization measures to identify troubled server configurations and prescribe improvement actions through multivariate analysis. In this paper, we present the findings from deploying our two-step machine learning model in the field. In particular, we describe the PASIR methodology, from ticket classification to the recommendation of modernization actions. We also assess the process of manual ticket labeling and the impact of noisy input data on our automatic classifier, and we demonstrate the model effectiveness by comparing predictions on the impact of prescriptive actions with actual system improvements.

Modeling process and product quality during maintenance

This paper contains data demonstrating our recent experiences with measuring quality of evolving systems. Both process and product quality measures are discussed. This is an area in which more effective collaboration between practitioners and researchers would be of great value. We note that access to industrial software by researchers is often blocked by proprietary restrictions. When such restrictions can be eased, publication of analysis results is often hampered by the industrial owners and developers of the software. I believe that practice can be significantly aided by the data and results of broadly based research studies. Thus, closer collaboration in this area will benefit both communities.