In terms of improvements for Denodo, regarding performance, in cases where there are multiple virtualizations—such as reading from one Denodo view that is virtualized, and from that view there's also virtualization, and another team is reading from that view—if multiple virtualizations happen with no caching in between, it becomes slow.

This occurs because it is cascading; whenever at the top level someone is reading data, that request is getting cascaded to the nth level, causing issues, especially in cases such as Power BI reports. We need to consider implementing some persistent layer in between.

The scaling process should improve because many things are getting automated. The scale-out part needs to be automated, though I am uncertain whether Denodo has already implemented that feature.

Denodo's complexity level is high since it connects multiple sources, sometimes creating glitches in the production environment. The stability of the features can be improved. The system has dependencies on other environments, like JVM, which can affect its performance.

As we are still in the initial phase, it's difficult to identify weaknesses. However, ensuring data caching is up to date is critical. While AI and ML integration is present, understanding its implementation better could be beneficial.

Denodo is like a Swiss army knife, but it's challenging to deploy it as a clear solution because it presents itself as a tool that can do anything. Denodo needs better communication on how the product can be deployed for specific solutions, and there are price issues, as it's considered expensive. It requires improving the story of how it can solve specific problems.

They need to invest more in the optimization engine. It will be a fantastic tool for them to enhance automation and improve the GUI interface, especially for Integration, as some tools are still not fully integrated, such as Iceberg. Batch processing is very good but not yet fully integrated. Therefore, they should focus on improving the GUI interface, enhancing cloud capabilities, and integrating modern technologies like Iceberg. Additionally, they should continue refining the optimization engine itself.

Data transformation must be improved. Denodo is not the best tool for data transformation.

I would like it if we could pull the data model or export the data model because Denodo has, based on how you build, something called associations. The data model gets very confusing when you go look at it. It depends on how you define it, but it's really huge. I wish there was a way to export it into Visio. It could be that they already have a way to do it, and I don't know, but it would be much easier for the architects and even for the developers to look at it than scrunching it into the screen and expanding every small portion of it. If there was an easy way to export the Denodo model into another modeling tool and view it, it would be great.

We are also trying to use Denodo as a cataloging tool so that we can put proper definitions, and once they are made available to our business users, it becomes easy. Using the Denodo layer, we can write queries to pull data. There is a virtualization layer. So, we can interact with it just like any other database. If there was a way to write a query and make my query public to other users so that they don't have to come in and write their queries, it would be great. There should be reusability of something that I did. For example, if I'm working in the healthcare industry and I want to get a view of the patient count for the whole year, I should be able to write a query and make it public so that whoever wants to use it can use that query instead of creating the same query again.

As a company, we first did a proof-of-concept for about four months to make sure the product was a perfect fit for us or not, and beyond that I have only used Denodo for another year or so, so I know that we haven't used the product to its fullest yet. Indeed, a lot of Denodo has changed since we had our first presentation on it with the Denodo sales team who gave us a rundown of all the features. Nevertheless, there are multiple ideas I could suggest in terms of improvement.

First of all, the visualization and reporting could be better. Of course, the data virtualization is good, but the data visualization could be improved with regards to the real-time dashboarding of the graphs, pie charts, etc. For the real-time data, the dashboard should preferably be updated automatically every hour. Let's say, as a CEO or CFO, I just want to know how much premium the company will get at any hour of the day. This data should be readily available on the dashboard. This is largely why we stick with Power BI's dashboarding features (besides the simple fact that Power BI works well hand-in-hand with Azure), and why we still haven't used Denodo's data visualization features as much as the data virtualization features. 

Another area we have been struggling with is the integration of Denodo with both Salesforce and MuleSoft, which we use to track the customers in our sales system, such as when sending insurance quotes. When we first tried to integrate Salesforce, we found that there was some type of version incompatibility. We had a hard time talking with Salesforce about this, but eventually we upgraded our version and the integration was resolved. Yet, this was a challenge that I feel we didn't need to go through, as we were not able to quickly map out the issue. And with MuleSoft, we have not been able to integrate it properly at all.

I have also seen our users complain about the availability of data sources, where they are sometimes not able to connect to all the sources they need. This kind of complaint, however, is difficult to diagnose, and I don't know for sure whether it is due to how we have Denodo set up in the company, or whether it's an actual issue with Denodo itself. These complaints were mainly made during the first few months of our usage, so it is possible that the problems stemmed from a lack of knowledge on how to use Denodo correctly, especially since the individual feedback would generally be something vague like, "Okay, I'm not able to do this". To help resolve these complaints, I would suggest that Denodo work on better documentation and perhaps some kind of virtual training.

