Teradata has a few AI models, but in data science, we need more flexibility. We can’t be limited to what's pre-built in the database. Typically, data science projects require experimenting with different models, so the limitation is that Teradata only has basic machine learning models in its database. Data science requires more advanced modeling, and you always want to search for the best possible approach. Combining the capabilities of Teradata with custom data science models will take time to mature, but it shows promise.

Teradata needs to promote it more. If they're the first to introduce things like in-database AI, they should really focus on promoting that. I haven't heard much about it, but maybe that's because the environment I’ve been working in recently has been mostly open-source. I’ve been doing applied research and freelance work that didn’t rely on robust vendor products, so I never got a chance to compare Teradata to others. I have heard about Databricks, though.

Teradata is somewhat late in adopting cloud technology. They need to focus on the adoption of cloud to remain competitive and target customers who prefer not to invest in capital expenditures and seek a more flexible, operational expenditure approach.

Replication needs improvement because we currently use DataMover once daily. Unlike SQL and Oracle, which have in-built replication capabilities, we don't have similar functionality with Teradata. 

Additionally, when backups are performed, it locks the database, preventing user access, which is a concern. Also, we need extra servers to perform backups, unlike SQL, which has embedded backup capabilities.

Teradata is an expensive tool. Like, if you're already using Microsoft products like Windows, they'll market all their products together. And with the rise of cloud technologies, companies will adopt solutions that offer them some privileges or facilities. 

Similar to how SAP does it in the market, so do Microsoft and other companies.  Even Oracle and other such tools are quite commonly seen compared to Teradata's competitors in everyday solutions.

The primary challenge with Teradata lies in its cost structure, encompassing subscription fees, software licenses, and hardware expenses. The management of these pricing components can be notably high. I believe there's room for improvement and investment in Teradata's ETL engine, making it more competitive with tools like IBM DataStage. Considering the growing importance of big data ecosystems, it could benefit from enhanced compatibility with platforms like Cloudera and tools like Apache Spark. It's essential to bridge the gaps and make Teradata's tools more accessible and user-friendly in the evolving landscape of data virtualization and analytics.

In our experience with Teradata, we haven't seen any significant weaknesses. However, when it comes to scalability, limited interest and success in some areas make us hesitate about upgrading. In the future, I would like to see Teradata incorporate features that allow it to handle unstructured data. Additionally, improvements in documentation to make it more user-friendly and up-to-date would enhance the overall user experience.

South Africa must significantly improve its internet infrastructure to reduce its high costs. I believe South Africa is one of the most expensive countries in the world when it comes to obtaining high-speed internet connections. We need to establish different service level agreements with clients to ensure we have backup lines and redundancy. The crucial question we need to ask ourselves is, how reliable do we want our internet connection to be? In the case of using cloud services, it's essential to have another line as a backup in case the primary one fails. In our situation, having a backup line is definitely necessary, but it also adds to the expenses. For a major bank like ours, relying on a single connection to the cloud is not feasible; we must have a second line in place to ensure uninterrupted service in case of any issues with the primary line.

The cost of Teradata Cloud Data Warehouse has room for improvement.

I don't have the exact details of the improvement needed in the performance of Teradata IntelliFlex because I work with the infrastructure department. The infrastructure department is in charge of getting infrastructure for the system to work, but before having Teradata, the data processing was taking a lot longer than it is now. We got into Teradata because we needed fast processing of huge amounts of data since, in the past, we were trying to achieve it with Oracle databases, and it wasn't good enough. In the last two years, the processing times have reduced by a third or a fourth of the time without causing any problems because of the efficiency of the Teradata engine.

The only issue our company has with Teradata IntelliFlex is that it is not cost-effective because of the way the product has been designed. We have Teradata IntelliFlex on-premises because we pay for a lot of muscle that we are not using. The problem we're seeing is that when we try to move into Teradata IntelliFlex's cloud services, we notice that Teradata doesn't have a flexible price structure to allow you to use a fraction of IntelliFlex's extension and charge just for the use of the solution. We receive messages from Teradata asking us to pay for the services we use on-premises, while in reality we use less than 30 percent of the machine's capacity.

There is something that is not working in favor of the business models for IntelliFlex when you plan to move to the cloud.

The tool's flexibility and capacity for expansion are areas of concern where improvements are required.

