Talend Open Studio [EOL] Reviews

We are using Talend Open Studio for building data pipelines in order to move data from source to the warehouse. My customers are using Talend Open Studio primarily in the retail industry.

We are not using the MDM, data quality, or data governance features in Talend Open Studio. We are just using it from an ETL and orchestration perspective.

We are using Talend Open Studio for jobs orchestration because we need to schedule the job. These jobs would be running after a particular data load. We have scheduled it using Talend Open Studio itself.

Talend Open Studio acts as an ETL tool. The ETL feature gives us the flexibility to build data pipelines the way we want. It also comes with a console which helps us to monitor the jobs we have built in, making that monitoring part easy. Additionally, it is scalable. Since Talend Open Studio has moved to Click and merged with Talend Cloud, the scalable option makes it easier for us to work on it.

The biggest benefit of Talend Open Studio for my customers is that it is a one-stop shop for orchestrating ETL and governance. When I say governance, it is more from a monitoring perspective.

There should be a dedicated support team for Talend Open Studio. We do have a support team, but at times it is critical for us to fix certain things within one or two hours. Usually, the support team takes some 12 hours, which is the standard edition of support. If we could have round-the-clock support, we would be able to resolve many issues which we encounter during the development part.

The initial setup for Talend Open Studio is a little complex. We need to have a dedicated person who has an understanding of Talend Open Studio as a tool. It takes some level of expertise, so it is not an easy process, but once we have a dedicated associate who is knowledgeable in that particular field, it becomes comparatively easier.

I have been working with Talend Open Studio for more than three years.

I have not faced any issues or complaints regarding stability in Talend Open Studio.

Since we have not used any other tool for doing the ETL or orchestrating, this becomes a tricky question. I cannot comment on whether my customers have seen value in money or return on investment with Talend Open Studio because unless we explore other tools, we will not be able to determine if there is a return on investment.

There are various technologies in the market which can do the things that Talend Open Studio is doing. If the focus is on orchestrating data pipelines and building logic, then Talend Open Studio is a good choice of tool. The key consideration should be the licensing cost that it comes with, and when we compare it with other available tools or open sources in the market, that becomes a challenge. Nonetheless, if the intent is to create data pipelines, build logic within an ETL, and also do orchestrating, then Talend Open Studio would definitely be one of the choice tools for me.

Nothing comes to mind regarding pre-built connectors in Talend Open Studio. The development team may be working on it, but I have not been close to the tool itself and its features.

Initially, there were some challenges with Talend Open Studio regarding issues which occurred because the tool was reaching its end of expiry. Since it has moved to Click Talend, that is Talend Open Studio on Cloud, we are not seeing those issues anymore. Certain functionalities which were not working before have been resolved. When developers used to open Talend Open Studio, they used to see some errors or certain things used to get stuck, but now it is not happening. Those were the issues I am talking about from six months ago.

My customers are using Talend Open Studio on the Talend-provided cloud.

On a scale of 1-10, I rate Talend Open Studio an 8.

We have customers who embed Talend Open Studio in their products like an OEM.

You can use Talend SDK to develop your own components and add them to the product.

The solution should integrate with a version control system in the subscription versions to make it easy to work with and manage the version control. Talend Open Studio is based on the Eclipse platform. Eclipse has a plug-in that you should install to compile it with Open Studio. Then, you will have access to work with version controls like Git and Bitbucket.

The solution could include visualization of the data, which is not available in the solution's subscription version.

The solution's performance and stability depend on the limitations of Java virtual machines and how you develop a Java program. Since it's auto-generated, it has better code-writing quality. Anyhow, you should evaluate the performance and the risks or failure and develop it accordingly. In Java coding, for example, you can write a program with a maximum of 65,000 lines of code. This is a Java limitation that developers should keep in mind while developing the program.

Sometimes, scalability is part of planning. It depends on what you mean by scalability. People talk a lot about it, but scalability is not always about system functionality. Sometimes, it may be planning the job you're doing. If you want to split it into several jobs or servers, you don't actually have to have it built in as a functionality. You can create a job using a loop, which runs and controls several jobs in a loop that may be controlled.

Scaling should not always be part of the infrastructure based on whether the engine can scale or not. I think it's your plan or project that should scale and split, and you can define these parameters. These parameters include how many servers you want to run or how many executions you want to do on different parts of the data. It's not always an issue of the engine running.

Sometimes, your database should be configured to support partitioning. The product may scale very well without partitioning, but if the basic response is very slow, you didn't solve the problem. You should solve the problems at a higher level, not just at the execution level. They should be solved at the database level and communication level, and you should have firewalls.

We are trying to add to the open source the ability to generate code for containers and Kubernetes that exist in the subscription version. Once you do this, Kubernetes will take care of the scaling, so there is no problem.

