AWS Step Functions Reviews

It is like a workflow to find a sequence of tasks, like in Zapier, Appian, and Camunda.

So, our use case was infrastructure-related orchestration. Not application-related infrastructure.

It was more Amazon API for infrastructure. We were trying some EFS-related stuff and some proprietary APIs from Amazon.

It wasn't easy to understand the licensing model. It's like if you use just a little, it's cheap, but it becomes more expensive as you use more. It's like a hook that ties you inside the Amazon ecosystem. So, it creates a dependency.

I don't like it because it creates a dependency, like Microsoft or other players who want you to stay within their ecosystem as much as possible.

I used this product for a couple of months because it was for an analysis project. So, in that case, it was a practical approach, comparing Lambda and Amazon Step Functions.

It is quite stable.

It's Amazon, it's scalable.

There's no technical support from Amazon unless you are from a bigger organization.

The support is there for critical stuff with it, but not for our particular project.

Zapier and Appian because I was studying them. So, they're suited for small to medium enterprises that mostly use SaaS services. But once you have a decent amount of work volume, they become prohibitively expensive.

However, they're not suited for critical enterprises that require execution in a secure environment. They're just stuck somewhere in the cloud. So, from my perspective, they are not suited.

And when it comes to Camunda, from a technical aspect, from a competitive analysis perspective, it's kind of old school that may not still fit the needs.

The initial setup is fast. There's no setup; it just runs. It's Amazon. You write it and spin it up. It's okay.

I would rate my experience with the initial setup a ten out of ten, where one is difficult and ten is easy.

It is a SaaS platform; it's like just go on a website, and this is not a problem. It's there from the start, and you can spin it up in a few seconds.  

We went for the Step Functions service. The pricing for Step Functions was quite convoluted and difficult to understand due to the numerous small factors involved, such as data transfers and limitations on runtimes. 

Additionally, replacing existing Step Functions tasks was not straightforward. As a result, we opted for a solution that provided clear and understandable outputs, allowing us to effectively monitor and manage our workflows. 

For we went for Lambda function for more decent output and to see what was going on. It was more Amazon API for infrastructure. We were trying some EFS-related stuff and some proprietary APIs from Amazon.

So, pricing depends on what I am doing. If you're consuming two or three things a day, it's fine. It's cheap.

But then there's no price control because Amazon is about, "Let me help you consume as much as possible," so then I can be, like, hell at the end of the month. 

Technically, it's okay, but otherwise, it's Amazon taking all your money if you're not careful.

Overall, I would rate the solution a five out of ten. 

We use the solution to mostly execute flows that contain parallel processing.

The solution's most valuable feature is its ability to parallelize tasks. In addition, the degree of parallelization can also be made scalable using its "map" state. Also, troubleshooting is way easier with an easy-to-understand graphical interface.

The solution has a data limit of 256 KB which means you can only pass under 256 KB of data to a state machine and between its state transitions. This can be improved.

We have been using the solution for a year.

The solution is very stable.

It is a scalable solution. We have four users in our organization.

The solution's initial setup process is straightforward. It has a graphical interface that provides drag-and-drop functionality. It's simple to configure and use.

The solution's price is reasonable.

The solution has a good interface and performance speed. I rate it a ten out of ten. I advise others to use the solution if they have to split their logic into a series of multiple lambda functions. It works better than Lambda functions in terms of orchestrating and managing parallel processing.

In terms of our use case, we are building a pipeline to manage data. It's pretty simple. We validate the data and then write it in the database.

It has enabled my organization to create workflows with a lot of different AWS services. It's a general solution that you can adapt to your own needs and is simple to use.

When it comes to valuable features, we like that it includes everything that’s connected to AWS which means that it can be integrated with everything in the AWS suite. We also like that the creation of the pipeline can be done using the graphical user interface. It's pretty easy because you don't have to write the code to create the tasks.

We are building our software and we just released our NDP internally, so I've been using it for almost three months.

It's stable and the responsibility is very good.

We are a four-person team, and for the moment I am the only one with access to Amazon Step Functions. We are an early-stage startup, so we are scaling and building our platform. We are looking for companies to test our software.

We met one of them just last week, so they are available. They are helpful.

Positive

When I used Apache Airflow, the main problems were that the debugging phase was not so easy and Apache Airflow only accepted Python code. It was only Python-based. On the other hand, Amazon Step Functions allows you to use different kinds of codes.

It was not so complex. It's pretty easy because as I mentioned, you can use the GUI to create the pipelines and the workflows.

It took about one day, because we had to set a couple of things. We had to understand some errors, but now everything is working good.

We did the implementation in-house.

I recommend speaking to a consultant if you are not familiar with workflow managers. Otherwise, it's very easy to understand how a workflow can be composed.

I rate it a 9 out of 10.

We use Step Functions to straighten out the events from the event grid.

One can rate all the calls and that is a good feature. 

The pricing of the solution can be improved. 

I have been using this solution for ten years.

The technical support team is good. 

I rate the overall solution a six out of ten as it needs a receipt feature to function better.

We use it for the execution of tasks quickly.

We mainly determine if the client is using a different cloud provider, or cloud solution, such as AWS, and then we use it.

Amazon Step Functions is good for AWS-specific solutions. If there are too many third-party products or solutions available, like Camunda and Airflow, it can make sense since they can work across the clouds. The price and support are areas with shortcomings where the solution needs to improve.

It is always good to have, like, across hybrid clouds or such things.

I have been working on Amazon Step Functions for the last six months. My company has a partnership while also being the distributor IT service provider of the solution.

The solution is stable.

The solution is easy to scale. Our clients are medium-sized businesses.

The customer support is good. They can be more supportive and responsive. I rate the solution's customer service a seven out of ten.

Neutral

The initial step of the solution is easy. With GUI or CLI, one can quickly do the deployment.

The deployment process takes two hours, depending on the use case. One engineer is enough to deploy it.

There is not much maintenance. Only some modifications are needed.

The solution is expensive.

I would recommend those looking into the product look into their use cases and see the functions and its limitations. Compare it with third-party provider solutions and develop a point of view to determine if it truly makes sense for the use cases or if alternative options are required. These limitations, as mentioned, primarily apply to AWS at the moment. Additionally, it is not sure whether there is a time limit on the number of invocations, as costs will increase. Therefore, all these aspects must be taken into consideration before finalizing the product.

Overall, I rate the solution a seven out of ten.