cancel
Showing results for 
Search instead for 
Did you mean: 
Data Engineering
cancel
Showing results for 
Search instead for 
Did you mean: 

Migrating DB from 7.3 LTS to 9.1 LTS

KC_1205
New Contributor III

Hi All,

I have a code in the dev and production using DB 7.3 LTS. Now, I would like to update the environment to 9.1 LTS as support is going to finish. I have gone through the documentation given in the following link.

https://docs.databricks.com/release-notes/runtime/9.1.html#new-features-and-improvements

As I am new to the process, I would like to understand what procedural precautions to be taken during the migration process.

Thank you in advance for the help.

Regards

Kiran

1 ACCEPTED SOLUTION

Accepted Solutions

Ravi
Valued Contributor
Valued Contributor
9 REPLIES 9

Anonymous
Not applicable

Hi @Kiran Chalasani​! My name is Piper, and I'm a moderator for Databricks. Welcome! It's nice to meet you!

Thanks for your question. We will give your peers a chance to respond and then we'll circle back if we need to.

KC_1205
New Contributor III

Hi Piper

Thank you for welcoming me.

Ravi
Valued Contributor
Valued Contributor

@Kiran Chalasani​ You can start with the DBR 9.1 LTS migration guide and Apache Spark 3.1.2 migration guide.

https://docs.databricks.com/release-notes/runtime/9.1-migration.html#databricks-runtime-91-lts-migra...

https://spark.apache.org/docs/3.1.2/migration-guide.html

KC_1205
New Contributor III

Hi Ravi,

Thanks for the reply with the useful links. Correct me if I am wrong. I am thinking to create a new cluster and attach the job to it. During this process I will be choosing the new 9.1 LTS and it automatically picks up the Spark 3.1.2. and new Python 3.8.8 environment. Then run the code in dev to check the potential errors and fix them. is that right? or anything different?

Thanks in adavnce.

Kiran

Ravi
Valued Contributor
Valued Contributor

@Kiran Chalasani​ Yes, that is the right approach. Always test in dev before moving the changes in prod!

KC_1205
New Contributor III

Thank you Ravi.

KC_1205
New Contributor III

Hi Ravi,

So, If I update the DB to 9.1 LTS, NumPy version will be 1.19 and later, if I update to 1.21 in the notebooks. At cluster I have Spark version related to the 9.1 LTS which will support 1.19 and notebook will be 1.21. One should expect any compatibility issues? or any thing else to be take care? any suggestions?

Thank you

Kiran C

-werners-
Esteemed Contributor III

The guides Ravi mentioned are definitely a must read.

The migration from 7.3 to 9.1 is pretty transparent imo. Nothing compared to spark 2.x to 3.x f.e.

gmondauto
New Contributor II

@Kiran Chalasani​  Hey Have you ever been able to run 7.3run time with multi_gpus before you migrated to 9.1?

Welcome to Databricks Community: Lets learn, network and celebrate together

Join our fast-growing data practitioner and expert community of 80K+ members, ready to discover, help and collaborate together while making meaningful connections. 

Click here to register and join today! 

Engage in exciting technical discussions, join a group with your peers and meet our Featured Members.