Jackie Brosamer, Head of AI, Data & Analytics at Block shares how Block accelerated GenAI production during the Data + AI Summit 2024 in San Francisco.

Databricks Data + AI Summit 2024 Keynote Day 1

Experts, researchers and open source contributors — from Databricks and across the data and AI community gathered in San Francisco June 10 - 13, 2024 to discuss the latest technologies in data management, data warehousing, data governance, generative AI for the enterprise, and data in the era of AI.

Hear from Databricks Co-founder and CEO Ali Ghodsi on building generative AI applications, putting your data to work, and how data + AI leads to data intelligence.

Plus a fireside chat between Ali Ghodsi and Nvidia Co-founder and CEO, Jensen Huang, on the expanded partnership between Nvidia and Databricks to accelerate enterprise data for the era of generative AI

Product announcements in the video include: - Databricks Data Intelligence Platform - Native support for NVIDIA GPU acceleration on the Databricks Data Intelligence Platform - Databricks open source model DBRX available as an NVIDIA NIM microservice - Shutterstock Image AI powered by Databricks - Databricks AI/BI - Databricks LakeFlow - Databricks Mosaic AI - Mosaic AI Agent Framework - Mosaic AI Agent Evaluation - Mosaic AI Tools Catalog - Mosaic AI Model Training - Mosaic AI Gateway

In this keynote hear from: - Ali Ghodsi, Co-founder and CEO, Databricks (1:45) - Brian Ames, General Motors (29:55) - Patrick Wendall, Co-founder and VP of Engineering, Databricks (38:00) - Jackie Brosamer, Head of AI, Data and Analytics, Block (1:14:42) - Fei Fei Li, Professor, Stanford University and Denning Co-Director, Stanford Institute for Human-Centered AI (1:23:15) - Jensen Huang, Co-founder and CEO of NVIDIA with Ali Ghodsi, Co-founder and CEO of Databricks (1:42:27) - Reynold Xin, Co-founder and Chief Architect, Databricks (2:07:43) - Ken Wong, Senior Director, Product Management, Databricks (2:31:15) - Ali Ghodsi, Co-founder and CEO, Databricks (2:48:16)

There are more data tools available than ever before, and it is easier to build a pipeline than it has ever been. These tools and advancements have created an explosion of innovation, resulting in data within today's organizations becoming increasingly distributed and can't be contained within a single brain, a single team, or a single platform. Data lineage can help by tracing the relationships between datasets and providing a map of your entire data universe.

OpenLineage provides a standard for lineage collection that spans multiple platforms, including Apache Airflow, Apache Spark™, Flink®, and dbt. This empowers teams to diagnose and address widespread data quality and efficiency issues in real time. In this session, we will show how to trace data lineage across Apache Spark and Apache Airflow. There will be a walk-through of the OpenLineage architecture and a live demo of a running pipeline with real-time data lineage.

Talk by: Julien Le Dem,Willy Lulciuc

Here’s more to explore: Data, Analytics, and AI Governance: https://dbricks.co/44gu3YU

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

While we may not all agree on a “If it ain’t broke, don’t fix it” approach, we can all agree that “If it shows any crack, migrate it to the cloud and completely re-architect it.” Akamai’s CSI (Cloud Security Intelligence) group is responsible for processing massive amounts of security events arriving from our edge network, which is estimated to process 30% of internet traffic, making it accessible by various internal consumers powering customer-facing products.

In this session, we will visit the reasons for migrating one of our mission critical security products and its 10GB ingest pipeline to the cloud, examine our new architecture and its benefits and touch on the challenges we faced during the process (and still do). While our requirements are unique and our solution contains a few proprietary components, this session will provide you with several concepts involving popular off-the-shelf products you can easily use in your own cloud environment.

Talk by: Yaniv Kunda

Here’s more to explore: Why the Data Lakehouse Is Your next Data Warehouse: https://dbricks.co/3Pt5unq Lakehouse Fundamentals Training: https://dbricks.co/44ancQs

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

YES! "Duck posting" has become an internet meme for praising DuckDB on Twitter. Nearly every quack using DuckDB has done it once or twice. But, why all the fuss? With advances in CPUs, memory, SSDs, and the software that enables it all, our personal machines are powerful beasts relegated to handling a few Chrome tabs and sitting 90% idle. As data engineers and data analysts, this seems like a waste that's not only expensive, but also impacting the environment.

