Royal Ahold Solves Mainframe Problems in Fraction of the Time Required in the Past
Industry: Supermarket Chain and Foodservice
Location: Global
Royal Ahold is one of the world’s leading supermarket chains and foodservice companies with operations in 28 countries throughout the United States, Europe, Latin America and Asia. Ahold owns and operates nearly 9,000 supermarkets, and provides service to 40 million customers a week. Ahold Information Services (AIS) is the information technology company of Ahold USA. Headquartered in Greenville, South Carolina, AIS is the advisor and supplier of information technology-related products and services to Ahold USA and its business units.
The Challenge
The AIS group plays an important strategic role in providing data services
to the operating units of the company and as such has an enormous
responsibility to respond promptly and effectively in any situation
where users are reporting operational difficulties. Meeting this goal
was difficult in the past because of the limited diagnostic capabilities
offered by MVS and CICS”. It typically
involved a trial-and-error approach that required a considerable amount
of time in the worst possible circumstances – while users were suffering
from slow performance or perhaps even a system outage. The time-consuming
diagnostic process also placed a considerable burden upon the technical
staff who were forced to juggle the large, unexpected blocks of time
with their day-to-day responsibilities.
The Solution
AIS management made the decision to seek out a comprehensive tool to monitor
CICS, MVS, and DB2® activity. They selected
ASG-TMON™ for DB2® because
it includes a very comprehensive toolset, as well as an exceptionally intuitive
user interface. TMON offers a drill-down approach to system and application
tuning that includes system overviews, point-and-shoot zoom and transaction-level
reporting granularity. This saves time by making it possible to quickly
scan a wide range of metrics and then quickly zero in to get more detailed
information when a problem is detected. The program also improves user productivity
by making it possible to set threshold alerts within monitored regions that
trigger advance warnings – allowing systems programmers to immediately react
to problems, or even head them off before they occur.
“I use the tool on a daily basis to keep tabs on system operations and to help diagnose and, sometimes, even solve any problems that are thrown my way,” said David Troutman, manager of database administration for Ahold. “Because our applications are very database-intensive, this is typically the first area that I suspect. TMON for DB2 tells me everything I need to know. I usually check the DB2 threads first to see if the number of ‘get pages’ is increasing. If this number is rising, there’s a good chance that the problem might be in DB2. To probe further, I usually look at another panel of the program that shows the time that the program has spent in DB2 compared to the total for the application. If one side is dominant, the chances are great that that’s where the problem is. Through a step-by-step process, I am able to quickly eliminate possible trouble areas and zero in on what is actually causing the problem.”
Measurable Results
TMON for DB2 performance management is helping Ahold solve problems and improve
operating efficiency of mainframe computers in a fraction of the time required in the
past. The software helps first by identifying the existence of a problem, in some cases
by tracking the number of ‘get pages’ generated by an application. Second, it can
determine in general terms where the problem lies, for example, by showing the amount
of CPU time spent in the application vs. DB2. Finally, it can zero in on the exact area of
the application that is causing the problem, substantially reducing the amount of time
required to implement a fix. As a result, typical turnaround time to identify, diagnose,
and fix problems has been reduced to just 90 minutes.
“We use the historical statistics generated by the TMON monitoring software to keep track of overall performance trends,” Troutman said. “We review our CICS transactions from time-to-time to see how frequently they run, we check DB2 transactions to see if we are having problems with deadlocks, and so on. Over time, patterns emerge that help us solve problems that are invisible on a short-term basis. In short, implementing performance monitoring with TMON has helped substantially reduce the time required to improve our users” experience while reducing the strain on technical staff.”
