Amazon Elastic Container Service Reviews

We host our production application on EC2. We host the application and take regular backups to create mirrors for our customers. We also use the solution’s firewall to save us against any threat.

The most valuable feature of the tool is its ease of usage. We spin off new servers and host our applications on those servers. The best thing I like about the product is hosting. It is super easy. Features for creating a server and hosting it is all builtin within the solution. I don't have to build a backup service. It is already available. I don't have to build a database as well. It is already built. I can just use any existing DB and then connect it directly to the DB instances and use it. Everything is ready to use.

We’ve been hosting our production application on the product. The most important criterion for us is downtime. In seven years, we have not faced a single second of downtime. It is the greatest thing. We have had zero downtime in the last seven years. Nothing can be better than that.

Billing is extremely complex. The whole concept of billing is so complicated that we never understand it. There is an upfront cost, and there is a network cost. It is impossible for me to calculate the cost, understand it, and inform the client about it.

We are never sure of our end cost. Sometimes the cost blows up and ends up being five times more than what we calculated. There is no way to understand the cost. I want the solution to have a better billing system and dashboard to understand the billing.

The solution could be more intuitive. If I'm using a service and my backup has gone down, there should be a notification or suggestion to help me understand what I can do. It should explain how I could have avoided the issue. EC2 is huge, and we'll never know all its services. The tool should have a way of suggesting to me what I can do better.

It can be through notifications, tutorials or through their customer service team. The product could provide CI/CD services in the future.

I have been using the product for seven years.

The tool’s stability is perfect.

The tool is scalable. It can be scaled according to our needs. I rate the scalability a ten out of ten. In our company, more than 100 people use the solution.

The technical support team is pretty good. The response time is very quick.

We have used Azure, and it provides similar services. However, we do not have a lot of experience in Azure.

The initial setup is super simple.

I rate the pricing a five out of ten.

The billing is really painful. The product must provide some intuitive notification to inform us if we are not doing something the best way. Overall, I rate the solution an eight out of ten.

It has helped our organization greatly and especially on weekends because we have many transactions as our users are buying some kind of tools and paying online.

The solution is easy to use. When you have nodes like Python, it's easy to plug and play.

The documentation and usage for the users can be better because for new users it can be very hard to understand and use the solution. They can add small images and accessories.

I have been using the solution for two years.

It is a stable solution. I rate it eight out of ten.

It is not a very scalable solution. I rate it five out of ten.

The technical support team is good but if you are looking for any documentation then it is not very helpful.

The initial setup is easy and it is deployed on the cloud. 

I have seen ROI with the product's use. 

The pricing is good.

I rate the solution an eight out of ten.

We are using the solution to run our apps and services. We make use of Docker to run our API services. So, we use ECS as the container service to run all of our Docker images.

Its most valuable feature is the ability to work with Docker images. We just have to give it a Docker image and run some parameters. It keeps the process going for us. It's been great. We don't have to worry about where and how it gets deployed. We just know that it's going to run and we can contact it at a particular address. It integrates perfectly with AWS's service mapping feature and allows one ECS container to contact any other ECS container using a service name. It's almost like an internal DNS thing. If we want to expose it to the internet, we just put a load balancer. It passes requests from the load balancer to the relative ECS services. It also helps for scaling–we just have to add one in a new container. That's pretty cool.

I see they have changed the console a bit to make it easier to use. Probably, they should include automated graphing, and monitoring solutions. Currently, I have to set up all of those by myself.

We have been using the solution since October 2021.

The whole engineering team of six people at our organization uses the solution.

There were some issues with the setup. It wasn't the easiest to get going as opposed to something like SFTP Gateway. It's a learning curve. You need to learn what a cluster is, what a task is, and what a service is. There are no industry standard concepts that could be applied to different technologies. You need to learn ECS-specific concepts. It's similar to Docker Swarm or other resource management technologies like Kubernetes. They have similar concepts, so if you come from that world, you might understand it a bit easier. There is some AWS-specific stuff that you just need to learn. Once you get over that learning curve, then it works pretty nicely. I would rate the setup procedure as a four out of ten.

The price of ECS is not bad. It is based on usage. It works on a per-hour pricing model.

I would definitely recommend the solution. Once you get the procedure right and set a pattern, it just works. I have used different use cases, languages, and frameworks. Everything works nicely once you get the pattern right. The solution scales by itself, and you can set the threshold of how to scale it. Once it goes to 80% of CPU for five minutes, it scales up automatically, and then if the CPU usage goes down, it scales down. You don't have to pay for that upscale cost over time. You need not have to worry about servers or machines. You just have to worry about your containers. It's great.

Also, deploying to ECS is pretty easy via whatever CI you use. It's generally quite easy. It also integrates with Amazon's ECR service, which holds the Docker images used for the services and containers. It works nicely with the ECS.

