DATA Pill #010 - MLflow on GCP, the Modern Data Stack is dead and trends in software development.

Hi everyone,

It’s still Monday and I'm back with the best news from the Big Data world!

I would be grateful for your feedback.?

Please don't hesitate to get in touch.

Let’s start!

ARTICLES?

An Open-Source Tool to Change Data Validation As You Know It | 8 min read | Data Validation | Madison Schott | Towards Data Science

If you’ve ever had to migrate your data warehouse to a new platform or account, then you know how time-consuming and painful this can be. Recently, I’ve had to migrate my data from one Snowflake account to another, reconfiguring all of the data ingestion pipelines and orchestration. You quickly realize how one little change can (and will) break everything in your data pipeline.

Deploying MLflow on the Google Cloud Platform using App Engine | 12 min read | Cloud | ??Marcin Zab?ocki | GetInData Blog

Read the step-by-step guide which will help you to deploy MLflow instances on the Google Cloud Platform using App Engine. In the article, Marcin Zab?ocki has described how to:

• Pre-configure OAuth 2.0 Client
• Build the docker image for MLflow on App Engine
• Prepare the Terraform inputs

How Airbnb Safeguards Changes in Production | 8 min read | Software Engineering | Michael LIn | Airbnb Tech Blog

With the statistical methods in place to evaluate business metrics in near real-time, we can now detect problems that were invisible to Spinnaker, or required too much lead time to rely on traditional ERF experiments.

{ MORE LINKS }

NEWS?

Trends in Software Development 2022? | 11 min read | Andrzej Frydryszak | ITMagination Blog?

Some of the fifteen most impactful trends in 2022:

• Observability is crucial
• It’s good practice to use both the “serverless” & “serverful” approaches
• Containerize everything! Kubernetes is a hot technology!
• Everything can be done “as a Service”
• A great long-term hiring strategy is to hire juniors and train them

{ MORE LINKS }

DISCUSS?

Modern Data Stack is Dead? | 4 min read

Lauren Balik argues that the Modern Data Stack is already dying, that this is a flawed concept and should be replaced with the “Postmodern Data Stack” that she defines… Do you agree? We advise you to go through the comments.?

TUTORIALS?

Multicloud reporting and analytics using Google Cloud SQL and Power BI | 7 min read | Google Cloud | Matthew Smith? | Google Cloud Blog??

The following guide demonstrates the key steps to configuring Power BI reporting from Cloud SQL.?

?

DATAtube

7 Jupyter architectures for 7 different organizations | 49 min | GetInData

As ML engineers, you often work on providing the Jupyter environment for Data Science teams, so you probably know that providing a platform that is both flexible and cost-effective is a challenge. In this video, you can learn about the different Jupyter setups, their pros and cons and listen to the lessons we learned.?

?

PODCAST

Why and When to Use Kubeflow for MLOps ?| 58? min | MLOps | ML Community

Kubeflow is an excellent platform if your team is already leveraging Kubernetes and allows for a truly collaborative experience. In this episode, Ryan Russon talks about the pros and cons of using Kubeflow in your MLOps.

{ MORE LINKS }

?

CONFS AND MEETUPS

Today, we would like to share the second part of an assessment of one of them with you.??

A Review of the Big Data Technology Warsaw Summit 2022! Part 2. Top 3 best-rated presentations |? 11 min read | ?? Micha? Rudko & Mariusz Strzelecki | GetInData Blog?

Furthermore,? we'd like to invite you to some of upcoming events:

Meetup #10: Service Mesh, GKE and Cloud Native applications | Google Cloud Warsaw | 23 July | Warsaw

{ MORE LINKS }

—------------------------------------

That’s all for now!

See you next week ??

Adam Kawa from GetInData