Milwaukee Tech Day

The multi-tenant option is critical to upgrading to Oracle Database 23c. If you have not embraced it yet, what are the things that you need to factor in as part of the upgrade.  We will discuss the upgrade path to Oracle Database 23c.  We will discuss what you can do to prepare for this journey, how to have the important conversation with your management on why this upgrade is critical to the organization, and most importantly, what are some of the options to reduce or eliminate your downtime.

MySQL HeatWave is designed to handle a wide variety of critical use cases, increase the performance of queries by up to several orders of magnitude, provide powerful no code machine learning, and run in-memory SQL queries against up to 500 TB of text files. This eliminates the need for ETL, specialty databases, advanced ML expertise, and data duplication. We’ll examine the architecture, design considerations, performance gains, and examples of each of these strengths. We’ll provide examples of how this solution is priced. Finally, we’ll have “what-if” examples of the impact against contemporary business challenges.

Tuning SQL statements in Oracle is more of an art than a science.  In this session we will cover more than just the basic steps to tuning a SQL statement.  Starting with finding the SQL statement and collecting the right information.  Covering questions like “How the Optimizer works to transform SQL into an execution plan?”  “What information is an execution plan is trying to show?”  “What do data access paths mean?” Learn about the detailed steps in SQL execution plans, SQL tracing and how it compares to the information in an explain plan.  If you are new to tuning SQL in Oracle or trying to freshen up your skills, this session should have enough information for everyone.

I'm naturally curious and fascinated by the rabbit hole ChatGPT universe. The user-friendly interface doesn't have intrusive ads which I love. My natural curiosity usually shows me the desired results. However, both ChatGPT and commercial LLMs have limitations. They tend to please their owners resulting in biased, lackluster, and occasionally strange responses. Most of the time they lack awareness of personal documents, so their answers are incomplete. This problem prompted me to see if utilizing LLMs to analyze my chosen documents would allow the documents to speak for themselves. We will start with just six lines of Python, and end with a fully functional web app. What's more, you can deploy it all on your laptop in just a few minutes. Let's jump into this rabbit hole together!