Hotsos Symposium Speaker – Andrew Zitelli
Mr. Zitelli is a prior Hotsos Symposium presenter.
Biography
Andrew Zitelli has thirty years of experience as a software developer, data architect and performance analyst. He has been a consultant to aerospace, semiconductor, steel, and pharmaceutical industries. He is proficient working with a broad range of operating systems and languages plus he has twenty-five years of experience working with ten different relational database products, including fifteen years working with Oracle. Mr. Zitelli possesses both a BA and an MS degree in Computer Science. Mr. Zitelli has presented at technical conferences on a wide range of software topics.
Presentation Title
Leveraging Oracle's Extended SQL Trace Data to Expedite Software Development
Abstract
During software development, the cost to repair software flaws tends to increase over time. A flaw which may be trivial and inexpensive to fix early in the development process can become extremely difficult and costly to repair later. Software flaws manifest themselves in a wide variety of ways. Some common symptoms include erroneous results, poor performance, low throughput, system instability, reduced functionality and propagation of errors to other system components. Early detection and timely repair of software flaws can be critical to the success of many software projects.
This presentation will discuss the use of Oracle's extended SQL trace data to assist in the early detection and repair of software flaws. Sometimes referred to as 10046 trace data, this data provides an unbroken record of the ordering, timing and content of SQL commands, executing within individual database sessions. Summarized contents of this trace data can quickly expose a variety of software flaws. Other types of flaws can be inferred from the specific sequence, timing and contents of the SQL commands being executed.
The techniques to be discussed have been applied over the past five years, to identify software flaws on several large software development projects. These techniques strongly complement the use of Hotsos' Method R, commonly used with 10046 trace data to diagnose system performance problems. Examples will be drawn from real-world software problems, discovered through the examination and analysis of 10046 trace data.
Presentation Materials
Presentation materials are available to attendees only. Attendees may find Mr. Zitelli's presentation on the
The speaker schedule can be found at the following links:Schedule