In this session, you will see how DuckDB brings SQL analytics capabilities to a 2MB standalone executable on your laptop that only recently required a large cluster. This session will explain the architecture of DuckDB that enables high performance analytics on a laptop: great query optimization, vectorized execution, continuous improvements in compression and more. We will show its capabilities using live demos, from the pandas library to WASM, to the command-line. We'll demonstrate performance on large datasets, and talk about how we're exploring using the laptop to augment cloud analytics workloads.

Talk by: Ryan Boyd

Here’s more to explore: Why the Data Lakehouse Is Your next Data Warehouse: https://dbricks.co/3Pt5unq Lakehouse Fundamentals Training: https://dbricks.co/44ancQs

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Ontada, a McKesson business, is an oncology real-world data and evidence, clinical education and provider of technology business dedicated to transforming the fight against cancer. Core to Ontada’s mission is using real-world data (RWD) and evidence generation to improve patient health outcomes and to accelerate life science research.

To support its mission, Ontada embarked on a journey to migrate its enterprise data warehouse (EDW) from an on-premise Oracle database to Databricks Lakehouse. This move allows Ontada to now consume data from any source, including structured and unstructured data from its own EHR and genomics lab results, and realize faster time to insight. In addition, using the Lakehouse has helped Ontada eliminate data silos, enabling the organization to realize the full potential of RWD – from running traditional descriptive analytics to extracting biomarkers from unstructured data. The session will cover the following topics:

• Oracle to Databricks: migration best practices and lessons learned
• People, process, and tools: expediting innovation while protecting patient information using Unity Catalog
• Getting the most out of the Databricks Lakehouse: from BI to genomics, running all analytics under one platform
• Hyperscale biomarker abstraction: reducing the manual effort needed to extract biomarkers from large unstructured data (medical notes, scanned/faxed documents) using spaCY and John Snow Lab NLP libraries

Join this session to hear how Ontada is transforming RWD to deliver safe and effective cancer treatment.

Talk by: Donghwa Kim

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

As the healthcare and life science (HLS) industry has grown and evolved, a need has emerged for scalable and cost-effective ETL solutions capable of processing billions of records at terabyte scale. IQVIA has the largest global healthcare data networks in the world, with over one million data sources providing access to 1.2B non-identified patient records and 100 billion healthcare records processed annually in over 100 countries. IQVIA’s ability to combine, centralize, and integrate various sources of HLS data enables clinical-to-commercial operational intelligence and omnichannel analytics for its clients. Databricks Lakehouse allows IQVIA to onboard the rapidly growing number of clients while delivering strong business value to customers, cost-efficiently and at scale. 

During this session, you will learn more about how IQVIA is leveraging Databricks Lakehouse as well as how HLS organizations can soon access IQVIA data assets though the Databricks Marketplace for quick and secure data sharing.

Talk by: Venkat Dasari and William Zanine

Here’s more to explore: State of Data + AI Report: https://dbricks.co/44i2HBp The Data Team's Guide to the Databricks Lakehouse Platform: https://dbricks.co/46nuDpI

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

he increasing popularity of data analytics and artificial intelligence (AI) has led to a dramatic increase in the volume of data being used in these fields, creating a growing need for an enhanced computational capability. Cache plays a crucial role as an accelerator for data and AI computations, but it is important to note that these domains have different data access patterns, requiring different cache strategies. In this session, you will see our observations on data access patterns in the analytical SQL and AI training domains based on practical experience with large-scale systems. We will discuss the evaluation results of various caching strategies for analytical SQL and AI and provide caching recommendations for different use cases. Over the years, we have learned some best practices from big internet companies about the following aspects of our journey:

1. Traffic pattern for analytical SQL and cache strategy recommendation
2. Traffic pattern for AI training and how we can measure the cache efficiency for different AI training process
3. Cache capacity planning based on real-time metrics of the working set
4. Adaptive caching admission and eviction for uncertain traffic patterns

Talk by: Chunxu Tang and Beinan Wang

Here’s more to explore: State of Data + AI Report: https://dbricks.co/44i2HBp Databricks named a Leader in 2022 Gartner® Magic QuadrantTM CDBMS: https://dbricks.co/3phw20d

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Too often business decisions in large organizations are based on time consuming and labor-intensive data extracts, fragile Excel or access sheets that require significant manual intervention. The teams that prepare these manual reports have invaluable heuristic knowledge that, when combined with meaningful data and tools, can make smart business decisions. Imagine a world where these business teams are empowered with tools that help them build meaningful reports despite their limited technical expertise.

