IBM DataStage vs. WSO2 Enterprise Integrator

DataStage is somewhat outdated for an ETL. I guess that's what makes it a bit lagged behind its competitors. It can be used for data processing, sure, but its performance seems to be lagging behind or quite slow given the server it is running from. I won’t depend on this application if it's handling a lot of mission-critical banking and business data.

The best-suited scenario is the service chain pattern or all patterns used in online mode. The less appropriate scenario is a batch service the duration time of the service is more than 10minutes because it is necessary to increase the HTTP timeout.

• Connect to multiple types of data-sources including Oracle, Teradata, Snowflake, SQl Server.
• Powerful tool to load large volumes of data.
• Transformation stages allow us to reduce the amount of code needed to create ETL scripts.
• Allow us to synchronize and refresh data as much as needed.

Incentivized

• Service chain pattern
• Message transformation
• Call more than one backend service
• Call service with different protocoll as the client use
• Use file as input
• Use mail as input

• Technical support is a key area IBM should improve for this product. Sometimes our case is assigned to a support engineer and he has no idea of the product or services.
• Provide custom reports for datastage jobs and performance such as job history reports, warning messages or error messages.
• Make it fully compatible with Oracle and users can direct use of Oracle ODBC drivers instead of Data Direct driver. Same for SQL server.

Incentivized

• View deploy artifact in date time order
• Easy way to map data
• Easy way to manage payload

Because it is robust, and it is being continuously improved. DS is one of the most used and recognized tools in the market. Large companies have implemented it in the first instance to develop their DW, but finding the advantages it has, they could use it for other types of projects such as migrations, application feeding, etc.

Incentivized

It could load thousands of records in seconds. But in the Parallel version, you need to understand how to particionate the data. If you use the algorithms erroneously, or the functionalities that it gives for the parsing of data, the performance can fall drastically, even with few records. It is necessary to have people with experience to be able to determine which algorithm to use and understand why.

Incentivized

IBM offers different levels of support but in my experience being and IBM shop helps to get direct support from more knowledgeable technicians from IBM. Not sure on the cost of having this kind of support, but I know there's also general support and community blogs and websites on the Internet make it easy to troubleshoot issues whenever there's need for that.

Incentivized

With effective capabilities and easy to manipulate the features and easy to produce accurate data analytics and the Cloud services Automation, this IBM platform is more reliable and easy to document management. The features on this platform are equipped with excellent big data management and easy to provide accurate data analytics.

Incentivized

We can start with the community version and then when we moved into production we can buy the supported version. The supported and community version have the same code so we can do every test before deciding to buy the supported version.

• It’s hard to say at this point, it delivers, but not quite as I expected. It takes a lot of resources to manage and sort this out (manpower, financial).
• Definitely, I don’t have the exact numbers, but given the data it processes, it is A LOT. So props to the developer of this application.
• Again, based on my experience, I’d choose other ETL apps if there is one that's more user-friendly.

• Reduce development time to the service integration
• We can use junior developer
• Clustered in easy way