Join us on Thursday, June 6th, for a Silicon Valley Elastic User Group meetup with Dremio. Alex Merced (Developer Advocate at Dremio) and Sherry Ger (Principal Customer Enterprise Architect at Elastic) will give presentations, followed by networking, light bites, and refreshments.

Date and time:

Thursday, June 6th, from 5:30 - 7:30 pm PDT

Location:

Dremio Office: 3970 Freedom Cir #110, Santa Clara, CA 95054

Parking:

You will see CHEGG building to your left as you enter parking lot, 3970 will be on the right side of parking lot about 200 yards into lot kiddy corner to/and passed CHEGG.

Agenda:

• 5:30 pm: Doors open; say hi, grab a seat, and eat some food.
• 6:00 pm: Better Together: Elasticsearch and the Dremio Lakehouse - Alex Merced (Developer Advocate at Dremio)
• 6:30 pm: Building blazing fast customer search experience with Elasticsearch and self hosted Foundation Model - Sherry Ger (Principal Customer Enterprise Architect at Elastic)
• 7:00-7:30 pm: Networking & refreshments
• 7:30 pm: Event ends

Talk Abstracts:

Better Together: Elasticsearch and the Dremio Lakehouse - Alex Merced (Developer Advocate at Dremio)

The synergy between Elasticsearch and the Dremio Lakehouse emerges as a transformative approach for analytics. This talk, titled "Better Together: Elasticsearch and the Dremio Lakehouse," delves into the innovative integration of Elasticsearch as a dynamic source within Dremio’s platform. By harnessing Elasticsearch's powerful search capabilities alongside Dremio’s efficient data lakehouse architecture, organizations can achieve a seamless, high-performance analytics environment.

The integration supports Elasticsearch versions 6.x and 7.x, with Dremio. This collaboration enables users to plan and execute queries efficiently by leveraging Elasticsearch’s rich metadata, including dataset discovery and details for enhanced query planning and execution.

Furthermore, Dremio’s elastic connector provides a user-friendly perspective of Elasticsearch indices and mappings, simplifying data access and analysis.

Key highlights include exposing data through Elastic aliases, support for synthetic tables via wildcards, and the parallelization of queries to optimize data retrieval speeds.

The talk will cover Dremio’s list promotion rules, special handling for geoshape data types, and how it maintains mapping consistency through schema learning and automatic discovery of new fields. Additionally, we will explore the extensive pushdown capabilities that allow for efficient query execution directly in Elasticsearch, reducing data movement and accelerating insights.

This session aims to provide attendees with practical knowledge on configuring and optimizing the Elasticsearch source within Dremio, ensuring a unified and efficient analytical platform. We will also touch on the technical considerations for integrating Elasticsearch with Dremio, including configuration specifics, debugging, logging, and working with Elasticsearch and x-pack.

By the end of the talk, participants will be well-equipped to leverage the combined strengths of Elasticsearch and Dremio, creating a robust foundation for their data lakehouse analytics endeavors.

Building blazing fast customer search experience with Elasticsearch and self hosted Foundation Model - Sherry Ger (Principal Customer Enterprise Architect at Elastic)

While GenAI services have exploded in popularity, many users are hesitant to adopt them due to privacy, security concerns, and the high cost. Open source LLMs offer a compelling alternative, showcasing impressive capabilities.

In this talk, we will explore:

• Set up and run open source foundation model locally
• Use a foundation model and Elasticsearch to extract knowledge, build blazing fast retrieval, and perform data analytics from unstructured text.