EightKB August 20th 2026 Schedule

UTC Time	Session	Speaker	Level	Length
13:00	Welcome!	EightKB Crew		15m
13:15	Enterprise Vibe Coding	Chrissy LeMaire	400	75m
Enterprise Vibe Coding While Spotify engineers stopped writing code and Anthropic is shipping 70-90% AI-generated code company-wide, SaaStr's Replit AI agent deleted their production database, fabricated 4,000 fake records to cover it, lied about recoverability, and scored a 95/100 severity rating. The difference between these outcomes is governance and specifically, automated enforcement baked into the code generation itself. This session walks through a concrete methodology for AI-assisted development aimed at DBAs, demonstrated through LeLab, a SQL Server lab automation framework that deploys full enterprise environments (Availability Groups, Failover Clusters, multi-node domains, chaos testing) on PowerShell 7, DSC v2, and Hyper-V. LeLab's AI hooks enforce infrastructure-as-code discipline at the point of code generation — not downstream in review, not after the PR is open, but as the code is being written. You'll walk out with the mental model, the hook patterns, and the workflow to bring AI-assisted development into your environment without losing control of it.
14:45	Vector Compression & Memory/Storage Economics: FP16 Vectors, and Matryoshka Embeddings	Mala Mahadevan	400	60m
Vector Compression & Memory/Storage Economics: FP16 Vectors, and Matryoshka Embeddings The single biggest issue with adopting vector search, despite its many benefits, is the cost of running it at scale. As vector dimensions and the datasets needed for them grow, the memory footprint, cache pressure, and storage bandwidth all start to increase. This talk examines vector search from a storage-first and systems engineering perspective, focusing on how representation directly impacts performance, cost, and retrieval quality. We start by understanding the basics of vector search, then delve into how high-dimensional float32 embeddings lead to a linear increase in storage and memory costs, which are further compounded across indexing, caching, and query execution. Using demos, we demonstrate that reducing vector size is an important design decision that can improve scalability and reduce costs. FP15 vectors can significantly contribute to this by cutting storage in half. We then get into an in-depth examination of various dimensionality strategies. Naive truncation and post‑hoc reduction are compared with more principled approaches that preserve retrieval quality. Reducing vector dimensionality directly removes information from the embedding space, often altering nearest‑neighbor relationships in subtle but impactful ways. The core of the session focuses on Matryoshka embeddings, a compression-based training paradigm that fundamentally changes how engineers think about vector size. We focus on understanding what these embeddings mean, how they are generated, and how useful they are compared to simply going from FP32 to FP16. We then conclude by reframing vector compression as an architectural choice rather than a compression trick. Audience takeaways will include a practical, systems‑level framework for choosing vector representations that balance memory, performance, and retrieval quality—without treating embeddings as immutable black boxes. The learning will help anyone dealing with vector search queries, although demos are specific to SQL Server 2025.
16:00	Unraveling the Layers of I/O: Your Next Step to Becoming a Comprehensive Performance Tuner	Andy Yun	400	75m
Unraveling the Layers of I/O: Your Next Step to Becoming a Comprehensive Performance Tuner Where are you on your SQL Server performance tuning journey? Many start by understanding execution plans and T-SQL anti-patterns. But some performance problems won't yield to query tuning alone. The best tuners tackle performance issues holistically, traversing logical to physical layers of the stack. Are you ready to take that next step? If so, then join me as we explore that next step together, unraveling the I/O lifecycle layer by layer. We will start by bridging execution plan operators to physical access methods. Then we'll look beyond SQL Server and explore the full I/O pathway. All along the way, you will learn about performance bottlenecks that cause you to get paged at 3AM. When you leave, you'll be equipped with a practical understanding of the I/O lifecycle, enabling you to dissect performance issues holistically across your entire SQL Server stack.
17:30	Accelerated Database Recovery - A Deep Dive Behind the Magic	John Morehouse	400	75m
Accelerated Database Recovery - A Deep Dive Behind the Magic We have all heard the stories. Horror story after horror story being regaled from database administrators all over the world of waiting for hours or sometimes days for a rollback operation to complete. DBA's hoping beyond hope that the rollback finishes soon before someone else gets the notion to reboot the server. With the release of SQL Server 2019, a new talisman was conjured that will save us all from hearing about future horror stories. Accelerated Database Recovery or also known as "ADR". This new mystical feature changes the way transaction rollback performs and will undoubtedly prove invaluable in your arsenals of magical tricks! In this session, we'll show you the secret behind it! You will - • Learn about new components of the transaction log • Discover the magic behind ADR • Determine when you wouldn't want to implement it Don't be the protagonist in a rollback horror story! After all, the time you save might be your own!
19:00	Always On Availability Groups and Failover Clustering Internals	Edwin M Sarmiento	400	75m
Always On Availability Groups and Failover Clustering Internals Windows Server Failover Clustering (WSFC) is the platform that makes SQL Server Always On Availability Groups highly available. There are external interactions between WSFC and SQL Server that you as a DBA cannot monitor from within SQL Server. You need to know the different components that make up the WSFC and how they work to properly deploy and manage Always On Availability Groups for high availability. In this session, you will dive into the internals of the WSFC, the SQL Server cluster resource DLL, and how they work together to make your SQL Server databases highly available. You will analyze the cluster error log for identifying the root cause of an unexpected or planned failover. Finally, you will learn administrative tasks that you're not supposed to do outside of SQL Server.

EightKB 2026 Registration is Open!

Who is EightKB?

• Andrew Pruski (T|B)
• Mark Wilkinson (T|B)
• Anthony Nocentino (T|B)

We created this event to fill a void in the SQL Server conference line-up. While there are many important topics to discuss and things to learn in the SQL Server world, we want this conference to focus on the core components of SQL Server, the things that make SQL Server work. EightKB is about expert topics, not expert speakers. If you have expert experience about any SQL Server topic, the community wants to hear you speak. If you aren’t yet an expert speaker, we will help you with your abstract and session and even pair you with a mentor if you like. A core goal of this conference is to enable SQL Server professionals to bring their whole selves to the event. In particular, we believe that a diverse and inclusive speaker community is foundational to everyone feeling welcome at our conference.