When you ask questions, the support team asks you additional questions to be more specific. It's a little complex to send all the logs. They guide you, but it's an interactive process. The support team should ensure we get a faster and better understanding of the issue. The technical support team should provide guidance on downloading log files for some issues.

Positive

The solution's initial setup is not complex. You should understand ETL like a Java or Python developer who works with scripts. You need to have knowledge of how to plan complex things and write code.

The solution's pricing depends on the case. In the past, the pricing was based on the number of users and not on the data volumes or the number of CPUs, like other products. From this perspective, Talend Open Studio was cheaper. The solution will be more expensive if you have a low data volume and a large number of developers.

Since Qlik purchased Talend, they've brought new pricing methods. The tool has now become more flexible in coping with the number of customers, developers, and volumes.

Most customers use the solution on-premises. However, it can be deployed anywhere in the cloud with the Java virtual machine environment. You can execute everything because you get the compiled executable of your programs that can be executed with any JavaScript. Java can be executed everywhere because the current version includes all the connectors. You don't have to install anything.

If you want to run a job with other products like Informatica, you need to install all the connectors for SAP, Salesforce, applications, databases, etc., on the server. With Talend Open Studio, the binary executive will already include all the connectors and the user passwords. It's already included and bound into Java, and we just need a Java virtual machine to run anywhere.

Some customers don't want to invest in an ETL and data integration infrastructure. I know at least one customer with huge data volumes using Talend data integration. We have another customer with huge volumes of data using Talend as an OEM, and it's running on Cloudera. Using the solution depends on the clients, the version, and the environment.

Our framework is based on code writing. There is a lot of software out there that you have to write code to develop. Many software products are not an integration tool but a code platform. They claim you must have ELT in the database to make it faster. While it is true in some cases, it isn't true in other cases. You need both ETL and ELT for good performance in a solution.

Talend Open Studio can do almost everything like a subscription platform. The problem is that it does not support version control and has its own operational cost. It was moving to subscription platforms to save time since time is money. People are using open source to save licenses, but the license is not the most important part. I suggest our customers move to subscriptions and have a more cooperative environment.

Overall, I rate the solution an eight out of ten.

It used to be Talend Open Studio, and I've got a paid version that's fully supported by Talend. Now, it's called Talend Cloud. Talend Open Studio parts are exactly the same as the open-source version. But with the cloud platform, when you build a job or an API, you publish it onto the cloud platform, and that manages all the API functionality really well. 

The best part is that if something goes wrong with an error or an issue, the logging is phenomenal. It tells you exactly what's wrong. That, coupled with Talend Open Studio, makes it really easy to resolve any issues. 

Talend Open Studio itself is fantastic. When you build a job or an API, it tells you what's wrong, what needs to be fixed, and how to fix it. Plus, Talend does the support documentation, and it's really straightforward to follow.

There's plenty of videos online on how to build jobs to transform data, build APIs, and link to other databases. Talend can handle quite major transformations and ETLs. So it's really simple and easy to follow, with all the documentation being intuitive. 

We use probably about 10% of Talend's functionality. But the primary use is keeping multiple databases (near 200 different databases) in sync. Talend integrates that information with REST APIs. We're moving a lot of our integration into the Talend platform, taking it from old XML-based integration to REST APIs. 

We use Crunchy Data's Crunchy Bridge solution for a lot of our databases in the back end, and Talend, coupled with Postgres, creates a very simple platform for us to build integrations really fast. It also means that we're using intelligence there on change events that occur in the database to keep multiple repositories in sync. 

We're moving towards more of a single source of truth platform rather than multiple sources of truth that we've been dealing with for the last fifteen or so years.

Data transformation has proven to be most effective for data management. That's amazing. We have a lot of external parties that gather information for us. We're a global government council, so we're transforming XML, JSON, comma-separated files, emails—every format that you can deliver data to us. GeoJSON, geometry, and spatial data—we're all feeding it through the Talend platform, and we're bringing it all into a single format from Postgres. 

It's just so simple to be able to take those foreign formats and transform them into the standard format that we have in-house. Likewise, we can take data from our database and transform it back into those other formats so that external parties can use their existing platforms without involving a human to do any data set in between. It's just made a huge improvement in how fast we can get data out and also consume data for the scientists that we have internally.

The error handling capability in Talend is really what sold us on the product. You can get everything from a full Java stack trace right down to putting a job through and stepping through it one step at a time. 

Each component within that job can be set to different levels, from warnings to the actual error itself. It's incredible how easy it is to find a problem and resolve it. Nine times out of ten, it prompts you on what you should be doing. So, if you've done something wrong, maybe you haven't got the schema quite right; it'll actually tell you that you have an error in the schema before you publish the job. 