Azure Synapse SQL has evolved from a solely dedicated support tool to a data lake. It can store data from multiple systems, not just traditional database management systems. On the other hand, Teradata has limitations in loading flat files or unstructured data directly into its warehouse. In Azure Synapse SQL, we can implement machine learning using Python scripts. Additionally, Azure Synapse SQL offers advanced analytical capabilities compared to Teradata. Teradata is also expensive. 

The setup is not straightforward.

I would like Teradata to get involved in South Africa. They aren't doing aggressive promotion of Teradata systems in South Africa. They really should.

The solution’s pricing, scalability, and technical support response time could be improved.

The current operational approach needs improvement. In the past, relying on infrastructure data staging provided flexibility in collaborating with any database provider. However, transitioning to Informatica and integrating it with various data services proved beneficial for serving multiple clients. Despite its advantages, this approach incurred substantial costs in terms of hardware and technical expertise. To address this, there is a need to develop and offer free content and workshops, providing guidelines and technical knowledge for efficient data operation.

Sometimes the large ingestion jobs takes days to load data, and some of our stored procedures take longer. I don't know whether it's the tool or the network, but it takes a long time to execute complex commands.

Teradata can improve the way it handles big data and unstructured data.

One issue we've identified is that its performance varies across different on-premises setups and doesn't consistently meet our standards. This limitation necessitates the purchase of Teradata's proprietary hardware to ensure the desired level of performance. The capability to implement it with comparable performance across various private cloud environments, ensuring adaptability to different infrastructure setups would be beneficial.

Teradata hardly supports unstructured data or semi-structured data. Teradata is very costly, and it's not easy to upgrade or expand the components in Teradata. A lot of infrastructure needs to be set up in order to use Teradata.

The market is enhancing, and every day, BigQuery, Snowflake, Redshift, and all these competitors are moving very, very fast. They're bringing out many new features. For example, Snowflake now has Time Travel and cloning and all these very interesting features. Teradata is an old data warehouse, and they're not improving in terms of new, innovative features. Basically, they're pretty much consistent and unchanging. They're not bringing new features. I've been working with it for eight years, and I have not seen so many new features. Maybe there have been only one or two in the last eight years. Innovation is something that is a bit slow with respect to Teradata. Other competitors are better in terms of new innovations. I'd like to see Teradata add features like cloning or time travel.

There is a query tool. However, we need to install the tool in order to fetch records. With other data warehouses, they have a web-based tool that allows for queries without having to download.

The initial setup is not too easy.

When we want to scale, we need to go to Teradata to initiate the process. 

Some of our clients are looking for on-premise installations as well. Although we don't have any, some of our prospects are also asking, and we are not sure if that part is easily doable or is as effective. We haven't tried on-premises, so we don't know how good it is. We are not confident about proposing on-premise to them, since we are more familiar with the cloud. Maybe some documentation on how on-prem works or what other things to look for in on-premise deployment would be helpful. We'd love to see more tutorials. 

Our company does data science and AI machine learning algorithms on the data that is deciding on the Teradata, so that's the value we add. From my perspective, it would be good if they gave better ITIN/R plugins to use the data for AI modeling, or data science modeling. We can do it now; however, it could be more elegant in terms of interfacing.

Teradata is good, but eventually, my company is moving to a higher-level system. You wouldn't want to work on a lower-level system.

As for the areas for improvement in Teradata, since I was working on the very basic, legacy systems, the memory thing was always a challenge. If Teradata is moving to the cloud, the space constraint or the memory issue that my company generally faces will eventually resolve, in time.

What I'd like to see in the next release of Teradata is that it becomes full-fledged on the cloud, apart from better connectivity to various systems. For example, if I have to read or include a Python script, if I write some basic codes, I should be able to read even unstructured data. I know that it's not supported even in Snowflake, but at least semi-structured data support, if that can be a little more enhanced, that would be good.

I'd like to see a better relationship with the business itself. Business users have trouble finding solutions when they run into issues in the data warehouse.

It can be a bit complex to work with.

Regarding data warehousing, I'd like to see more data and business analysis to help with the design and development. 

It's primarily designed for big projects and therefore, the pricing is pretty high. It's not suitable for smaller companies. 

Any value this product provides is offset by the high cost and lack of scalability. We're moving to Snowflake because of the scalability and cost differentials. There are better, more flexible, robust, and economical options out there.

Teradata's UI could be more user-friendly.

