What is our primary use case?
My current use case is an analytical system for an oil and gas company. They have a data lake and they are getting their data from different IoTs and different SAP systems. They wanted to have some feedback about their system and check on their production.
The IoT center's data was actually our use case, then how we can use that data to prevent disaster from happening. Previously, this work was all done manually. If there is a leak in the pipes then it is a large area to search because there are hundreds of kilometers that have been run.
To tackle this problem, they use IoT sensor data. It helps them to understand how the flow is working at a specific point in time, what happens when they're drilling the oil and other types of data. To complete the analysis, we use an analytical system or a Teradata Azure system that we designed.
The ECL was designed in a manner where we were doing the transformation on Hadoop. It is more like a balanced system where you use Hadoop for what it is good for, and where you use Teradata for its purpose.
I have worked in several different environments. The largest Teradata implementation on Azure was done by me. Right now, I'm working with AWS but previously, I have implemented both on-premises and hybrid deployments.
What is most valuable?
Performance-wise, Teradata beats everything.
The flexibility in design is very good.
When you want to tune it, it is very easy to do. I have worked a lot as a DB and I know that when there is a query that is stuck for hours, and you want to tune it, the task is quite easy. You can use the indexes and architecture, and exploit the power of Teradata.
What needs improvement?
There are lots of improvements that Teradata is working on and many that have already been done. However, recently I was in a session with people who are working on the data sciences side of the business. The data science group is not using the Teradata modeling features. Rather, they are using third-party tools for the prediction and forecasting models that they are building. Teradata should focus on functionality for building predictive models because, in that regard, it can definitely improve.
In my experience with Teradata, I have seen their sales strategy and that needs to change. As an organization marketing a product, they need to be cloud-focused instead of the on-premises thinking that they currently have.
There are a few things where you are totally dependent on the customer service and support from Teradata. This is an area where they really need to improve a bit, especially when you're talking about cloud service integration. Depending on the environment you may need to contact the AWS team, or the Azure team, or the GCP team, or the Teradata team. In all of these environments, improvements need to be made.
For how long have I used the solution?
I have been working with Teradata for more than a decade, probably close to 11 years. I have been in the industry for approximately 18 years.
What do I think about the stability of the solution?
We are satisfied with the stability.
What do I think about the scalability of the solution?
It is quite easy to scale. For a cloud-based deployment, it's even simpler now.
Previously, it was easy if we just wanted to add power but there was an impediment when you had to add a node. Similarly, it was difficult when you had to shift to the newer nodes. Now, however, with the cloud, it's easy to do.
How was the initial setup?
For the on-premises deployment, a lot of people consider the initial setup to be difficult. Keep in mind, however, that Teradata is a million-dollar product, so they set it up for you.
For a cloud-based deployment, it is now pretty simple to deploy.
What's my experience with pricing, setup cost, and licensing?
The price of Teradata is on the higher side, and I think that it where they lose out on some of their business. The price should be reduced to make it more competitive.
Which other solutions did I evaluate?
One of my clients recently tried to switch solutions but they have rolled back to Teradata. The same thing happened with another company that I know of in Pakistan, where they tried working with an IBM product.
What other advice do I have?
My advice for anybody who is implementing this product is that they really need to hire the right people. Get people who actually have worked on Teradata to guide them.
You can buy Teradata and you can get it up and running, but until you hire a few good people, you are not going to get the full benefits from it. These people are not resources that you would ask Teradata to provide you. Rather, you can hire them from the market.
The important part is that you need to have people who have good experience with Teradata because they can truly exploit its power for you.
Overall, this is a good product. It is close to the best but I cannot rate it perfectly because of the price.
I would rate this solution a nine out of ten.
Which deployment model are you using for this solution?
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Amazon Web Services (AWS)