What is our primary use case?
I am not using
Cognosys SQL Server currently, but in my past project, I used it for approximately one to 1.5 years.
In that project, we had several applications, and we used Cognosys SQL Server to load the operational and transactional data into this database. After that, we had a Snowflake layer as well, a Snowflake database. From Cognosys SQL Server, we pushed the data to our Snowflake database for the analytics use case. Technically, we used Cognosys SQL Server only to store the operational data and the transactional data, not for the analytics.
Because it is quite difficult to integrate Snowflake with our application due to some compatibility issues, the application we were using was not compatible with the Snowflake database. For that reason, we had to adopt a SQL Server database. Previously we evaluated Microsoft SQL Server as well, but the initial setup takes a lot of time, which is why we chose Cognosys SQL Server at that time. Once the operational and transactional data was stored in Cognosys SQL Server, we performed the test cases on top of that. Once that was done, we pushed the data to Snowflake database for the analytics purpose.
The integration was quite good because we did not face any issue integrating Cognosys SQL Server with Snowflake database. That was simply done by the APIs. Once the data was available in Snowflake database, we had some downstream applications or downstream reporting as well, which we built on Power BI, and that we used to showcase the data. On Cognosys SQL Server side, because in the previous project we had six to seven applications, before adopting Cognosys SQL Server, we fetched the data from the application and the transactional data. We maintained Excel reports and maintained the data in Excel, and then we pushed that data into Snowflake database. It was quite a complex task and a manual task. For that reason, we adopted Cognosys SQL Server to integrate all the applications, all the six to seven applications, with a single database only. There we could perform all our database operations before pushing the clean data into Snowflake.
One use case I already provided is as follows: when we did not have Cognosys SQL Server, we were fetching the transactional data directly from the application and doing the transformation and cleaning the data in Excel itself. Once our final data was ready in Excel, we loaded that Excel sheet or Excel data into Snowflake database. That work was fully manual, and we had many chances of error in that manual work. After adopting Cognosys SQL Server, we created a data pipeline in SSIS which pulls the data from the application and loads the data into Cognosys SQL Server. Then and there, we performed the transformation by using the stored procedure or by using the views. The chances of error have been reduced significantly. We were saving a lot of time because previously, for one of the Excel sheets, we were spending one and a half days to transform the data. After adopting Cognosys SQL Server, the transformation and the ETL part could be completed in 15-20 minutes.
We used it as a database only. We were fetching the data from the application using the SSIS tool and doing the transformation. At that time, we did not explore much of the capability or the AI capability of it. However, if they have really implemented some of the AI capability, then I would like to explore it more. Currently, I do not have anything to share regarding the AI capabilities.
What is most valuable?
Cognosys SQL Server is a Microsoft SQL Server, so whatever the features we have in Microsoft SQL Server it has provided, including the ACID properties, the transactional properties, and it is very scalable because you do not have to worry about the infrastructure. It is maintained or it is cloud-hosted. It is very scalable, so whenever we get a large amount of data, it is very easy to manage it in Cognosys SQL Server. Apart from that, we can take the backup. The backup mechanism is also very good in Cognosys SQL Server because you can schedule it as per your demand or as per the project demand. That is not manual; that is completely automated, so you do not have to worry about it. If at any time our database crashes or is damaged, it is very easy to recover it.
Because this is also a Microsoft SQL Server, before choosing Cognosys SQL Server, we evaluated Microsoft SQL Server as well. However, the problem was that for initial setup and regarding the installation, we have to own the complete activity. In the case of Cognosys SQL Server, we do not have to worry about it. It is easy to install or easy to set up. For that reason, we chose Cognosys SQL Server instead of the traditional Microsoft SQL Server.
What needs improvement?
When the data grows, every time we have to troubleshoot the query. It is a SQL Server. We have the dimension tables over there. Every time the data grows, our queries show some lag or we face some performance issues on the performance side. If they can work on the improvement parts so that whenever the data grows, the performance can be improved automatically on the database side, that would be a great feature. Otherwise, every time the data engineer will have to troubleshoot the SQL query, they will have to look into the table and create the proper indexing over there. That work is still manual. On the performance side, if they can work on it and that can be automated, that would be great.
If they can work on the integration part as well, because right now the market share of Cognosys SQL Server is very less compared to the other databases. As far as I know, it can be integrated with Power BI and Tableau, but not with all the data visualization tools. If they can work on the integration part and scale the integration to all the data visualization tools, that could be also useful.
For how long have I used the solution?
I am not using Cognosys SQL Server currently, but in my past project, I used it for approximately one to 1.5 years.
How are customer service and support?
In the previous project I used it, that was completely owned by our infrastructure team. Our infrastructure team purchased the licenses for different projects and they did the initial setup. Regarding the installation and the setup part, they got the full support from Cognosys support team. However, that was completely owned by the infrastructure team, not by us as a developer.
I know that during the initial setup and the installation process, our infrastructure team reached out to Cognosys support, and they got full support at that time regarding the setup. Regarding the setup only, we reached out to them and we got full support. After the setup, I think we did not face any issues.
Which solution did I use previously and why did I switch?
As I mentioned, previously we were using Excel only. We were fetching the transactional and the operational data from the application. We were storing it into the Excel sheet. Then we were doing the transformation and cleaning the data. Once the data was cleaned, we were loading that Excel data into Snowflake. That was completely manual. For example, one of the Excel sheets took one and a half days to clean the data, transform the data, and
finally load it into Snowflake database. After adopting Cognosys SQL Server, we created data pipelines in SSIS which pull the data from the application and load the data into Cognosys SQL Server. Then we had the stored procedures as well, which we were using to clean the data and transform the data. Finally, we were loading the data from the SQL Server reporting layer. Those pipelines were completely automated, and that could process the data within 15 to 20 minutes. We could see a lot of improvement in terms of time saved, as well as reduced errors. Because manual work requires many chances for error, that has been reduced significantly.
How was the initial setup?
Cognosys SQL Server is a Microsoft SQL Server, so whatever the features we have in Microsoft SQL Server it has provided, including the ACID properties, the transactional properties, and it is very scalable because you do not have to worry about the infrastructure. It is maintained or it is cloud-hosted. It is very scalable, so whenever we get a large amount of data, it is very easy to manage it in Cognosys SQL Server. Apart from that, we can take the backup. The backup mechanism is also very good in Cognosys SQL Server because you can schedule it as per your demand or as per the project demand. That is not manual; that is completely automated, so you do not have to worry about it. If at any time our database crashes or is damaged, it is very easy to recover it.
What other advice do I have?
Cognosys SQL Server is very scalable. So whenever our data grows and we have to onboard new sources or integrate new sources with Cognosys SQL Server, we do not have to worry about the storage or the performance. This is hosted on the cloud, and it can be scaled automatically.
One piece of advice I want to give is regarding the data size. If you are working on transactional data or operational data, then that can be a great fit for this. However, if you have a very large amount of data and on top of that, you work on machine learning or you have to work on big data, then instead of choosing Cognosys SQL Server, you can evaluate other databases as well, such as Databricks or Snowflake, because they are the perfect solution for analytics work. I would rate this product an 8 out of 10 based on my experience with it.
