Parallelizing SAIGE Across Hundreds of Cores
As population genetics datasets grow exponentially, it is becoming impractical to work with genetic data without leveraging Apache Spark™. There are many ways to use Spark to derive novel insights into the role of genetic variation on disease processes. For example, Regeneron works directly on Spark SQL DataFrames, and the open-source Hail package can be...
A Guide to Training Sessions at Spark + AI Summit, Europe
Education and the pursuit of knowledge are lifelong journeys: they never complete; there is always something new to learn; a new professional certification to add to your credit; a knowledge gap to fill. Training at Spark + AI Summit, Europe is not only about becoming an Apache Spark expert. Nor is it only about being...
Brand Safety with Structured Streaming, Delta Lake, and Databricks
The original blog is from Eyeview Engineering’s blog Brand Safety with Spark Streaming and Delta Lake reproduced with permission. Eyeview serves data-driven online video ads for its clients. Brands know the importance of ad placement and ensuring their ads are not placed next to unfavorable content. The Internet Advertising Bureau (IAB), defines brand safety as...
Making the Move to Amsterdam: Miles Yucht
While we are proud of our Berkeley roots, Databricks now calls many cities around the world our home. In addition to offices in London, Singapore, New York and our headquarters in San Francisco, we have one of our major engineering hubs in the fast-growing European Development Center, Amsterdam. Databricks offers the exciting opportunity to relocate...
A Guide to MLflow Talks at Spark + AI Summit 2019 Europe
We are thrilled to see how well MLflow has been welcomed by the community since we launched it last summer. With now over 800K monthly downloads, 130 code contributors and dozens of contributing organizations including RStudio and Microsoft, it is one of the fastest growing open source projects in the field of Machine Learning, confirming...
Diving Into Delta Lake: Schema Enforcement & Evolution
Think back to when you were in high school - so fresh and full of ideas. Since then, undoubtedly, both the world and the way you see things have changed in many ways, as you have gained new experiences. Data, like our experiences, is always evolving and accumulating. To keep up, our mental models of...
Engineering population scale Genome-Wide Association Studies with Apache Spark™, Delta Lake, and MLflow
The advent of genome-wide association studies (GWAS) in the late 2000s enabled scientists to begin to understand the causes of complex diseases such as diabetes and Crohn’s disease at their most fundamental level. However, academic bioinformatics tools to perform GWAS have not kept pace with the growth of genomic data, which has been doubling globally...
Productionizing Machine Learning: From Deployment to Drift Detection
Try this notebook to reproduce the steps outlined below and watch our on-demand webinar to learn more. In many articles and blogs the machine learning workflow starts with data prep and ends with deploying a model to production. But in reality, that's just the beginning of the lifecycle of a machine learning model. As they say,...
A Guide to AI, Data Science, Machine Learning and Deep Learning Talks at Spark+AI Summit Europe 2019
The Spark + AI Summit Europe is just around the corner, and it's a great opportunity for data scientists and Machine Learning (ML) practitioners to get up to speed on the latest tools and innovations in the field! Below is a selection of talks on ML best practices for productionizing ML at scale, real-life use...
Adventures in the TCP stack: Uncovering performance regressions in the TCP SACKs vulnerability fixes
Last month, we announced that the Databricks platform was experiencing network performance regressions due to Linux patches for the TCP SACKs vulnerabilities. The regressions were observed in less than 0.2% of cases when running the Databricks Runtime (DBR) on the Amazon Web Services (AWS) platform. In this post, we will dive deeper into our analysis that determined the TCP stack was the source of the degradation. We will discuss the symptoms we were seeing, walk through how we debugged the TCP connections, and explain the root cause in the Linux source. As a quick note before we jump in, Canonical is working on an Ubuntu 16.04 image that resolves these performance regressions. We plan to update the Databricks platform once that image is available and has passed our regression tests.
