What needs improvement with Oracle GoldenGate?

Microservices Architecture is not user-friendly. It would have been great if the product would have stayed with the classic architecture and expanded into Microservices. However, it has completely moved away to Microservices. We can't use the same commands or scripts we had on the classic architecture. We have to redo everything.

In the future, I would appreciate a more user-friendly monitoring interface for easier management and troubleshooting. I believe enhancing the monitoring tools would make GoldenGate even more efficient.

Monitoring must be a bit more enhanced.

Having a graphical interface would greatly enhance the tool. A clean and user-friendly graphical interface based on common design principles would be beneficial. While the current interface is functional, incorporating a more visually intuitive design, perhaps inspired by widely used interfaces, would make it even more straightforward. Looking ahead to future releases, there are specific features I'd appreciate having. One such feature is the incorporation of pipelines, similar to those in Java, for seamless integration. Additionally, configuring for the Amazon Cloud is currently not straightforward; it often requires adding Python code to customize integrations. I would like to see an enhancement in this aspect, making the configuration process more straightforward and less reliant on intricate coding, particularly during meetings or collaborative sessions.

Oracle GoldenGate hangs for large tables. You cannot put everything in Golden Gate because it's licensed to hold systems. If you connect to another system with 20 CPUs, Golden Gate would need 20 more CPUs. You cannot put everything in Golden Gate for the license and large models. For that reason, we use Golden Gate and DB Link together. In some cases, we prefer to use DB Link; in others, Oracle GoldenGate.

The major problem in Oracle GoldenGate is the monitoring part. If you need to use the monitoring feature of Oracle GoldenGate, you need to pay a lot, meaning it is very expensive to use the plug-ins from Oracle, so sometimes a user needs to avoid using it. A user directly needs to use the scripts to monitor the integration, which works perfectly. Sometimes, a user finds it more difficult to discover the problem, though solving the problem is easy. It is difficult to discover what was the problem with the integration using Oracle GoldenGate since the logs were not very clear. With Oracle GoldenGate, the documentation was not very clear, even for the parts related to the table or DDL. In general, Oracle GoldenGate can't deliver good feedback from the logs.

Regarding improvements, the focus should be on addressing data quality issues and minimizing delays, particularly in Oracle databases. The technical support should be improved as well. Additionally, in terms of features, it would be great if the solution could support APIs.

The GoldenGate license is very costly.

IBM CDC has one central control while Oracle GoldenGate has two controls. In IBM CDC we can do all things in frontend. The solution needs to incorporate a feature where we can connect it to a standby database.

The solution's performance needs improvement. It could be faster.

The solution needs to improve its latency, monitoring and support.

The product lacks some features and it's expensive. I'd like to see a repair table function and a simplified installation process and configuration.

My only issue with this solution is that it doesn't become a full-scale ETL tool.

It's pretty close to perfect. There are no areas we see that are in need of improvement. It is not very easy to use, however, for what it does, it's not too difficult to manage in general. The solution costs a lot.

The main problem of Oracle GoldenGate is that sometimes, the processes get abended. Data consistency is also challenging. For example, my team's problem with using Oracle GoldenGate is that occasionally, you'll notice missing rows on the table, and it's hard to find that data. In the future, an additional feature I want to see in Oracle GoldenGate is the microservices feature.

What needs improvement in Oracle GoldenGate is the support. Another feature that needs to be improved in the solution is its GUI. It would be useful for programmers or users to be able to work from the GUI, not just from the command line. Simplifying how Oracle GoldenGate is used would also make the solution better.

The front-end management isn't very good. Its pricing and licensing can also be better.

The problem with GoldenGate is it is very complex to use. You need heavyweight skills to use it. When there's a problem, you have to delve quite deeply down to this sort of data layer where the problem is.

Troubleshooting issues is not always the easiest (can't always find info on errors online or at MyOracleSupport). Oracle's professional support can also be improved (take a long time to resond to ticket, when they do it is obvious they don't read the issue before responding; they frequently ask for info already provided in the ticket.) GoldenGate for big data is a separately licensed product. Regarding big data, I've found multiple prop file parameters not documented, not in the sample programs, yet were required.

There are some DDLs that have problems with GoldenGate. These issues detract from stability and it is something that should be improved. I would like to see the interface more graphical and easier to use. I like Microsoft products, for example, because they are easy to use. When I work in SQL Server, which is integrated with Oracle GoldenGate, I find it better to work in than Oracle.

Right now, the solution is working quite well for us as it is. The solution is quite costly, and if they could reduce their pricing, it would be more economical to implement.

While we like the CDC feature, it has some limitations in terms of what we can do and how to configure what we need to do. We are trying to fix those issues by implementing some workarounds. Memory consumption is also an issue in OGG architecture. Each OGG data replicator consumes more and less 200 MB of RAM. We have 10 OGG hubs, each one with almost 300 data replicators to the central database. We need 64 GB for each Hub. a total of 640 GB just for data replication.

It should be easier to integrate this solution with non-Oracle databases, including cloud-based solutions hosted on Azure.

I use an on-premises installation of GoldenGate — a standard GoldenGate installation — to do our test deployment for our projects. Part of the Oracle solution that needs some improvement is the ability to test features extensively before purchase and deployment. I am from Egypt and the economy is not so good. To create proper test environments for complex enterprise deployments is very expensive. Right now there is no place in Egypt that has enough servers to provide a proper test environment for a large deployment like the one we have planned. I do my best to do the test here with what I have. It is a poor model to test all the GoldenGate features in an improper environment because a valid test of the features of GoldenGate may require a huge investment in accessories like memory and processors. I still need to do the testing. I need to write the code to test out the deployment without a proper environment and without using some features that may be valuable in achieving our goals. I am going to join a project to perform testing with Oracle Exadata in hopes to have free equipment to test this with. In this way, I hope to have a good enough environment to test the product. The project will be huge will be much larger in the future. It could be that I will find I need to test out the cloud as a potential solution. I think I need one month or two months to decide what is the solution. But assistance from Oracle in temporary testing facilities would be useful and create the opportunity for investment.

Monitoring needs improvement. Right now we lose a lot of of time and effort. To have a real monitoring system is the most important thing. If you replicate between systems 24/7, you need a strong monitoring tool. It's not taken from a lot of effort and a lot of support and involvement with Nexus adapters to plug into our Oracle enterprise manager. Scalability also needs to be addressed. There's not a straightforward way to handle it. The solution, in general, should be easier to use, with less need to perform workarounds within the system.

We are having a challenge in that we want to replicate different types of data, from one database to another, that have different types of elements. The process of transforming the data between two different databases is inefficient and needs to be improved. Technical support for this solution could be improved. The integration is not straightforward and should be made simpler.

I find the user interface to be difficult for non-technical people. It was suggested to me that I use the command prompt, which is fine for me because I am a technical person. However, when showing others, such as managers, how to use the product to provide details such as the system status, it is a bit difficult. The inclusion of monitoring and management tools would be very nice. The Oracle documentation is very complex, and some of the documents are very long. I prefer to use sites that are on the topic of Oracle, but in fact, are not run by Oracle itself. In our experience, there is zero technical support for this solution.

The User Interface should be improved in newer versions, but without putting any restrictions on the customization capabilities. Additional performance metrics should be added and visible in a dashboard, displayed in a detailed manner. The licensing model should be improved to be more cost-effective for the end users.

I would like it to be simplified, easier to manage.