If you have credentials or secrets that you might want to use as environment variables in your applications, you could inject them at run time into your containers. ECS caters to this with a pretty easy task definition like syntax. You just have to specify the secret name in the task definition and ECS will know when it starts, and when it needs to pull this secret into the container at run time. Thus it's not vulnerable to an attack where someone can snoop on your task definition. It's not in plain text and it only works if you give your container explicit permission to pull the secret.

I would rate the solution as an eight out of ten.

I am using AWS ECS combined with AWS Fargate in order to orchestrate our containers. To explain what that means, I will give an example. Suppose you are running a microservices architecture in which you have multiple containers within multiple services. ECS is the tool that helps you orchestrate all of that, by providing you with the ability to scale up / down your containers and monitor them. Also, with ECS, you can identify any problems or bugs with your containers.

We had a microservices architecture operating in this manner, and we needed some kind of solution to help us manage and orchestrate all our containers. Here we had a choice between something like Kubernetes or ECS, and we chose to use ECS because we didn't have a very large or complex architecture with many containers. Instead, we only have a few containers, yet they still needed orchestration in many ways, which ECS was able to provide.

For me, the best feature of ECS is Fargate because I don't have to manage anything. Instead, everything is managed by AWS and all I have to do, in essence, is configure my containers and deploy them.

That said, whether this feature is appropriate for everyone depends on the sensitivity of your data. I was able to make good use of Fargate since I do not have especially sensitive data running on the containers.

Visualization is an important factor for me, and I don't think that the visuals within ECS are good enough because it doesn't show you all the details you might need to see at a glance.

Another aspect that could be improved is that our monitoring within ECS depends on other services like CloudWatch, for example, and with this arrangement you have to constantly switch between screens when navigating around. Because of this, it's difficult to take a look at your monitoring data, or even just to set up, and it can be very frustrating.

Perhaps it would be possible to improve this situation by having the ability to include everything on one status page. For example, if there is a service or container that is exhibiting a bug or is in some type of loopback mode, when I click on it, I could be shown the screen with all the errors displayed right there.

I didn't see any issues, especially in regards to ECS with Fargate. There are two ways you can run ECS; one way is with EC2 (Elastic Cloud Compute) and the other is Fargate. When you run ECS with EC2, it means that you manage your own servers and your own containers. However, with Fargate, AWS manages all the servers and containers behind the account. For some people, it might be the case that when you're using ECS with EC2, your issues are more related to your own management methods rather than anything to do with AWS itself.

I contacted support one time when I needed to find some information related to launching containers. I wanted to find out how to access certain container data, such as container ID and so on. Once I reached their support, they managed to help me find this information.

The quality of support you receive depends on your subscription because when you have an enterprise subscription, their response is very quick. However, when you are not part of an enterprise, they might not help you right away. Regardless, when we're talking about how well they help, they are indeed able to offer technical solutions and they are straightforward about it. 

Positive

The setup is not that hard. When I began using it, I found that it's not very intuitive, but at the same time it's not difficult.

I would suggest, however, that AWS offer more courses or tutorials on how to use it. I know that they teach how to use ECS through their certifications, but even a few two to three minute tutorials on their YouTube channel would help people a lot. Personally, I had to look through many resources to find good knowledge on it, but when you eventually find the solution, it gets more intuitive. 

When it comes to how long it takes to deploy your containers within ECS or their orchestration tools, this really depends on your architecture, and how you have set things up. If you have set everything up properly, it takes only a matter of minutes to have your application up and running.

I don't know the exact amount we were charged for our use of ECS, but I do know that it can be costly, especially when there is a bug or an error caused by default configurations. When you configure your containers to be launched with special configurations, such as with CloudWatch events, sometimes they fail to launch and they enter into a locked state. Each time this happens, all the configuration behind the container creates itself again, making it such that the costs can quickly go up if you have any bugs in your configuration.

Besides that issue, I would say it's not that expensive, but can still be costly in a way.

We considered using Kubernetes instead of ECS, but we went with ECS because our architecture and amount of containers did not warrant the complexity of using Kubernetes.

My advice, first of all, is to do your research, and do it in detail. Verify that ECS really satisfies all your requirements, especially when you are also using Fargate because with Fargate you are not managing the servers yourself. A good tip is to watch some tutorials that already exist online, so you can start your process with that.

To be honest, using a tool like Kubernetes to orchestrate your containers can be a very difficult process, especially when setting up Kubernetes clusters, and there are a lot of small things that you need to do and understand. However, compared to Kubernetes, ECS is very intuitive and extremely quick to learn. This is why I would definitely recommend ECS over Kubernetes if you don't have a highly complex microservices architecture and you simply need to set up your containers quickly.

I would recommend ECS mainly for its stability and its ease-of-use in helping to manage containers, despite that there are some improvements that they could make, such as better visualization and other improvements to the technology itself in order to orchestrate even more complex architecture.

Overall, I would rate AWS ECS a six out of ten.

We use Amazon EC2 Container Service for Kubernetes and hosting sites. 

Amazon EC2 Container Service is a flexible product from Amazon. You can put it in the Auto Scaling group for high availability. There are also a lot of choices for pricing. 

