My first Pass Summit on the V.20

After a great week in Seattle, for the Pass Summit v.20, I would like to share with you how the pass was going on, then my impression about the session I presented and the main trends I followed during the event.

How the pass is going on ?

The Pass Summit is the largest technical conference for Microsoft Data platform community. We get together to learn, share, meet expert and have some fun.

During 3 days, you are fully immersed at the Pass. From the breakfast to the diner.

You can follow conferences, from 8:00 AM to 6:00 PM. Each conference lasts one hour, with 15 minutes break between them. You have a dozen of them each time. If you don’t have enough, you can follow other conferences earlier with the breakfast’s conference (from 6:45 am) or during the lunch time.

You have many activities on offer next door: dinner, communities’s exchange, debates …

Another part of the Pass is the sponsor’s stand. You can’t miss it. A plethora of exhibitors are doing their best to catch your attention, showing their solution / technology and giving some gifts. It’s perfect place to exchange with experts and meet new people.

It is also a place where you have a lot of goodies for free, or almost. You just have to give your email address, and in exchange you receive some stickers, cup, bath duck … but I warn you, your mail box will be full the week after the Pass. I didn’t think about that in the moment, I was naive ~ “Jura, mais un peu tard, qu’on ne l’y prendrait plus” — La Fontaine.

My first Pass Summit as a Speaker

It was a great honor to be selected as a speaker at the Pass Summit, with Jonathan Petit, with this subject:

Azure Cosmos DB — Accelerate a Real-Time Big Data Solution

After presenting Cosmos DB main features, we demonstrated how to build an end to end Big Data Solution on Azure, from ingestion to visualization, through a concrete use case. You can see the architecture as follow:

It was very impressive for me, especially because I had never been to the Pass before. The first day was a real challenge to me, feeling very nervous, seeing such amazing people and experts giving their speech. Hopefully I could enjoy the conferences and learn as much as I could.

The second day wasn’t the same, I had to make mine at the beginning of the afternoon, so I preferred practicing over and over, to be prepared for the best. I know that practicing a lot just before your conference isn’t the best idea, but after this experience, I think I will feel less stressed out, especially because everything went well.

Through the feedback, we found that the audience like our presentation and I’m very proud of this first big experience.

I’m going to work on my accent, trying to “unfrench” it! It sometimes leads to misunderstandings from the audience. But this will provide me a good reason to travel and meet people!

If you want to know more about the architecture, don’t hesite to have a look on the presentation, over here.

A bucket of session

Building a Modern Data Warehouse on Azure

As you can see during the last few years, Azure is the new nuclear power of Microsoft. You can’t think of Microsoft without it now.

The same appears for Data Warehousing.

A lot of sessions during the Pass was on it, just some of them that I followed :

• Building a Modern Data Warehouse on Azure by Bill Gibson and Anand Subbaraj.
• Learnings and Best Practices in Building an Enterprise Data Warehouse Using Azure SQL DW by Mike Weiner, John Hoang and Alexei Khalyako.
• Implementing a Hybrid Data Warehouse by Johan Ludvig Brattås.

The architecture above is the new architecture provided by Microsoft for building data warehousing and transform your data into a real outcome. The MSBI tools changed for their contemporaries but the main features stay unchanged, just improved. You can now easily interact with Azure Databricks or Azure Blob storage for a better understanding and bring more impact from your data.

Once on Azure, you can interact with any module you want, and let your imagination bring you forward. Here is an example of one of the few architectures that you can build deeply.

A last word about something I just discovered there: Common Data Model (CDM). CDM simplifies data management and app development by unifying data into a known form and applying structural and semantic consistency across multiple apps and deployments.

Just have a look on the documentation.

Advanced Analytics

You’re not on the cloud, and you can’t use Azure Databricks to use your favorite machine learning algorithms, don’t worry ! Python and R integration are fully integrated on the SQL Server environment.

Available since the SQL Server 2016 release for R and 2017 for python, you can manage all your data life-cycle in one place, even truly with SQL Server 2019 with the integration of Spark and HDFS.

There are three ways of integrating your data with R or Python:

I followed a very interesting conference, giving by Enrico van de Laar: “Moving Advanced Analytics to Your SQL Server Databases”. In one hour you understand the different ways to use R in SQL Server 2017, with funny examples and all the material that you can be found on the conference page.

The expanse of Power BI

I followed different sessions about Power BI, all explained with passion (a common characteristic for people who are presenting this technology):

• Building Enterprise-Grade BI Models with Microsoft Power BI by Christian Wade.
• Introducing Advanced Data Preparation Using Power BI Dataflows by Cristian Petculescu, Roy Gal and Adi Regev.

The two conferences as followed were on two new features that I didn’t know before: using aggregations on Power BI Desktop and the new cloud module Power BI Dataflows.

It’s a perfect occasion to broaden your knowledge and ask questions directly to the expert.

Azure Cosmos DB

What would be the Pass Summit without a session about one of the most powerful database ? Of course I’m talking about Cosmos DB.

Speaking directly with the Azure Cosmos DB team was a real opportunity. They could help me with deeply issues I encountered, or if they didn’t have the solution by their own, giving me the right person to contact directly.

I followed two sessions, but the one given by Deborah Chen was very interesting. A perfect session to acquire a better understanding about modeling and indexing with Cosmos DB. You can find a good demo on her github.

Hoping I will have the chance to go back next year, and even more, try other Microsoft’s Experiences.