Our company uses one of the latest containerized versions of Vertica. When the curated processes are completed, our company uses Vertica at the end of all pipelines. In our organization, we usually store the curated datasets in parquet files, but we offer Vertica storage access for data consumers to manipulate and query the data easily. Our company also stores data in the Vertica database for analytics. 

In one of our organization's prior investigation reports on Vertica, it was highlighted that the solution was executed quite quickly due to its columnar storage underground, which is the most valuable feature for our company.

Vertica is fast enough in data copying processes. The data digestion process is quickly completed using a certain copy command. The solution gets accelerated in processes if the queries are properly designed. 

I have previously worked with Microsoft Secret Server and Deep Secure. I was familiar with the system's assets and transaction habits, so when I switched to Vertica, I had some expectations in terms of features from the other systems, which didn't quite match. The transactional operations and rollback were missing in Vertica, which I am trying to implement using the AUTOCOMMIT setting, but I am unsure whether the features will work according to my expectations. 

I have been using Vertica for three months. 

I would rate the stability a nine out of ten. 

In our company, we have faced difficulties in scaling the solution for certain use cases. The solution didn't scale as expected in some cases, for which our company team witnessed performance issues. But in our company, we had doubts in the aforementioned cases if it was a query optimization or scaling problem. There are around 100 Vertica users in our organization. I would rate the scalability a seven out of ten. 

I would rate the technical support an eight out of ten. 

Positive

I have previously used Redshift, MySQL, PostgreSQL and Microsoft Secret Server. 

When I was dealing with an onboarding task for the solution to implement a specific data pipeline with Vertica, I had to ideally and precisely configure the infrastructure and data architecture so that Vertica worked well enough in integration with Spark. If you have multiple systems and you want to establish communication between them, some drivers need to be provided, and some installations need to be made using JAR files. 

For the deployment of Vertica, no client agents are required because it only involves passing the secret code to Vertica and executing the solution using Airflow DAGs, Vertica operators, Vertica hook and Spark solution, which work satisfyingly together. After processing datasets in our company, we store them in Vertica directly using Spark and Spark connector for Vertica. Our company uses Spark as a computational engine. Our company's operational team mainly takes care of Vertica's maintenance. 

It's an expensive product. I would rate the pricing a five out of ten. 

In one of the projects in our company, we dealt with a huge amount of customer data, and the configuration of this data was quite specific as it arrived from multiple business units. Our customers can have contact with different business units, and each unit is a company in itself. The same customer might pass, for example, one shopping mall of a business unit and might shift to another unit and identify themselves in both business units depending upon the identification process of how the data is collected. 

The aforementioned project required identifying a unique user or customer with contacts with multiple business units and commercial centers. Our company processed a huge amount of the mentioned data type with the help of data science and machine learning models. Using the clustering process and leveraging the identified cluster users stored in the Vertica database, we executed some basic analytics on transactions involving customer spend, frequency of visits to a commercial center, and types of product selection. 

I found the swapped partition of tables option to be extremely useful for our company which allowed me to perform some quick operations that would be otherwise impossible to guarantee the rollback in operations involving all assets. I would overall rate the solution an eight out of ten. I am unsure if my company would stick with Vertica in the future and it also depends on the vendor's promises on Vertica's scaling capacity.