SSIS Reviews

• Compatibility with the rest of Microsoft BI stack
• Scripting component which opens unlimited possibilities: by having the full .Net stack/libraries available, if you need a yellow rubber duck to quack every million rows, it will quack
• Inter-stream dependency mechanism built directly into the tool; the whole ETL solution can be driven off it without having to employ third party software
• Parallel processing of a data flow: let's say there's five steps in the flow and ech step consumes output from the preceding step; once step #1 is completed and its data passed to step #2, the engine will feed the next batch of rows to step #1. Once the first batch of rows reaches the last step, there are five concurrent steps working on five different data sets, maximizing utilization of server resources
• All the necessary types of enumerators (files, xml nodes, rows and many more)

• SSIS packages are stored in XML-like format; we keep them in SVN which makes it very easy to track changes
• Built-in dependency tool is really great; you can not only react to success/failure results, but you can evaluate any expression in the flow and execute (or bypass) any components based on the result of that expression. This gives you full control to build decision trees of any complexity

• Writing custom components and sharing them across multiple ETL streams is tricky and requires specialistic voodoo knowledge
• Make the GUI less eye-candy and more responsive; especially the scripting component needs some TLC regarding UI responsiveness

It's easy to deploy, you just need a simple file-copy mechanism that is sufficient for most deployment scenarios.

The good:

• Straightforward, intuitive, quick to learn
• Built-in debugger: variables, data viewers, breakpoints - everything you would expect from a modern software development tool
• Sleek, snappy UI
• Great flexibility with storing and deploying ETL components

The bad:

• Third party plug-ins may become incompatible with future versions of SSIS
• Upgrading is usually painful and time consuming despite what MS says
• Compatibility issues may arise when used with non-Microsoft technologies

We use this solution for extracting data from various databases and saving it in our data warehouse.

We use the on-premise deployment model.

My team is facing problems regarding the database connectors, which are not available. The MySQL connectors need to be purchased from outside vendors. They should provide connections for more SQL databases, free of charge.

The performance of this solution is not as good as other tools in the market. Compared to the same job is running in a different tool, it will take longer using SSIS.

This is a stable solution.

This solution is scalable.

The number of people I have using this solution depends on the size of the project. Normally, I need three to five ETL developers. Sometimes, if the project is big enough, then I will need more.

I have not contacted Microsoft Technical Support for this solution, although we have sometimes accessed the internet to research problems that we face.

In my previous company, I was leading a team who were working with Informatica. Here, they stick to Microsoft technologies and are unwilling to change.

The initial setup of this solution is very straightforward. In a few hours, everything was up and running.

The decision to use this particular solution includes many factors. Some companies do not want to purchase a license for another product because this one comes included with the database.

SSIS worked well for small or medium-sized Projects. For larger projects with huge data, I believe that you should search for another solution as you will need to do manual fine tuning. Additionally, some components such as SCD will show unexpected errors with huge data.

 As Microsoft is very slow in providing updates and enhancements to SSIS, I see that the future for Integration projects in Saudi Arabia goes towards other vendors products such as  Informatica powercenter, IBM DataStage, and Oracle ODI. 

Compared to other Integration tools, I would rate it a six out of ten.

Movement of data and creation of files. ALl the typical things that you would have a ETL solution do. Data movements were in the millions and no calculations were completed. This means it was always a select * from where ever it was coming from and going to. Light translations like concatenation was being used. 

SSIS was easily adaptable to our group. It was cheaper than the other tools that we compared it to, however I feel that we got what we paid for.

The packaging and how it is organized is good for someone that really has never seen ETL before. 

Scalability of SSIS needs some improvement. Seems to get sluggish as soon as we hit a high volume of data.

Again it failed a lot and by a lot I mean every day. The failures were false alarms and caused many sleepless night for our company that I used to work for.

SSIS is good for smaller shops that don't really have a high volume of data.

I would rate the customer service as poor. 

We were using Hyperion Application Link. We switched because HAL was being sunset.

Never participated in the initial setup.

In-house.

For the money, it's a decent tool. However, if the budget was larger I would have gone with a different tool

Look at how this product is sold to you. Ask yourself, am I getting everything that I need. Its more expensive to get the additional adapters after the fact.