In this session, we will discuss: - The value derived from investing in developing citizen data personas within a business organization - How we successfully built a citizen data analytics culture within Michelin - Real examples of the impact of this initiative on the business and on the people themselves

The audience will walk away with some convincing arguments for building a citizen data culture in their organization and a how-to cookbook that they can use to cultivate citizen data personas. Finally, they can interactively uncover key success factors in the case of Michelin that can help drive a similar initiative in their respective companies.

Talk by: Philippe Leonhart and Fabien Cochet

Here’s more to explore: Why the Data Lakehouse Is Your next Data Warehouse: https://dbricks.co/3Pt5unq Lakehouse Fundamentals Training: https://dbricks.co/44ancQs

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Rust is a unique language whose traits make it very appealing for data engineering. In this session, we'll walk through the different aspects of the language that make it such a good fit for big data processing including: how it improves performance and how it provides greater safety guarantees and compatibility with a wide range of existing tools that make it well positioned to become a major building block for the future of analytics.

We will also take a hands-on look through real code examples at a few emerging technologies built on top of Rust that utilize these capabilities, and learn how to apply them to our modern lakehouse architecture.

Talk by: Oz Katz

Here’s more to explore: Why the Data Lakehouse Is Your next Data Warehouse: https://dbricks.co/3Pt5unq Lakehouse Fundamentals Training: https://dbricks.co/44ancQs

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

The biggest critical factor to success in a cloud transformation is people. As such, having a change management process in place to manage the impact of the transformation and user enablement is foundational to any large program. In this session, we will dive into how TD bank democratizes data, mobilizes a community of over 2000 analytics users and the tactics we used to successfully enable new use cases on Cloud. The session will focus on the following:

To democratize data: - Centralize a data platform that is accessible to all employees and allow for easy data sharing - Implement privacy and security to protect data and use data ethically - Compliance and governance for using data in responsible and compliant way - Simplification of processes and procedures to reduce redundancy and faster adoption

To mobilize end users: - Increase data literacy: provide training and resources for employees to increase their abilities and skills - Foster a culture of collaboration and openness: cross-functional teams to collaborate and share ideas - Encourage exploration of innovative ideas that impact the organization's values and customers technical enablement and adoption tactics we've used at TD Bank:

1. Hands-on training for over 1300+ analytics users with emphasis on learn by doing, to relate to real-life situations
2. Online tutorials and documentations to be used as self-paced study
3. Workshops and office hours on specific topics to empower business users
4. Coaching to work with teams on a specific use case/complex issue and provide recommendations for a faster, cost effective solutions
5. Offer certification and encourage continuous education for employees to keep up to date with latest
6. Feedback loop: get user feedback on training and user experience to improve future trainings

Talk by: Ellie Hajarian

Here’s more to explore: State of Data + AI Report: https://dbricks.co/44i2HBp The Data Team's Guide to the Databricks Lakehouse Platform: https://dbricks.co/46nuDpI

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Over the past two years, Providence has developed a robust streaming data platform (SDP) leveraging Databricks in Azure. The SDP enables us to ingest and process real-time data reflecting clinical operations across our 52 hospitals and roughly 1000 ambulatory clinics. The HL7 messages generated by Epic are parsed using Databricks in our secure cloud environment and used to generate an up-to-the minute picture of exactly what is happening at the point of care.

We are already leveraging this information to minimize hospital overcrowding and have been actively integrating AI/ML to accurately forecast future conditions (e.g., arrivals, length of stay, acuity, and discharge requirements.) This allows us to both improve resource utilization (e.g., nurse staffing levels) and to optimize patient throughput. The result is both improved patient care and operational efficiency.

In this session, we will share how these outcomes are only possible with the power and elegance afforded by our investments in Azure, Databricks, and increasingly Lakehouse. We will demonstrate Providence's blueprint for enabling real-time analytics which can be generalized to other healthcare providers.