It's subtle, so when it actually even finds errors, it won't go live to production or development until you fix those errors. And if an email does go through, or if there's an ETL with data, it tells you exactly what's wrong. And, generally, most of the time, it gives you some indication of what you need to do to fix it.

Data Governance Maturity

We're on the low end of maturity when it comes to data governance. But one of the beauties of Talend is that we now have one place to go to define business rules around data quality and data ownership. We're moving down the path of a more domain-oriented architecture, giving business owners more control of their own data by structuring our architecture around our business services.

Empowering Analysts

Talend Cloud is making it easy for us to have both developers and data analysts. Data analysts are able to manipulate the data through the cloud solution without having to write jobs and APIs. We publish those APIs and jobs, and they can even manipulate things through pipelines within Talend Cloud. 

This means you don't need a highly trained developer. You can just have an analyst who knows how to use the toolset but doesn't need to know the details or mechanics of how everything works.

Streamlined Data Management

This makes it easier for us to align our services with our data. Validation and other processes are pre-built into the jobs themselves, which means that data quality, data control, and data ownership are a lot clearer and in line with our business services. 

This streamlines things and makes them a lot easier. However, our maturity level around data governance is still low. We didn't really have any until about two years ago. It was a bit of a wild west when it came to data.

The interaction with Talend is minimal, mainly because the documentation is very mature. For most of the issues we've had or things that we've needed to try that we weren't quite sure about, we just went straight back to the documentation, and there was an example of pretty much everything that we needed to do. 

We have had quite a bit of interaction with the support side of Talend, but almost every instance has come back to something with our firewall rules or the way we've set up our security internally. Our operations team just started to learn about Azure. We've been on Azure for about four years, and our operations team still hasn't quite gotten the grasp of turning out all of our hands there. 

But most of the issues that we have had, our calls with Talend, have been around security and how we basically get out of our Azure environment to keep somebody outside of our banks.  

Talend is doing a lot of work at the moment, and it's not there yet, but the whole platform could be managed in a SaaS-type environment. You still need to have the Studio running on a virtual desktop or a PC. They will get to be able to do the whole thing inside your browser, so you don't need to install anything locally. It's down the track, and it's the nirvana that we were looking for in Boomi. 

But the biggest challenge they have is that the platform is so focused on the Studio for all of its development. They'll probably get there, but they have such a mature Studio client that it's a huge amount of work to get all of that functionality into a browser or SaaS platform. That's pretty much the biggest flaw with the Talend environment—being reliant on the Studio, which needed to be on a local machine.

The only other thing is that you have to integrate into an API gateway. We're in Azure, so we use Microsoft Azure Gateway. It doesn't come with its own gateway, which is another sort of big plus side that we saw in Boomi. Talend isn't quite there yet with the API gateway.

Other than that, it's bloody hard to find something because it just seems to be all good.

We've been using it now for 24 months. We ran it parallel with Boomi at the same time, so we ran Boomi for about a year and a half. And we've been using Talend now for about 24 months. It'll be 24 months in June. What took us about six months in Boomi, we accomplished in about two months in Talend. It was a lot easier to pick up and use.

We haven't noticed any slowness. We have an internal KPI for APIs to be responsive within 400 milliseconds, and if not, we have thresholds and alerts set up. We haven't had any APIs even come close to those thresholds.

• Robust Architecture:

This is partly due to our architecture, which uses storage queues or message queues for a lot of the ETL or data transfer. The APIs themselves get a sort of special treatment that ensures they have all the resources they need.

• Standard Configuration Sufficient:

As for the Talend solution itself, we haven't experienced any impact from updates causing slowness or needing to increase the service to a larger amount of RAM beyond the standard configuration.

Scalability is part of our licensing agreement. We can scale both horizontally and vertically. Adding additional servers is quite easy; we just spin up a new engine.

Load Balancing:

The nice thing is that when you create engines, you load them into a cluster pool. If a certain job or API is under heavy load, it will automatically move the load across multiple servers. So, it does a kind of load-balancing

Resource Optimization:

So, it does a kind of load balancing. We use Azure load balancer, but Talend is able to see load balancing across different environments. A job comes in, and Talend determines which server is free. If five servers are all under stress, it will bring up another server and use that at the same time. It's very easy for us to scale both horizontally and vertically.

Talend's support is good. We've got a lot of solution internally, and they provide third-party support. 

The only negative is that their support all comes out of Singapore, so we do have about a five-hour time gap between New Zealand and Singapore. But the people are great, even though English isn't their first language. 

They provide really good documentation, so 90% of the time we can sort things out just by reading the FAQs or the documents they have.

For the times we do need to talk to people, nine times out of ten, they respond within two or three hours of us logging a call. And pretty much every job we've logged with them has been a problem in our environment, and we created ourselves within our own security. We actually haven't had a problem with the Talend software itself.