We looked at ODI, Informatica, and DataStage. All three we had in-house. ODI was the better option and after dealing with SSIS for only a few months, we ended up using ODI.

We use this solution for ETL, which includes data summation and cleaning. 

This solution used in an on-premise deployment, for now.

I have used most of the standard SQL features, but the ones that stand out are the Data Flows and Bulk Import.

The synchronous processing needs to be improved. For batch processing, it works fine, but when you start to do real-time processing, I find that this solution is not strong, depending on how you use it. If you use it for short batches, micro batching, that might work, but it is not as good for queuing real-time processing. This solution needs full support for real-time processing.

The solution needs better support for XML and JSON.

I use this product extensively, on a daily basis. It is stable.

Over time, I think we'll most likely start to decrease usage. This will happen as we move to more real-time processing, and we will most likely start to do more queue processing. In terms of batch processing, it will scale down considerably.

In the catalog, it's supposed to be scalable. I think that it has support for an SQL cluster.

In my opinion, I think it's a bit more limited in terms of scalability, although it scales with the database. I would say that the scalability is intermediate in terms of being able to launch multiple instances, or it could do load balancing as well. I think that would be a bit more challenging.

We haven't needed to contact Microsoft technical support. When we have trouble we usually use Google to search for what we need to find out. Also, in terms of issues, there's a lot of information on SQLServerCentral and Stack Overflow.

For the most part, we started with this solution. 

The initial setup of this solution is pretty much straightforward.

If you want to develop with Visual Studio then you have to install the data service add-ons afterward, so it is a bit cumbersome. Then, if you want to use the catalog on the database, with deployment there are often security issues and you have to get an SQL catalog up and running. This can also be a bit cumbersome.

I would say that it takes a day or two to deploy this solution in a new environment, and it can be completed by one or two people. A single developer may be sufficient to deploy and maintain the system.

I implemented some of this myself and had help in terms of setting up the security. There are often security settings that require the assistance of a DBA.

This solution was already in place. That's what is available and it's what people know. Going forward, this will most likely change.

My advice to anybody implementing this solution is to look into whether to use it on a catalog in a database, versus using package deployment. There are pros and cons to both approaches in terms of deployment and security. I would say that's something that needs to be evaluated quite early. There are lots of benefits to the catalog, but also a bit more admin attached to it.

Another consideration is real-time processing needs. If this is a requirement then I would recommend against using this solution, unless the next version has a new set of features specifically for real-time processing.

I would rate this solution an eight out of ten.

For the full version of the SQL Server, SSIS, SSAS and SSRS will come as additional features for free. Hence, my organization need not spend extra money for other ETL, reporting and analysis tools. This can give very good flexibility.

In SSIS, the scope is not only to handle ETL challenges, but it will allow us to do so many other tasks, such as DBA activities, scripting, calling any .exe or scripts, etc.

SSIS can improve in handling different data sources like Salesforce connectivity, Oracle Cloud's connectivity, etc. Also, handling of the different data types will be a big challenge here; so expecting improvement in these areas.

There were no stability issues.

There were no scalability issues.

Technical support is very good.

Based on the client's requirements, we switched over to this solution.

The setup is straightforward.

We looked at Informatica and DataStage.

Don't worry, go ahead.

Our primary use for this solution is to move data between points and applications.

This is a flexible tool to use and comes with the MSSQL server package.

This solution is easy to implement, has a wide variety of connectors, has support for Visual Basic, and supports the C language.

Tuning using this solution requires extensive expertise to improve performance.

This solution is included with the MSSQL server package.

I am an ETL developer working as an information zone leader. We are an outsourcing company to our customers.

The most valuable feature of SSIS is that you can take data from other servers which are not MS SQL Server or Oracle. 

I would like to see SSIS improve the collection of data trail servers and reporting. Future releases should improve the data lineage, as it currently is not good.

I have been using SQL Server Integration Services for five years.

This solution is stable, but the performance is not good. I believe it can be improved. Informatica is better.

I have not been restricted by this solution.

Prior to using SSIS I used Informatica. I switched because my organization is using SSIS.

The initial setup of this solution is not complex. 

Most of the time we do the implementation in-house, however, we have used third parties in the past.

Our license with SSIS is annual.

If an organization has the money, I recommend they use Informatica. 

I would rate SSIS a seven out of 10 overall.