Teradata Vantage Base (DIY)

Positive

My main use case for Teradata  is mostly for data warehousing and analytics. Teradata  is used to store customer data; specifically, accounting transactions from multiple sources. The data is kept for analytical purposes for users. Since the data is very large-scale, an MPP architecture RDBMS  was needed to store it. Hence, Teradata was chosen.

Teradata's MPP feature allowed the team to handle huge data sets on the scale of petabytes and perform analytics on those in near real-time when users queried for that data. Teradata has positively impacted the organization because it was selected after extensive market research to identify the best tool for data warehousing. Teradata helped to scale better as the initial minimal data set started increasing rapidly. The scaling capabilities of Teradata really solved that problem.

There is nothing much on the improvement side that I wish was different or better for Teradata. Nothing comes to mind regarding improvements. I give it a nine because I still believe there is always room for improvement, especially when it comes to enhanced availability and possibly better performance and agility. There are no improvements needed for Teradata that have not been mentioned yet.

I have been working in my current field for almost 15 years.

Teradata is stable.

Teradata was very scalable, and that was one reason why it was chosen. Compared to a cloud-native tool, Teradata's scalability was good, and it did scale well.

Teradata's customer support was excellent.

Positive

Before Teradata, Redshift was used initially, mostly due to cost constraints and flexibility. The move was made to be mostly cloud-agnostic.

After implementing Teradata, there was not noticeable latency. I would not say it was very performance efficient, but it was definitely good. Performance was really good, but I could not compare it with anything else because I have not extensively worked on any other data warehousing system. Teradata is mostly what I worked on, so it was really good.

I was not directly involved in pricing, but there was a team handling that. Overall, I think it was good or on par with expectations, otherwise the finance team would not have approved it.

I have definitely seen a return on investment because with any other tool, more manpower would have been needed to maintain it.

Redshift was evaluated before choosing Teradata.

My advice for others looking into using Teradata is to do homework; there is no silver bullet for your problem. You have to find out what works well for your specific needs. If the priority is to find the best tool that can be relied on with less maintenance and good performance, Teradata is definitely a good option, but again, homework must be done to see which one suits the data workloads. I give this product a rating of 9 out of 10.

I have been using Teradata for more than one or two years.

My main use case for Teradata is creating tables, views, and procedures for our applications, along with creating BTEQ scripts to schedule some pipelines.

A specific example of how I use Teradata in one of my projects is for a large retail chain that wants to understand the customer's buying pattern across different regions to improve their store sales and inventory planning, where we have used Teradata for data integration, collecting data from multiple source systems such as POS systems, online transactions, and CRM, handling billions of transactions efficiently with advanced analytical queries.

BTEQ is the best feature of Teradata, as it allows me to execute multiple queries such as select, insert, update, and delete in a single script, handle errors, automate running multiple SQL statements, and schedule ETL jobs, making it lightweight and scriptable, unlike SQL Assistant.

Parallelism helps my team mainly for scripting and automation in projects; for instance, we utilize Teradata's parallel processing engine to execute queries across multiple model processors simultaneously, making data migration more efficient by splitting large datasets into smaller files and running multiple BTEQ scripts in a parallel fashion.

One challenge I have faced regarding the main use case is the integration with AI, as Teradata does not have the AI models that other OLAP systems such as BigQuery provide, making it difficult for us to give proper recommendations without using different tools for AI integration.

Teradata can be improved on the cloud side by integrating with some OLAP features and providing capabilities similar to Delta Lake for storing semi-structured, structured, and unstructured data.

I want to highlight two features for improvement: first, storing data in various formats without requiring a tabular structure, accommodating unstructured data; and second, adding AI ML features to better integrate Gen AI, LLM concepts, and user-friendly experiences such as text-to-SQL capabilities.

I have been working in my current field for almost more than four years.

Teradata is stable in my experience.

Teradata's scalability is good.

Teradata's customer support is good.

Neutral

I have not used any other solutions before Teradata.

I purchased Teradata through the AWS marketplace and believe it would benefit from easier documentation for deployment, as it is somewhat complicated for new users.

My experience with pricing, setup cost, and licensing of Teradata is almost good, although the cost is slightly high, but they provide good features.

Before choosing Teradata, we evaluated other options such as Databricks, which we preferred due to their strong AI ML features and options such as Delta Lake.

My advice for others considering using Teradata is that if their use case involves daily transactions with commands such as insert, update, or delete, they should go ahead and use Teradata, as many companies in retail, CPG, and banking find it effective for those operations compared to other platforms. I rate this product 7.5 out of 10.

There are two major use cases for Teradata . One is that whatever data we cleanse and aggregate, we push it to Teradata  for our business users. We create some ETL pipelines and automate them. The second use case is for data wrangling. Whatever data we publish to Teradata is used for various analyses, various SQLs, and a lot of dashboards sit on top of Teradata.

My most recent project was for inferring the Net Promoter  Score for one of the largest Australian banks where I used Teradata for ETL and data analysis. The entire cleaned data of the bank was stored in Teradata, wherein we had eight to ten different datasets coming in from different sources that were aggregated or converged into Teradata. Using that data, we developed certain business rules on top of that aggregated dataset, which was further fed into Tableau that sat on top of Teradata. Using that data, we were able to infer the customer Net Promoter  Score for a rolling six-week average.

The first thing that I appreciate about Teradata is its multi-parallel processing. Whatever queries we execute on Teradata, they are blazingly fast, so it offers really fast connectivity. Secondly, it also provides the MultiLoad feature, by which I can upload my Excels directly to Teradata or CSVs to analyze the data. The third feature is the QUALIFY or ROW_NUMBER keywords that I really appreciate about Teradata. The fourth thing is the way Teradata stores data in a columnar format for faster query processing, which is also one of the best features.

The multi-parallel processing and fast query execution of Teradata have benefited me and my team greatly. What really happens is that we store multiple copies of the data, one in Teradata and the other in our HDFS or object storage. When we had to query the data from the object storage, it was really slow, but when we discovered that this dataset is also available in Teradata, it was really fast, especially related to the NPS  project that I was discussing. That is probably one of the use cases that I can recall.

Teradata has positively impacted my organization since its inception. Earlier, whatever data we used to house was in HDFS, then we migrated to cloud, and now we are using Teradata, but Teradata has also moved to cloud. Teradata has immensely helped our organization to fetch the data at a faster rate, which has saved us quite a lot of time. That is probably the very best thing about Teradata.

Teradata could be improved by having a web interface that can really help users to plug and play. Right now, what is required is that I have to install a desktop app for Teradata and then set up the connections. If the same thing were available in a web interface, that would be really helpful.

Since I started my career, I have been using Teradata. It has been more than seven years that I have been using Teradata.

Teradata is stable.

The scalability of Teradata is really great. Whenever we need more resources, we can add that in Teradata, and when not needed, we can scale it down as well. All in all, it is very good.

I have not used other solutions personally, but I have seen a use case of Redshift getting used earlier in my current organization. I have also seen the use of object storage prior to using Teradata fully. What I can see now is that we are moving away from object storage because we want faster results, which is why we use Teradata.

I have seen a return on investment through time saved, specifically saving fifteen to twenty percent of the time.

At least fifteen to twenty percent of our time has been saved using Teradata, which has positively affected team productivity and business outcomes.

Before choosing Teradata, I evaluated other options such as Snowflake  and Redshift. These were some of the options available.

My advice to others looking into using Teradata is to go for it if you need faster processing, multi-parallel processing, or more security. I would rate this product an eight out of ten.