Talend's support is really good. They're actually very, very patient and helped us resolve the issues we've had within our own environment. It's not actually part of the Talend platform. It's interesting enough that our security team learns more from them than they do from our security provider.

Positive

We used Boomi. We ended up switching from Boomi to Talend Open Studio. When things started getting really challenging, we had another project that was already using a solution that was very similar to Boomi, and it turned out that that solution was actually better suited to our needs than the Boomi platform. 

Mainly around debugging or finding errors and stuff like that, so many other platforms can be a little bit easier to use or even troubleshoot.  

Talend is a lot easier to manage and a little bit more mature product.

With Boomi, we were constantly moving from one screen to another, and you could tell it wasn't reaching its potential. So, we ended up going for Talend's own platform, and that's when we cut out development time for APIs and data ETL jobs down to just a few hours, if not minutes.

With Boomi, we were spending quite a bit of time trying to figure things out. Talend is just a phenomenal product. Boomi will get there eventually, they're just putting a lot of time and money into development, but Talend is already able to deliver the functionality that we're after.

It's actually one of the simplest things that we've had to deploy. You quite literally go to the manual and follow it step by step. Our main operations guy did the installation. He was absent sometime, and we had a new contractor start up. We just gave them the documentation from the website, and they followed it step by step. 

Everything just seemed to work. There was no real fireworks. The fireworks did come about when we put on the security certificate onto the machine. All the certificates that we had in the Java environment were using relative paths, and we found that they needed to use a full path instead. It took us two or three weeks to figure it out. 

We even worked with Talend support for a while until we figured it out. The documentation is an example of what most vendors strive for, but Talend has documented everything really, really well.

Our main operations guy did the installation. 

Deployment time: 

For deployment time, a remote engine, which is what everything runs on, takes about 45 minutes to do a full install and configuration. And that's not just double-clicking the executable; it's doing all the configuration as well. 

We've had to do that because we've upgraded to Java 17, which required us to set everything up on a new VM. That was an internal decision, not a Talend decision. 

We just wanted to set up a fresh server. It was 45 minutes to an hour, and it was all installed and running. We were able to deploy APIs and jobs onto that and do the testing. 

If it's just a job or an API that we've created, you can build it and deploy it within 30 seconds to a minute after you've built it. It's really easy for developers to deploy new jobs onto the server, test them, and give you feedback very quickly.

Deployment resources:

We generally have two people working on it at any one time. We've created several different templates, so if someone wants a new API or a new job to transfer data, we have pre-built templates to speed up development.

Most of the customization is done via configuration. They just type the information into variables, and in nine times out of ten, they don't actually have to change anything in the job itself. The entire job can be deployed as a complete brand new job, without having to build everything from scratch.

These templates make it really easy to deploy new work. My job is to define what models or templates to use, and then I hand them over to the developers. We've gone from six or seven people spending a week building things in Python or various other languages to just two developers spending half a day a week building jobs.

We've been able to realign our resources to move on to tougher challenges. Our integration now is significantly easier than what it used to be a few years ago.

Return on investment is kind of hard to quantify, but initially we sped up development with nine people for at least three years. 

After the first year, we really had enough with three people. But as a council, it's hard to calculate with a caveat or put it under internal costs. 

The easiest way to quantify it is that rather than years for full deployment, we're probably about two and a half. So, we've reduced the number of staff needed to support the platform by about 300%. If we figure out that in reference to an investment, that'd be fantastic.

Pricing is always a challenge. It is quite an expensive model, but because the platform is so simple to use, we haven't had to purchase any additional licenses.

Unexpected Efficiency:
We started out with three developer licenses. I'm not sure we've even used all of them. After the first year, we didn't actually need any more licenses because the guys that were using it were getting through ten times as much work as we expected. 

So we don't actually have enough work for six developers, let alone the three developers that use it one day a week. We've spent time on building templates and standard patterns, and as a result, within a year, our workload was pretty much down to zero. We're probably about a year ahead of schedule compared to our original roadmap.

Overall Value:
So, it is very pricey, but we've been able to chew through the work so fast that it's actually hard to justify buying more licenses.

Overall, I would rate it a nine out of ten. I've been recommending Talend for six or seven months now to almost anybody who's interested in that space. 

Compared to other products, we had about 12 on the list that we analyzed, and we ended up choosing Boomi and Talend. And even then, those two were quite significantly out in front. It's a low-code, no-code, simple-to-use platform that's easy to support. 

Talend is so far out in front of everybody else that we know it's very hard not to recommend them to anybody. It's all around error handling and ease of use. Boomi does so many tasks so well, but in some areas, it's really quite hard to use. 