Talk by: Lindsay Mico and Deylo Woo

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Most have experienced the frustration and disappointment of a flight delay or cancelation due to aircraft issues. The Collins Aerospace business unit at Raytheon Technologies is committed to redefining aerospace by using data to deliver a more reliable, sustainable, efficient, and enjoyable aviation industry.

Ascentia is a product example of this with focus on helping airlines make smarter and more sustainable decisions by anticipating aircraft maintenance issues in advance, leading to more reliable flight schedules and fewer delays. Over the past five years a variety of products from the Databricks technology suite were employed to achieve this. Leveraging cloud infrastructure and harnessing the Databricks Lakehouse, Apache Spark™ development, and Databricks’ dynamic platform, Collins has been able to accelerate development and deployment of predictive health monitoring (PHM) analytics to generate Ascentia’s aircraft maintenance recommendations.

American Airlines takes great pride in caring for customers travel, and recognize the importance of supporting the dignity and independence of everyone who travels with us. As we work to improve the customer experience, we're committed to making our airline more accessible to everyone. Our work to ensure that travel that is accessible to all is well underway. We have been particularly focused on making the journey smoother for customers who rely on wheelchairs or other mobility devices. We have implemented the use of a bag tag specifically for wheelchairs and scooters that gives team members more information, like the mobility device’s weight and battery type, or whether it needs to be returned to a customer before a connecting flight.

As a data engineering and analytics team, we at American Airlines are building a passenger service request data product that will provide timely insights on expected mobility device traffic at each airport so that the front-line team members can provide seamless travel experience to the passengers.

Talk by: Teja Tangeda and Madhan Venkatesan

Here’s more to explore: Why the Data Lakehouse Is Your next Data Warehouse: https://dbricks.co/3Pt5unq Lakehouse Fundamentals Training: https://dbricks.co/44ancQs

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Sensitive data sets can be encrypted directly by new Apache Spark™ versions (3.2 and higher). Setting several configuration parameters and DataFrame options will trigger the Apache Parquet modular encryption mechanism that protects select columns with column-specific keys. The upcoming Spark 3.4 version will also support uniform encryption, where all DataFrame columns are encrypted with the same key.

Spark data encryption is already leveraged by a number of companies to protect personal or business confidential data in their production environments. The main integration effort is focused on key access control and on building a Spark/Parquet plug-in code that can interact with company’s key management service (KMS).

In this session, we will briefly cover the basics of Spark/Parquet encryption usage, and dive into the details of encryption key management that will help in integrating this Spark data protection mechanism in your deployment. You will learn how to run a HelloWorld encryption sample, and how to extend it into a real world production code integrated with your organization’s KMS and access control policies. We will talk about the standard envelope encryption approach to big data protection, the performance-vs-security trade-offs between single and double envelope wrapping, internal and external key metadata storage. We will see a demo, and discuss the new features such as uniform encryption and two-tier management of encryption keys.

Talk by: Gidon Gershinsky

Here’s more to explore: Data, Analytics, and AI Governance: https://dbricks.co/44gu3YU

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

As one of the world's largest fast-food chains, McDonald's manages massive amounts of data for customers, sales, inventory, marketing, and more. And at that scale, ensuring the accuracy, reliability, and quality of all that data comes with a new set of complex challenges. Developing manual data quality checks with legacy tools was too time consuming and resource-intensive, requiring developer support and data domain expertise. Ultimately, they struggled to scale their checks across their enterprise data pipelines.

Join our featured customer session, where you’ll hear from Matt Sandler, Senior Director of Data and Analytics at McDonald’s, about how they use the Lightup Deep Data Quality platform to deploy pushdown data quality checks in minutes, not months — without developer support. From reactive to proactive, the McDonald’s data team leverages Lightup to scale their data quality checks across petabytes of data, ensuring high-quality data and reliable analytics for their products and services. During the session, you’ll learn:

• The key challenges of scaling Data Quality checks with legacy tools
• Why fixing data quality (fast) was critical to launching their new loyalty program and personalized marketing initiatives
• How quickly McDonald’s ramped up with Lightup, transforming their data quality struggles into success

After the session, you’ll understand:

• Why McDonald’s phased out their legacy Data Quality tools
• The benefits of using pushdown data quality checks, AI-powered anomaly detection, and incident alerts
• Best practices for scaling data quality checks in your own organization

Talk by: Matt Sandler and Manu Bansal

Here’s more to explore: Data, Analytics, and AI Governance: https://dbricks.co/44gu3YU

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Composable analytics is the next progression of business intelligence. We will discuss how current analytics rely on two key principles: composability and agility. Through modularizing our analytics capabilities, we can rapidly “compose” new data applications. An organization uses these building blocks to deliver customized analytics experiences at a customer level.