If I want to implement an upgrade, I'd like to see how it will be different. Ideally, Data Lab should help me test production items and also do future things. Future releases should be downloadable and testable in Data Lab.

Teradata is a bit late for the cloud. They maybe need to have some room for cloud improvements.

Scalability is limited on-premises. 

The solution could use more AI support.

Apart from Control-M, it would be nice if it could integrate with other tools. Specifically, with respect to the integrations with the automation tools, if that can be increased, and if the tester could be able to test more automation scripts in a better way it would help. The efficiency could always be increased.

Occasionally, at the outset of the implementation process, we struggled a little bit.

I've been using the same UI for 20 years in Teradata. It could use some updating. Adding more stability around Teradata Studio would be outstanding. Teradata Studio is a Java-based version of their tool. It's much better now, but it still has some room for improvement.

An area that needs improvement is the cloud-based version of the solution because it's designed for on-prem and then transferred to the cloud version. Teradata needs to pay attention to the cloud-based solution to make sure it runs smoothly. In the next release, I would like Teradata to be cloud-agnostic in order to not limit customers' choices.

Teradata could improve by providing better training. For example, Oracle provides good overall training. It is not easy to find information on the internet.

In terms of improvements, I can only speak from the dashboard perspective. 

The most important thing is that I don't know how far the dashboard integration goes. People, for example, were using the Cloud Data Warehousing product, however, they were actually building their own dashboards, et cetera. That's like double the amount of work. The company should have the ability to integrate with something like Tableau to make everything more standardized.  

With the customer data, we were able to integrate very successfully with Tableau and we were able to use it well from an end-user perspective. We were using the Tableau interface for understanding what our customer buys, et cetera. For example, for the customer financial data aspect of it, the business data, we were able to use Tableau dashboards. It would be better to use back-end data warehousing and also have dashboard reporting and have everything integrated properly. It doesn't need to be Tableau. It can be some other dashboard. However, it needs to be standardized. 

The usability could be a bit better. It could be a bit more user-friendly.

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.

I'm not sure about the unstructured data management capabilities. It could be improved.

They should, I think, try to do more of this open-source initiative to encourage more people to use it. Otherwise, it's my opinion that they might lose our business in the long run.

I think that a couple of things can be improved with this product. One is, of course, that there should be an option for using it on the Cloud. A lot of customers ask me how they can get up and running fast, and this would be the fastest way with the least client effort. I know Teradata already has its own Cloud solution, but I think that is one area that needs more maturation and exploration. We need to do more testing before we begin to recommend it to clients and we can do this because we work directly with Teradata. We are using the Cloud solution and it is more scalable and also the costs are fairly elastic. It should be something we are using more with clients in the future.  

At an organization level, I think sometimes the support for development is not forthcoming. It should become more of a part of their ongoing process. If they dedicate more resources to developing along with the latest trends in the industry, Teradata could then bring that functionality out for the current customers and maybe attract other customers as well.  

I would say that the company should consider more capabilities in handling unstructured data in a better way. Right now most of the latest solutions that are coming to the market address the need for how to take unstructured data and create a standard map for that. I think that would really help. Right now, I only see most of the data through Teradata with its unstructured processing and some of it is not available. They can do more with this. I think Teradata's mappings are good and very easy to use. I would use it for customer databases as it is. But there still some ways that the handling of unstructured data could be improved and that would really help.  

Teradata's UI could be improved.

Teradata's pricing is quite high compared to Redshift, Synapse, or GCP alternatives. In the next release, I would like Teradata to refine their Object Store to be more compatible with cloud environments.

Teradata needs to expand the kind of training that's available to customers. Teradata only offers training directly and doesn't delegate to any third-party companies. As a result, it's harder to find people trained on Teradata in our market relative to Oracle.

They should diversify their sales channels and develop more resellers. Teradata could also be more active in marketing because I rarely see any Teradata events in Russia. They've always had several regional conferences for Teradata users, but I hardly ever see marketing events geared toward new customers.

The scalability could be better. The on-premises solution is always more complicated to scale.

The flexibility could be improved. The product is good if you know what you're planning to do. If you're designing as well, upfront, then it's fine. However, if you've made a mistake in the design work or your system, then it could be quite complex and expensive, and it may not preform well.

I would like to see more integration with many different types of data.

It could use some more advanced analytics relating to structured and semi-structured data. In other words, there should be a process that makes advanced analytics easier.

The integration capabilities could be improved.