Those areas are where Talend is so easy. It's a real big difference. Functionally wise, they're both providing the same functions, but Talend is significantly easier to use.

There are two main use cases related to the solution, one of which is building an ETL for a data warehouse and creating an automatic process related to jobs with some targets and some goals to prepare data for the company.

For two years now, I have been working with the SSIS. I can see a great contrast between SSIS and Talend Open Studio. Though Talend Open Studio offers performance for some small operations, it needs more technical capabilities than SSIS.

The physical appearance of the product is an area of concern where improvements are required. As Talend Open Studio is a tool we use for ETL-related purposes, it needs to be improved to make it easier for beginners to use. Talend Open Studio is not easy for a beginner to use, and I know this because before I started my internship, I never knew about Talend Open Studio. The aforementioned aspect can be considered for improvement.

I have been using Talend Open Studio for six months. I use the solution's latest version.

Stability-wise, I rate the solution an eight out of ten.

Scalability-wise, I rate the solution a seven out of ten. I did not face any problems with the product's scalability. The tool is not very easy for a beginner to manage.

Only five engineers in my company work with the solution.

The solution's technical support is responsive and helpful. Talend Open Studio's technical support team helped me grow my skills related to the solution since I had never used it before my internship.

I rate the technical support a nine out of ten.

Positive

I have experience with SSIS. Between SSIS and Talend Open Studio, I would choose the latter.

I did not take care of the product's installation phase since my company had an infrastructure team consisting of five engineers who did the installation for us.

The solution is deployed on an on-premises model.

From an ROI perspective, the tool helps save time in our company by 30 percent.

I recommend the solution to others who plan to use it.

I rate the overall tool an eight.

The data quality features are quite noteworthy. Data quality features are important.

There is a need for mastery in some areas.

Stability feels fine.

Customer service needs a better approach.

The initial setup was okay.

The implementation team used Studio.

The setup cost is expensive, so there are considerations.

I'd rate the solution eight out of ten.

Our client wanted an open-source ETL tool to load data from his ERP to PostgreSQL Data Warehouse. Our company usually uses Microsoft technologies, but since the client wanted an open-source product, we used Talend and Postgres based on the client's requirements. So they just used it to pull data from the ERP source to PostgreSQL Data Warehouse.

You can do pretty advanced tasks, like using a code generator at times. You can even go to the Java code, and Talend Open Studio can be customized as much as you want. With other products, like Microsoft SQL Server Integration Services, you are limited to the UI interface, because of which you cannot do a lot of customization. The community of Talend Open Studio is very large, and it's an open-source tool to which others can add features.

When speaking of the ETL capabilities of Microsoft, I would say that the documentation of Microsoft is very much better than Talend.

Usually, I go to the community of Stack Overflow to check answers to my doubts or queries, so Microsoft provides all details since its documentation is great. The graphical interface of Microsoft is much better than Talend. Talend Open Studio's interface does not provide a good experience, especially when compared to Microsoft products.

In the future, I would like to see better documentation and a more enhanced graphical interface in Talend Open Studio.

The performance of Talend Open Studio compared to Microsoft is weak, and I think it is because Talend Open Studio runs on Java. Java is slow compared to Microsoft, which uses C-Sharp in the backend. Performance-wise, Talend Open Studio is a bit slow.

Microsoft products are usually used in Microsoft's environment, and maybe because it's a native environment where everything is integrated, causing the performance to be better. In general, I cannot clearly state whether Microsoft or Talend offers better performance. With my experience, I believe Microsoft's products are better in terms of performance than Talend Open Studio.

The installation process of Talend Open Studio can be improved. Suppose you download Talend Open Studio during the installation process. In that case, Talend should check if Java exists in the user's system, and if it doesn't, it should take you to the link where you can download Java's compatible version with Talend.

I have been using Talend Open Studio for approximately two to three years. I currently use Talend Open Studio Version 8.

Stability-wise, I rate the solution an eight out of ten.

Five users, including engineers, business intelligence developers, data management consultants, and developers in my company, are using the solution.

I don't have a clear idea about the scalability of Talend Open Studio. I think Microsoft's scalability is better since you can install it on a cluster.

I did not use certain aspects of scalability to check how much I can distribute the tool in the architecture and find where we can have multiple nodes around the same ETL. I worked with one server and could run just one instance using it.

Regarding Talend Data Management Platform, I want to communicate with the support team of Talend because the installation media is not available on their website, and I have to order it. It's not good for a developer who wants to start with their product since there is no trial version.
One has to first purchase a license and start learning about it. If Talend wants to have new clients, it's not a good idea not to provide a client version. Microsoft does provide developer versions, so anyone can install them as a developer, and it is in the production phase, they push out the license.
I rate the technical support a seven out of ten.