This session will orientate business intelligence leaders to composable data and analytics.

• How data teams can use composable analytics to decrease application development time.
• How an organization can leverage existing and new tools to maximize value-based, data-driven insights.
  • Requirements for effectively deploying composable analytics.
  • Utilizing no, low-code and high-code analytics capabilities.
  • Extracting full value from your customer data and metadata.
  • Leveraging analytics building blocks to create new products and revenue streams.

Talk by: Scott Castle

Here’s more to explore: Why the Data Lakehouse Is Your next Data Warehouse: https://dbricks.co/3Pt5unq Lakehouse Fundamentals Training: https://dbricks.co/44ancQs

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

This session begins with data warehouse trivia and lessons learned from production implementations of multicloud data architecture. You will learn to design future-proof low latency data systems that focus on openness and interoperability. You will also gain a gentle introduction to Cloud FinOps principles that can help your organization reduce compute spend and increase efficiency. 

Most enterprises today are multicloud. While an assortment of low-code connectors boasts the ability to make data available for analytics in real time, they post long-lasting challenges:

• Inefficient EDW targets
• Inability to evolve schema
• Forbiddingly expensive data exports due to cloud and vendor lock-in

The alternative is an open data lake that unifies batch and streaming workloads. Bronze landing zones in open format eliminate the data extraction costs required by proprietary EDW. Apache Spark™ Structured Streaming provides a unified ingestion interface. Streaming triggers allow us to switch back and forth between batch and stream with one-line code changes. Streaming aggregation enables us to incrementally compute on data that arrives near each other.

Specific examples are given on how to use Autoloader to discover newly arrived data and ensure exactly once, incremental processing. How DLT can be configured effectively to further simplify streaming jobs and accelerate the development cycle. How to apply SWE best practices to Workflows and integrate with popular Git providers, either using the Databricks Project or Databricks Terraform provider. 

Talk by: Christina Taylor

Here’s more to explore: Big Book of Data Engineering: 2nd Edition: https://dbricks.co/3XpPgNV The Data Team's Guide to the Databricks Lakehouse Platform: https://dbricks.co/46nuDpI

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Legacy data warehouses are costly to maintain, unscalable and cannot deliver on data science, ML and real-time analytics use cases. Migrating from your enterprise data warehouse to Databricks lets you scale as your business needs grow and accelerate innovation by running all your data, analytics and AI workloads on a single unified data platform.

In the first part of this session we will guide you through the well-designed process and tools that will help you from the assessment phase to the actual implementation of an EDW migration project. Also, we will address ways to convert PL/SQL proprietary code to an open standard python code and take advantage of PySpark for ETL workloads and Databricks SQL’s data analytics workload power.

The second part of this session will be based on an EDW migration project of SNCF (French national railways); one of the major enterprise customers of Databricks in France. Databricks partnered with SNCF to migrate its real estate entity from Oracle DW and IBM DataStage to Databricks on AWS. We will walk you through the customer context, urgency to migration, challenges, target architecture, nitty-gritty details of implementation, best practices, recommendations, and learnings in order to execute a successful migration project in a very accelerated time frame.

Talk by: Himanshu Arora and Amine Benhamza

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc

Geospatial data analysis is critical to understanding the impact of agricultural operations on environmental sustainability with respect to water quality, soil health, greenhouse gasses, and more. Outside of a few specialized software products, however, support for spatial data types is often limited or missing from analytics and visualization platforms. In this session, we show how Truterra is using Databricks, Apache Sedona, and R to analyze spatial data at scale. Additionally, learn how Truterra uses spatial insights to educate and promote practices that optimize profitability, sustainability, and stewardship outcomes at the farm.

In this session, you will see how Databricks and Apache Sedona are used to process large spatial datasets including field, watershed, and hydrologic boundaries. You will see dynamic widgets, SQL and R used in tandem to generate map visuals, display them, and enable download all from a Databricks notebook.

Talk by: Nara Khou and Cort Lunke

Connect with us: Website: https://databricks.com Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc Facebook: https://www.facebook.com/databricksinc