They should improve their marketing to make it easier to sell to customers.

It is hard for some of our users to set up rules for cleansing and transforming data, so this is something that could be improved. This is necessary because the data that comes from different sources has different formatting, and sometimes there is duplication. This means that some degree of transformation is required.

I would like to see more options for visualization so that the data can be better presented to management.

To run our reports I have to upload data from past tables and it is always a problem, so there is a lot of room for improvement there. It just struggles when I have to load data and I would like to see better file processing power.

I have seen instability in automation when working with very large datasets.

Technical support needs improvement.

What really needs to improve, is their general cloud services. That is their biggest downfall at the moment. 

They should add more connectors to different platforms. 

It could use a more powerful CPU and a faster IO. 

The reporting side wasn't very good in the past, but with the latest versions, it's getting better. Still, the friendliness of the PDC reporting and functionality needs to be improved. 

The user interface needs to be improved.

The price needs to be more competitive as Hadoop, Redshift, Snowflake, etc are constantly making way into EDW space.

I would like to see an improved Knowledge Base on the web.

I would like more security and speed. These days, everyone is not just moving to cloud, but they are moving to clout. In a multi-cloud environment, you have to be cognizant of data transfer, bandwidth, and egress charges, not just from a performance and scalability point of view, but also how to keep costs low.

Teradata wants to come out of the hardware business and to run its own cloud on other public clouds. I think that is a great way to go. However, there are challenges to doing this. There are multiple layers of security and caching. E.g., if you have a lot of compute or a caching that will offload a lot of compute activity because it's already there, it will save its cycles for something which needs to be done. This way, you are offloading your compute, so you can go with a lesser capacity than a higher capacity. That is what this appliance does.  If they are moving off of the appliance, I am curious to see if their software will withstand the change, or will Teradata still run but need a bigger box?

I would like more security and speed. These days, everyone is not just moving to cloud, but they are moving to clout. In a multi-cloud environment, you have to be cognizant of data transfer, bandwidth, and egress charges, not just from a performance and scalability point of view, but also how to keep costs low.

• A more transparent integration with Aster solutions.
• An even deeper focus in data sciences.
• A teaching web site with more training on third-party tools used for BI (data sciences).

We are using an older version at the moment. We are looking for a more flexible cost model for the next version that we use, whether it be cloud or on-premise.

There is a need to improve performance in high transaction processes, as well as the reporting system.

Query language and its functionality are rather limited, compared to Oracle or even SQL Server. However, it is possible to perform any kind of logic in it (though some workarounds may be required).

Data ingestion is done via external utilities and not by the query language itself. It would be more convenient to have that functionality within its SQL dialect.

Compatibility with more Big Data platforms.

• Cost
• Scalability

It would help to make scaling easier with a reduced cost.

• Cloud implementation
• User-friendly products

I think the UI is not there yet. It could be improved by being more user-friendly.

• Licensing
• Architecture openness
• Integration with other tools

GUI of administrative tools is really outdated.

There is some improvement required on OLTP level and some analytical function is missing.

Cost, but Teradata is currently making improvements in this area.

It should be compatible and free on CLOUD/ AWS. Currently many DB's on AWS are free , Teradata is also on Amazon Web Services, but its charged and no documentation available to practice those LABS .

The cloud is the new challenge and the new opportunity.

Query Log maintenance should be free in all versions and should be made easy.

Data synchronization to the DR site.

My company’s management has to decide whether to invest in this area. This is more an internal issue at the company and definitely not a problem with the Teradata product. If we end up migrating to another platform, this will be the reason.

The Query Response time needs improving.

The increasing volumes of data demand more and more performance. I truly believe that in five years, we will breach the effective limits of Teradata to perform in a multi-petabyte environment, i.e., without significant changes to its underlying architecture. If this occurs, we will be forced to migrate to a more traditional platform (such as IBM Informix).

The SQL Assistant is very basic. This tool can be improved for usability.

Concurrency is still a challenge, pricing is still high.

I'd like to see better processing on a transactional level and star schema optimization.

Teradata could improve by being less complicated. There are some aspects that are not available on the Unix server and a Unix system is required to access some data, such as in case of an emergency.

The solution could improve by having a cloud version or a cloud component. We have to use other solutions, such as Amazon AWS, Microsoft Azure, or Snowflake for the cloud.

The solution overall needs improvement in its stability, support and pricing.