Neutral

Previously, I used Microsoft SQL Server Integration Services, Microsoft Active Directory, along with Microsoft's business intelligence stack, consisting of SQL Server Database, MDM integration services, OLAP cubes, and Power BI for its dashboards. I have also used Pentaho Data Integration.

Talend Open Studio is deployed on-premises.

Talend Open Studio's installation process is easy. One just needs to install Java before installing the product.

It is an open-source tool which means it is a free solution.

The cloud-based version of the product is paid.

The additional costs of the solution would be the operation systems, servers, and hardware. If the OS is Linux, you can use the tool directly without a server from Windows. Some costs might be attached to the infrastructure after the tool's deployment, which is mostly applicable to all products. All ETL tools require a team to support the project after they go live.

Maintenance is required not because of Talend, so there is a team to help each client adapt to the ETL account and change the data source.

There is a team of two members for handling the solution's maintenance of our clients. It is not that there is always a change in the data source, but when it happens, our clients contact us and ask us to edit the ETL according to the changes.

I would recommend the solution to those planning to use it.

When comparing Talend Open Studio, I saw many other products with better documentation and UI that are more scalable.

I rate the solution a seven and a half to eight out of ten.

Talend Open Studio is mainly with data warehouses. It is also used for data masks. There are so many types of projects we can do with Talend.

There are many things I like about Talend since I'm a Talend enthusiast. There are many architectures: hybrid, cloud, and on-prem. We can use the solution on a Linux or Windows architecture. There are many solutions for the same problem when we get experience and build a relationship with the tool. The tool also has a strong community.

I've used Talend Open Studio since 2016.

I rate Talend Open Studio's stability an eight out of ten. Talend has some problems sometimes.

The tool is not scalable, but the solutions you create with the tool are.

Open Studio's initial setup is pretty easy. When installing the solution, you get installation on the cloud console, and you get the installation with Java integrated. The easy configuration allows a quick start.

However, you need some lessons. I give lessons to customer companies. The course has four lessons per year. It's quite simple to learn how to use Talend.

Open Studio is free to use. Without a license, you cannot use versioning, and you have to use an external scheduler. The product is not so expensive compared to other alternatives. Talend Open Studio costs about 11,000 a year.

I rate Talend Open Studio a seven out of ten. I recommend this solution.

We are currently using the free version. We use the solution to extract loads from the source system into the entire house. We are also doing API and data integrations from TXT or XML files.

Though we are not currently using it for big data, there is a plan to extract data from MongoDB.

The API integration and big data approach are very good because of how you extract data from JSP files or big data web repositories like MongoDB. This is an excellent tool for big data and API integration.

Comparing Talend Open Studio with SSIS, Informatica, or Data Services, I think the overall workspace concept can be improved since these other solutions have workflows and data flows.

I have been using Talend Open Studio for the past three months.

The solution is stable enough. It's a mature tool. It is a bit complex in a few areas, such as when writing your ETL Jobs, because tLoop is slightly confusing.

I have used SSIS, SAP Data Services, and Informatica to a small extent in my previous companies. My current company uses Talend.

We currently use the free version, where we don't get the built-in scheduler. We had to deploy the job on the server as a file and, using the CRON jobs, schedule the batch or shell file. We can deploy the solution and the ETL jobs in ten or 15 minutes.

We are using the tool's free version because the enterprise version is a little expensive.

You would need three enterprise servers if you are going for a full-scale lifecycle, like development, quality, and production. That would be expensive because the servers have yearly costs.

We can only have one user in the free version. We are now wondering whether we should go for the enterprise version or switch to a different platform, like the HANA platform, HANA cloud, or Datasphere. Or another platform called Encarta. There are multiple technologies we are exploring.

It's a good tool overall, but that depends on whether you have the enterprise version. The free version is not for a team in which you have more than one member. In this case, you have to go for the enterprise version, which is a bit expensive.

Likewise, you do not get support with the free version, so you won't speak with tech support. There is, however, a community with a knowledge base in forums.

You also get the scheduler, the deployment, and everything with the enterprise version. You could deploy the solution on the server and then schedule the jobs. Likewise, you get logging and an audit trail of your jobs. Currently, we maintain these jobs manually.

They should make purchasing the servers for the enterprise version a one-time payment, where you buy the server, and that should be it. The yearly cost is what makes Talend not a great tool.

Talend is fine if you are only using it for the ETL tools. However, there are other options, such as SSIS, Informatica, and SAP Data Services, which are not promising for Talend Open Studio. Some things are easier with other tools. Talend is logical, but a few things are not very straightforward. If there is an option to choose from other ETL tools, I would rate Informatica number one, in second place, maybe SSIS, and then maybe Talend.

If Talend Open Studio is cheaper than other solutions, you can go for it.

Our primary use case is for data migration and validations.

