Ippokratis Pandis

@Hippotas

Distinguished Engineer at Databricks. Mostly focused on large scale analytics and transaction processing. Tweets are personal opinions.

Menlo Park, CA pandis.net Joined April 2009

344 Following 2.5K Followers 1.4K Posts

Latest posts 5

Ippokratis Pandis @Hippotas May 29

I suspect that Marc had a similarly disappointing experience with me over the past week.

Marc Brooker @MarcJBrooker May 28

Today I'm thankful that my career success doesn't depend on three reviewers with strong opinions about my work, at least one of whom doesn't seem to have read it, and another who is startlingly ignorant about the state-of-the art.
View quoted post on X

Ippokratis Pandis @Hippotas Jan 22, 2025

History Mode for Zero-ETL.

When we announced the beginning of the Zero-ETL program 2 years ago (at ReInvent'22) the first question/feature request I got was whether there is an option to not be deleting in Redshift the records that get deleted in Aurora. Today we are excited to announce the launch of History Mode for Zero-ETL. History Mode allows past versions of records to remain visible in Redshift.

By preserving the complete history of data changes with History Mode, users can easily run advanced analytics on historical data, build look back reports, and perform trend analysis across multiple zero-ETL data sources, including Amazon Aurora, RDS and DynamoDB.

With History Mode the differentiating Zero-ETL capability of AWS is getting even better!

aws.amazon.com/about-aws/what…

aws.amazon.com Amazon Redshift announces support for History Mode for zero-ETL integrations - AWS Discover more about what's new at AWS with Amazon Redshift announces support for History Mode for zero-ETL integrations

Ippokratis Pandis @Hippotas Dec 5, 2024

@tim_kraska and his team have developed impressive NL2SQL technology and it is enabling Bedrock Knowledge Bases for structure data from the Lakehouse, including Redshift Storage, S3 and S3 Tables.

#nl2sql #kb #bedrock #structured #data #lakehouse #redshift #s3 #s3tables

Tim Kraska Verified account @tim_kraska Dec 4, 2024

Very proud to see Bedrock Knowledge Bases for structured data stored being launched at #reinvent24. Under the hood it uses our novel NL2SQL engine, which we developed over the last year, to make it easy to access structured data as part of Generative AI assistants.

View quoted post on X

Ippokratis Pandis @Hippotas Dec 5, 2024

It was great to be able to talk a bit about the next generation of Amazon Sagemaker that becomes the center of your Data, Analytics and AI needs, with enhancements such as the SageMaker Lakehouse and SageMaker Unified Studio.

You can watch our session at: youtu.be/80ImA4-zycw?si…

#amazon #sagemaker #lakehouse #unifiedstudio #reinvent24

Ippokratis Pandis @Hippotas Dec 4, 2024

Today we are excited to annouce the next generation of Amazon SageMaker, which becomes the center for all of your Data, Analytics and AI needs!

As part of that, we are announcing Amazon SageMaker Unified Studio (in preview) as well as SageMaker Lakehouse (in GA), which brings together the analytics-optimized capabilities of Redshift Managed Storage, as well as tables in S3 and the newly announced S3 Tables under one catalog, accessible through an Iceberg REST API and governed by powerful Fine-grained Access Controls (FGACs). This is Lakehouse re-imagined.

We are also announcing that we are enhancing our ZeroETL capabilities to SageMaker Lakehouse by adding support for Enterprise Applications, such as Salesforce, SAP, and ServiceNow.

#amazon #sagemaker #data #analytics #ai #unifiedstudio #lakehouse