The product can become expensive if you don't choose what you want. 

I have been using the product for two years. 

I rate Amazon EC2 Container Service's stability a nine out of ten. 

I rate the tool's scalability a nine out of ten. 

Amazon EC2 Container Service's deployment is easy and takes three minutes to complete. 

I rate Amazon EC2 Container Service an eight out of ten. 

Our services are relatively straightforward to manage. We have a limited number of servers compared to some others. When it comes to billing, our plan is flexible, but it may not be the most suitable option for medium-sized enterprises.

They handle the backup process quite well. They automatically encapsulate it, including container backups, without relying heavily on the client's involvement. This is a significant advantage compared to other providers where clients often need to manage the process more independently. It's a feature that I find suitable and beneficial.        

For instance, we've opted for a fixed monthly fee, which covers us for a broad range of data usage. However, with Amazon, the billing tends to increase exponentially as data usage grows, which poses some financial challenges for us. We've also integrated AWS and Synchronoss services into our setup.

I also believe there are limitations in terms of upgrading. The software has the concept of dedicated servers that you can manage. However, an issue arises when you can't match one operating system with another that you've already purchased. You can't simply merge them; instead, you have to buy a completely new one. This limitation has caused some challenges for us.

I have been using Amazon EC2 Container Service for two years.

Initially, we operated our servers on-premises. However, as we aimed to improve the quality of service for our clients, managing local infrastructure became challenging. To address this, we made the decision to move to cloud providers. We chose providers like OVH, where we procured Bandwidth servers and Ultra servers. These providers offer a flat-rate billing plan, which includes servers with specific resources like RAM, processors, and databases. You configure your setup, and they provide you with a monthly bill. On the other hand, Amazon's billing system is more dynamic. It adjusts the billing based on the demand and usage of the server. This dynamic billing approach has posed certain challenges.

There are no specific issues with the service itself; it's quite good. The only concern we have is the billing structure, which may not be the best fit for medium-sized enterprises like ours. The service is more aligned with the needs of larger corporations.

My advice would be for them to stay up-to-date with the current market trends, especially in the cloud services sector. While Amazon is a leader, it should be open to adapting and being competitive in the evolving market. The flexibility in server configuration is crucial, and the billing plan can remain as it is. However, most cloud server providers tend to have similar billing plans based on the server's performance and usage, so it's essential to align with these industry standards.

I rate the overall solution a six out of ten. 

The solution is to compute instances. Basically, it's a debit. It's a host. For example, if you need a computer that you want to install your enterprise application, instead of buying a server on-premise, you can spin up an instance online.

I like the per-second billing. Every second I use is billed. Any other time I am not using it, it's not billed, so it's better. It’s cheaper. It's good.

It’s easy to set up.

The solution is stable.

The solution can still be expensive, even with per-second billing.

We’ve used the solution for the last year.

The solution is extremely stable and reliable. There are no bugs or glitches. it doesn’t crash or freeze.

It is very scalable.

We started with AWS this year. We have one client so far.

Technical support is perfect.

The solution is very straightforward. The initial setup is very cheap also.

It's very easy to replace. You don't need a lot of teams to do it. One person can do everything.

We are integrators and can handle the initial setup.

You pay for only the time you have used the on-demand service. For example, if you are using software that does a lot of computation or analyzing your data for your company, you only pay for that time spent analyzing that data. The rest of the time, you don't pay for it.

Although I like the per second billing, the cost of it, even if you are being billed per second, it is still higher than competitor products.

We are integrators. We use a lot of Amazon products.

I would rate the solution eight out of ten.

I would recommend a competitor's product. I would recommend, for example, Oracle Compute Instances or Oracle Cloud. I wouldn't recommend AWS EC2 due to the cost. It can get out of control. For example, once I left my EC2 on for a few days with some services running, and in three days I lost a thousand dollars. It's quite costly if you are not very careful.

I've primarily used the solution for web hosting. It can also be used for development purposes. 

It's very easy to deploy.

The solution is simple to access.

The stability has been very good.

It's a product that can scale as needed. 

There is nothing I would like to change about the solution at this time. 

The pricing could be a bit better. The Amazon Cloud expense can get high.

I've been using the solution for one month.

The stability has been good. There are no bugs or glitches. It doesn't crash or freeze. 

The solution can scale. It's not a problem.

Right now, I'm the only person who uses the solution. 

I've never used technical support. I can't speak to how their services are. 

We did use a different solution. It was an on-premises product.

We have found the product easy to deploy. It's not overly complex or difficult. 

In terms of the implementation, for the object storage, the installation is mostly configuration. Once you buy the Amazon Cloud, this feature is available, so the only thing you'll do is you'll need to choose which type of E-service you're going to use.

The solution can be handled internally, or a company can install it with the help of an integrator.

The pricing depends on the region in which you're using it. Most probably if you're using it in the UAE or South Africa, the price may get lower.

I'd recommend the solution to others.

I'd rate it an eight out of ten. We have been happy with it.