This solution has helped our organization by allowing us to query from different sources and transform and send it to APIs through API calls.

The most valuable feature for me when it comes to this solution is that it's easy to use.

I would say that writing to JSON is kind of a pain. It reads from a JSON file pretty well, but writing to a JSON file is not so great because its components are not good. I would say this is the most important.

I have been using this solution for about six months.

I would rate the stability of this solution a nine, on a scale from one to 10, with one being the worst and 10 being the best.

I would rate the scalability of this solution an eight, on a scale from one to 10, with one being the worst and 10 being the best. There are around seven or eight users of this solution in our company. Eventually, we plan to scale to 20 people.

I would rate the technical support of this solution a nine, on a scale from one to 10, with one being the worst and 10 being the best.

Positive

I would say that the initial setup was medium-difficult.

I would recommend other people looking into this solution to pay attention to the learning curve and the documentation, which is the solution's strong suit. The Talend academy is a good place to start.

I would rate this solution as a whole a nine, on a scale from one to 10, with one being the worst and 10 being the best.

We have used the solution for a real estate agency in Singapore to stream their data on a more or less real-time basis to a mobile app. Their backend system is primarily built on SAP. They used a combination of Kafka message querying and Talend to stream the data into a NoSQL environment, which is MongoDB. The solution basically streams the data from SAP backend systems, all the way to a NoSQL MongoDB database, to support a real-time or near real-time data flow into a mobile app. 

This mobile app will be a consumer-facing app. The end consumers will be able to look at the status of their real estate space. Users are mostly the tenants who either rent or lease commercial or residential spaces. The status of their leased property will be made available via a mobile app. The data solution comprised of technology such as Kafka for message queuing, Talend for data flow streaming, and MongoDB for storing the data. That's one use case.

The second problem solved using Talend is to build the data lake for internal data analytics at a government organization, again based in Singapore. Their backend systems are primarily supported by Sybase databases. From the Sybase databases, selective data is brought into a staging area via Qlik Attunity, which is a real-time data synchronization tool that uses the change data capture mechanism. Once the data comes to the staging environment through Attunity, then Talend takes over. Talend pulls that data and moves it to Cloudera's big data solution, the big data platform. Then, Tableau is used for data analytics and data visualization. This solution is set up for internal data analytics within the government organization.

The tool offered essential connectors to connect to various data sources and our big data platform. Our data sources were SAP/ MS SQL and Sybase. The Hadoop/ hive connectors are used to connect to the Cloudera-based big data platform. The components available to loop and dynamically pass source/target connections helped us to stream data based on source tables' metadata. This helps us to onboard new data sources quickly, rather than building a custom data pipeline for each new data source.

The main differentiator that I have seen between Talend and other data integration tools is the ability to view the data pipeline in the form of a program/ code. The other tools hide the backend code behind the scene, however, Talend shows the backend code that gets generated from the UI (user interface) components used for creating a data pipeline. I.e., In most of the data integration tools, all we do is, leverage certain UI components meant for data processing such as reading, selection, filtering, sorting, writing et cetera to build data pipelines. The tools such as Pentaho or Informatica, will convert the pipeline into a backend code, which is written probably in C++ or java program. This is not transparent to the developer or the programmer. But in the case of Talend, the tool converts the pipeline into a programmatic format and shows it to the user. This helps in debugging errors. 

The json parsers can flatten the API response body into a structured output. This is useful when we try to use Googleapis for YouTube video analysis and to check corporate status uding ACRA/IRAS apis from Singapore governments.  

If somebody has a passion and interest in software programming, they will like this feature. Of course, we can do without it, however, this additional feature available in Talend helps the data engineer to understand the code better, understand how to better tune for performance, and so on. The differentiation is the ability to view the data pipeline in the form of a traditional software program.

Comparing the plus and minus of other tools and will try and suggest what can be improved in Talend based on how other tools implemented certain features. 

Both Talend and Pentaho leverage open-source plugins. There are lots of open-source committees that build solutions as per their requirement;  Say, for example, someone may have had a need to create a Tableau data extract from a certain database for the purpose of data visualization in Tableau. They may have created a plugin for this purpose and shared it with the market. We can always go into the market and search for a plugin that someone else created (for ex., to generate TDE files) when we have similar needs. Both Talend and Pentaho offer the 3rd party plugins in their market. This plugin market feature automatically extends/improves the tool capability. If somebody needs a particular feature, they can always build one and publish it for the other developers to use. 

Other than this, maybe if Talend had a few components, it's easier to deploy solutions; job scheduling is one. Currently, the job scheduler is a separate component within Nexus and is accessible over browser. It can be integrated within TOS (Talend Open Studio) so that we dont need to switch between the two. Nexus holds the repository of compiled program code. However, Nexus doesn't have the ability to keep the repository of the source code programs. That can be improved. 