For example, there's an insurance software company called Guidewire, and when we first brought Guidewire into the company they sent us a lot of training videos even before the actual integration took place. For Denodo, it would be beneficial to make sure that the team that will be using it has some kind of training on how to use the product at least a month beforehand, and there could even be some kind of feedback or Q&A sessions to go along with the training. If Denodo were able to provide this kind of training, it would be very helpful to users in insurance and banking companies because the staff are typically older and not always technically-minded. They say, "You are pushing us too hard", so they need encouragement when it comes to adoption of a new software product.

The cache configuration is more complicated.

Sometimes, Windows-related functions do not work properly in Denodo. The analytic functions in SQL do not work properly. The aforementioned area in the product needs to be taken care of for improvements.

Tasks such as conversion of a date format or conversion of a number format that can be done in a very easy way in different languages, like SQL or Oracle, are not so easy to do in Denodo.

For example, if you want to convert a date from one format to another, in Oracle it's pretty easy; in Denodo, however, it requires so many lines of code. Simple things that can be done very quickly in other database languages require more lines of code in Denodo.

The scheduler UI in 8.0 is not as good as that in 7.0. We have thousands of jobs, and if you search one job, get it, and click on it, when you come back, your search will be gone. You have to do your search all over again. The 7.0 UI was much better.

It's not really scalable. We have a couple of problems with a high demand for the servers. It also doesn't have very good integration with Git, so it's not possible to work with branches with the developers. When it goes to production, then we cannot use it anymore.

The git configuration really should be improved. It would help a lot. Then the developers could have different values of the same project and use, for example, GitFlow for the projects. That would be amazing. If they had it, the scalability would be very good. And we can actually have one external metadata database for starting the Denodo configuration. However, it's very slow. 

If it were possible to work around Kubernetes, for example, it would help. They have some implementations. However, it doesn't really work as expected.

The solution is expensive as you are charged by the CPU. 

There are a couple of areas that can be improved in Denodo. From a stability point of view, sometimes we see issues in the data management functionality. This only happens now and then, however, and usually takes place when we add in our own customization.

There are also certain limitations on customizing Denodo, in general. It would be better if Denodo provided more mechanisms for users to develop custom products where they could easily build in their own logic with automated means. In the case of complex customization, we will usually use Spring Boot instead of Denodo, especially when we have tons of data in production and we need to segregate it based on certain logic. Otherwise, when it's a matter of minimal data points that are required, we will simply build it as a web service instead of writing the same logic in Spring Boot.

A feature that we have wanted since we started with Denodo is to have a fallback option. After we migrated all our web services to Denodo, it would have been nice to have a ad-hoc fallback option where if we ever do want to use something else, that option is available. For example, something where we have those built-in read-only views which we can reuse and, without too much time or fuss, build a web service on top of that by simply plugging in views, details, or any other part of the Denodo platform architecture.

I think Denodo needs a bit of work with performance management. When data sets are huge and a heavy query has been pushed out to the database, it can take quite some time to get what you need without clear optimization techniques. It requires clearer instructions. Some features should be added on the performance management and optimization side of things. I'd like to see them provide more on data governance. In addition, it would be good to have some support for analytical queries in SQL as that can be a challenge at times. There is currently no advanced analytic support for the SQLs.

User-specific security at the column and row levels needs to be improved. Instead of applying security at every individual level, it would be better if it were at the group or tier level. It will save a lot of time.

Denodo could also improve the UI, data governance, and cataloging.

The feature that you have to connect on LDAP needs improvement.

I am not sure if it's on Denodo's side, but sometimes it is very difficult to connect on LDAP, and you can take advantage of this by using the roles that have already been created there.

I have no complaints about Denodo's features, except perhaps the price.

Monitoring event logs can be improved. In the older version, there was a monitoring schedule to get event reports and properly audit the reports. In the newer version, it is not there, and we have to manually configure data and audit events.

Denodo could create a UI for us to monitor the events. They could also provide a good training environment or basic training on how to create a view. At present, we have been creating our own material for training. It would be better if Denodo could provide that training material for beginners.

I'm waiting for them to introduce the generative AI capabilities they mentioned would be available by the end of the year. That would be a game-changer.

Moreover, I would like to see the ability to utilize vector databases. 

For some internal strategies, I need to publish data into vector databases. Denodo currently integrates with ChatGPT, but the ability to manage and utilize them directly within Denodo would be a significant improvement.

