Migrating Amazon Redshift to Autonomous Data Warehouse Cloud

“Big Data wins games but Data Warehousing wins championships” says Michael Jordan. Data Scientists create the algorithm, but as Todd Goldman says, if there is no data engineer to put it into production for use by the business, does it have any value?

If you google for Amazon Redshift vs Oracle, you will find lots of articles on how to migrate Oracle to Redshift. Is it worth it? Perhaps in some cases before Oracle Autonomous Data Warehouse Cloud existed.

Now, things look quite different. “Oracle Autonomous Data Warehouse processes data 8-14 times faster than AWS Redshift. In addition, Autonomous Data Warehouse Cloud costs 5 to 8x less than AWS Redshift. Oracle performs in an hour what Redshift does in 10 hours.” At least according to Oracle Autonomous Data Warehouse Cloud white paper. And I have nothing but great experiences with ADWC. For the past half an year or so.

But, what are the major issues and problems reported by Redshift users?

One of the most common complaints involves how Amazon Redshift handles large updates. In particular, the process of moving massive data sets across the internet requires substantial bandwidth. While Redshift is set up for high performance with large data sets, “there have been some reports of less than optimal performance,” for the largest data sets. An article by Alan R. Earls entitled Amazon Redshift review reveals quirks, frustrations claims that reviewers want more from the big data service. So:

Why to migrate from Amazon Redshift to Autonomous Data Warehouse Cloud?

1. Amazon Redshift is ranked 2nd in Cloud Data Warehouse with 14 reviews vs Oracle Exadata which is ranked 1st in Data Warehouse with 55 reviews.

The top reviewer of Amazon Redshift writes “It processes petabytes of data and supports many file formats. Restoring huge snapshots takes too long”. The top reviewer of Oracle Exadata writes “Thanks to smart scans, the amount of data transferred from storage to database nodes significantly decreases”.

2. Oracle Autonomous dominates in features and capabilities:

DB-engines shows an excellent system properties comparison of Amazon Redshift vs. Oracle.

In addition, reading through these thoughts on using Amazon Redshift as a replacement for an Oracle Data Warehouse can be worthwhile. It shows how Amazon Redshift compares with a more traditional DW approach. But Enterprises have some Redshift concerns, including:

– The difference between versions of PostgreSQL and the version Amazon uses with Redshift
– The scalability of very large data volume is limited and performance suffers
– The query interface is not modern, interface is a bit behind
– Redshift needs more flexibility to create user-defined functions
– Access to the underlying operating system and certain database functions and capabilities aren’t available
– Starting sizes may be too large for some use cases
– Redshift also resides in a single AWS availability zone

3. Amazon Redshift has several limitation: Limits in Amazon Redshift. On the other hand, you can hardly find a database feature not yet implemented by Oracle.

4. But the most important reason why to migrate to ADWC is that the Oracle Autonomous Database Cloud offers total automation based on machine learning and eliminates human labor, human error, and manual tuning.

How to migrate from Amazon Redshift to Autonomous Data Warehouse Cloud?

Use the SQL Developer Amazon Redshift Migration Assistant which is available with SQL Developer 17.4. It provides easy migration of Amazon Redshift environments on a per-schema basis.

Here are the 5 steps on how to migarte from Amazon Redshift to Autonomous Data Warehouse Cloud:

1. Connect to Amazon Redshift
2. Start the Cloud Migration Wizard
3. Review and Finish the Amazon Redshift Migration
4. Use the Generated Amazon Redshift Migration Scripts
5. Perform the Post Migration Tasks

Check out what Paul Way says about why Oracle thinks Autonomous IT can ultimately win the Cloud War.

Finally, here is what Amazon CTO Werner Vogels is saying: Our cloud offers any database you need. And I agree with him that a one size fits all database doesn’t fit anyone. But mission and business critical enterprise systems with huge requirements and resource needs deserve only the best.