Currently, Talend has a versioning feature, meaning the previous version can be named as 0.1 or 1.0 and the subsequent minor/major versions can be increased with incremental version numbers, however, it is not that great as a version control repository (such as VSS, CVS, Git, etc). If a Nexus-like tool can be used for maintaining versions of the underlying source code (not just for the compiled code), that would be great. Basically, version control can be improved.

I used the solution from 2019-2021. Currently using Fivetran, DBT etc for ETL.

The stability of the solution is mainly based on the server architecture behind the scenes. We have seen some instability issues on other tools around the Talend platform. For ex., Qlik Attunity had been throwing some exceptions every now and then, however, in the case of Talend, we haven't seen many. At times the server would run out of memory. Based on the capacity we have, we may have to go and clean up a few logs and other systems’ data before resuming the processes manually. 

The system's stability is okay, especially in the case of our data warehouse. Users do understand that a data warehouse is less critical when compared to operational systems. The users' reports are normally generated at the end of the day for their availability, the next day. We have a workable level of stability, which is good enough. We don't see any instability issues.

Talend operates on a clustered job server setup, available in both Linux & windows. The clusters can be scaled up with additional nodes as per demand. In the case of an on-premise setup, we assess the load and define clusters based on the load. We did not have auto-scaling.

Talend has a support portal and we can get access to the support portal. The initial process could be a little difficult, to get into the support. The initial access provisioning could be improved. It took about a week's time to get a support login created. Once that is done, then you will get quality answers to the support questions.

There are two ways to get technical support, either from the dedicated support team from Talend, or from the open-source community. The dedicated team provides an answer for sure, however, we may not get answers to all our questions from the open-source forums.

Neutral

We are aware of the other solutions such as Pentaho. But, there was no switching. For most of the customers, I do the evaluation of tools, over one to two months and then finalize on one particular tool. They don't really move around from one ETL tool to another ETL tool, unless the tool itself is getting obsolete or if it is no longer supported. In the case of Talend, from what I have seen, there is no switching. They must have initially evaluated a few tools and finalized with one. The main reason could be that Talend supports a big data environment. Talend has the ability to integrate with market solutions. Third-party integration is possible. Talend is also becoming popular in Southeast Asia, so skilled resources are readily available in the market if we need people to maintain the environment. Those could be some of the factors that influence the selection of Talend.

Initial setup requires a systems engineer, who has knowledge in Linux/ Windows OS, depending on the server type, to work with Talend support. It takes about 2-3 weeks for the basic setup. For additional plugins/ components, guidance from Talend support team is required.

After the initial setup, most of the ETL data pipelines take about a week at the minimum to build. Typically, we reap the benefits after a month or two from a particular data stream. Of course, it is based on the different use cases and depends on the number of data pipelines involved in solving that particular use case, however, smaller scoped items can be delivered in one to three months' time. 

After development and testing, the time required for the deployment varies, depending on the process we adopt and depending on the version control tools that we have as well. Certain tools will easily integrate into the Talend environment. In those cases, I mean, it would be a medium-complexity deployment. We can't say it's very simple, we can't say it's too complex either. There is a medium level of difficulty involved in deployment.

The time it takes depends on the processes a particular organization follows in terms of change management and deployment. Some will require at least one week of review with the change management board. In certain organizations, the approval processes may take up to three weeks to get something deployed. If the process is a little simpler, if the developer has a fair level of control, it could be just one week.

A vendor team was in charge of the server maintenance and their scope of service included deployment as well, as developers do not have access to production networks in established firms. Developers normally provide a step by step instructions in a word document that the vendor team will follow. Some familiarity with the tool is enough, but the vendor personnel we worked with had the reasonable experience to an extent where they recommended code improvements. 

Typically before embarking on a project, we analyze the return on investment, which could be over two to three years. It could be as quick as three months, depending on the scope of the data integration project, however, in the case of a data warehousing solution, it could be two to three years. 

The real estate company we worked with, must have already seen the return on investment. Their mobile apps should be live with the data integrated at the backend. 

For the internal data analytics use case, they would've also seen the benefits, as the data is already available in the data lake. The users will be able to do the analysis with the integrated data pool in the data lake. 

Both the companies have seen the returns.

Pricing is competitive in comparison with Pentaho. Both are at par. Solutions such as Informatica could be more expensive. Certain packaged solutions that use Talend as ETL may be available, if you are looking for a pre-built data warehouse model.

Yes, Apache NiFi was evaluated. Based on the POC (proof of concept) results, we chose Talend.

Some of my comments are based on my limited understanding of the tool. Please do due diligence before selecting the tool. I recommend a POC over 1-2 months on shortlisted tools, before finalizing one.