There are a few cosmetic changes that would make Denodo better, including offering holistic governance. It would be beneficial if you could see what inquiries are going around Denodo on the remote system, including who has logged in and which one is using which part of the deviant spaces. 

Denodo has provided several modules with their scheduler that also have room for improvement. 

When thinking about future releases, Denodo should consider their platform availability because our administrator has seen some challenges in this regard. It requires some optimization on the developer side as well. 

Denodo would benefit from having a more user-friendly view. It would be good if the solution provided a much-needed cellular platform.

I would like to see a proper way to avoid killing the sourcing systems. There is some parameterization in this case if you push down the query to some systems, but I think there are still issues with giving this tool to end users because they could write queries that could cause a lot of issues in the operation systems.

We can't scale it to meet digital requirements.

Denodo has some difficulty supporting large numbers of records. Its error reporting is also unreliable and doesn't give clear information on the problem. We've also recently had an issue with roles getting deleted.

I would like to see a connectivity option with third-party apps, for example, JDBC, and ODBC drivers. Currently, we need to install it separately from the Denodo side and then connect it. If they provided an option directly in Denodo to connect to these Power BI, and Tableau, Denodo would be even stronger. Microsoft provides SharePoint, and they have this option where you can directly connect to Power BI, and you can directly connect to the visualization tools. If Denodo can do this, then it would be great.

Denodo's training documentation could be improved by providing more material. From an administrative standpoint, I've found that only Denodo websites provide the usual tutorials. It may be because it's a bit of a restricted tool, but it results in trouble with learning. Normally, I can find help and solutions from other sources, but I haven't been able to find any for Denodo. Other that, it's fine and it performs well. I only have six months of experience, so I can't accurately suggest improvements. 

I am still waiting for Denodo to support vector databases. Currently, Denodo does not provide a vector database, which my customers build on their own to use AI functionalities.

The solution is slow when there are many virtualization layers. The solution is also slow when we connect the on-premise solution with the one on the cloud through a network gateway.

Denodo can improve usage management-related aspects. If you deal with the mini views, it gets stuck. The performance is very slow when we go with a large number of views and high volume.

In a future release of Denodo, they could add more features, such as remote tables and metadata for views. 

The dropdown menus feel antiquated to me, and the administrative portals need improvement.

In the next release, I would like to see a better user experience and the interface needs modernization.

There have been some issues when you are at a table. Currently, Denodo exports data sets for a tabular model. When you are finished modeling your database or data warehouse they export a link to be used in Tableau. They should support other tools like Power BI. 

The support is not the best and should be improved.

In the future, I would like to see a different structure for how they manage the data sources and connectors. Having a better data lineage tool would also be an improvement.

Better audit capabilities would be a useful enhancement.

We would like this solution to be more universally user-friendly. At present it is really only aimed at IT specialists.

In our country, we do have to get the training and do the end-to-end delivery, and that can be a bit hectic when it comes to customer systems development. We need a bit more external support from the company. Support has been a real issue for us.

We occasionally have some integration issues that we need to work through.

In terms of features, et cetera, I don't know if there is anything really missing from that perspective. We are also new to this product. We have to analyze it and other products and that process often takes some time. It's not fair to comment on what's missing without surveying the entire landscape.

Documentation needs to be improved, they need to lift their game. Finding information is difficult. I'm doing the training and they provide courses but it could be improved. 

I haven't tested the solution for long enough to be able to suggest any additional features. 

The integration could use improvement, it's a lot of non-speed line processes that we have discovered, in the country. The configurations could use a lot more improvement. 

The solution should have its own acceleration technology.

The connectivity is not so great. It uses standard protocols to connect to the data center, it could improve the interface in terms of automation. The connection should be easier for the end-user. At the moment it's a bit too complex. You need a lot of documentation.

The data catalog certainly has room for improvement. It is functional but we look forward to development. We are in constant contact with Denodo and they are fully aware of our needs.

I'd like to see an improvement to data catalog functionality which would immensely increase the value of this product.

In our organization, when we have to analyze a lot of data, the possibility to represent the very meaning of data to the analysts, which is actually the functionality of data catalog, is very important. Denodo is very good at creating unified access to data. It's not that good yet, in the integration of the semantics of the data, in determining what the data means. In a statistics organization, the integration with a metadata system is extremely important. That's the reason why we have found a way to integrate our metadata system with Denodo data catalog. It's easy to imagine much more functionality and we know that Denodo is fully aware of that. Denodo is improving its data catalog functionality as a result. It's not to say that we are not satisfied with the existing one but we'd like to see improvements.

I'd like to see integrations with AWS, GCP and the like.