We just raised a $30M Series A: Read our story

Amazon AWS Competitors and Alternatives

Get our free report covering Red Hat, Microsoft, SAP, and other competitors of Amazon AWS. Updated: September 2021.
540,884 professionals have used our research since 2012.

Read reviews of Amazon AWS competitors and alternatives

Julio Graham
Managing Director at Olive Professional Services Ltd
Real User
Top 5
Helpful automation scripts, good documentation, responsive support, easy to use and manage

Pros and Cons

  • "The ability to fire up a virtual machine, use it, and then kill it, is quite a valuable feature for me."
  • "Before they changed the dashboard, I found some of the more granular options easier to find."

What is our primary use case?

Linode is a service that hosts virtual machines for you.

How has it helped my organization?

It is important to me that Linode offers a small, but well-focused set of cloud computing services, and there are two reasons. I've got a lot of choices because you can do a lot of things with Linode, but the core of what they offer is the ability to host your own cloud.

The size of the applications that you can use is significant. For example, you can run some massive infrastructure through Linode if you need to. The focused set of services is important because if you compare with Amazon, for example, they have AWS cloud, and it has a lot of things that get really very confusing. I understand technology, but I'm a business person more than a technologist. So, for me, the fact that they're focused means that they are working on the latest technology.

They are also practical and don't just use a component because it's there. They've got specific paths, such as migration paths, and they understand what I call fundamental IT. They understand that very well and it's invaluable. For example, one time I had to contact support because someone on my end deleted something. I phoned them at two o'clock in the morning to open a support ticket and a human being answered the phone. They said, "Oh yeah, no, we can do that. All done." Ten minutes later, the problem was resolved and life was fine. Essentially, they understand the practicalities of IT the way it should be in terms of the fundamentals. This is something that a lot of people don't understand.

An example that I'm thinking of is where you have a large call center that is located somewhere else in the world and people don't speak English, which is always a problem. They've got scripts that they go through and based on what you describe, they try to tell you the problem or ask you more questions. You'll be forced to go through a poor process, whereas Linode doesn't do that. Linode is just human beings who are IT literate, for want of a better word, dealing with you at every step of the way. That's important.

To me, it's very important that they haven't lost their focus, even though their capabilities seem to have expanded quite a lot.

Another way that Linode has improved the way my organization functions is that I don't have to have the physical infrastructure, anywhere. This means that I can work from anywhere in the world with the same infrastructure. Whether I'm in the United Kingdom, whether I'm in South Africa, whether I'm in Costa Rica, makes no difference. This is a massive advantage.

Many people think that because it's the web, and it's all-pervasive, it is just "old hat". But to have infrastructure that you can access from anywhere is absolutely brilliant. So, that's one area.

I used to host all my own stuff but I've gotten to the point now where I think I've got one server in-house, and that is full of what I call pet projects. I don't even know if I've got backups of that.

On the topic of backups, human beings don't like doing them. Also, IT people set backups and never look at them again. At least at Linode, you choose their backups and you know you've got three backups a day that you could fall back on, in the worst case. It's brilliant for me from that point of view.

Because of the ease of use, I can offer it to more clients. It's a doorway for me to customers where maybe I would have had to have a much more technical staff. As it is now, I don't have to, and it's not a deal-breaker. The customer is not saying that I don't have a server expert because I do. It's Linode.

In terms of helping me to accelerate innovation, Linode has got a whole bunch of storage options now. They've got features where you can attach data in different ways, which is something that they have addressed in the past couple of years. You can have Amazon buckets, as well as other clever things. I don't know whether they were the first to support or offer things in this way, but they definitely made the accessibility to some of these more obscure storage options easier. If you want to attach to an S3 bucket, it's always been quite a challenge, whereas with Linode, you just put in your credentials and it'll attach to the bucket for you. From that point of view, I see them changing the underlying technology constantly because you see the upgrades as they come through. Without even having a strategy to remain on top of things, Linode has enabled that for me.

If I want to fire up anything that is attached to any of the common data types, it's not a difficult thing to do at all because that is what they're doing. For example, I've just seen something new on their site, which is called a cloud firewall. It's in beta testing. So, looking around, I can see that there's going to be a new service that they're going to add back on top of all of that. It means that if I've got a cloud firewall, I don't have to worry about other firewalls on my machines. I just stick them all behind one firewall. This means one setup, one cost, etc. That's innovation for you because, one, they're making life easier for me. I don't have to set up a whole bunch of things. Secondly, they make a new income stream for themselves, which is brilliant.

What is most valuable?

The ability to fire up a virtual machine, use it, and then kill it, is quite a valuable feature for me. They have a lot of startup scripts, I think they are called stack scripts, whereby you can install something at a click of a button. For instance, you can install a whole server at the click of a button. Linode gives the users a lot of control.

Another thing that I use quite a lot is their documentation. They have documentation on how to perform tasks and often, I use that to educate a client on how to do something or on how to maintain something, et cetera. This is because a lot of customers are used to simpler systems like an iPhone, where if you want an app then you just download it from a store, press a button and it installs, or press a button and it updates.

The truth of the matter is that with servers and things like that, whilst people like the push button idea, it's a lot more complex than that. With these stack scripts, the people at Linode have thought about all of the things that a new user would not think about. They do all of that stuff and then walk you through it, and that's where Linode's documentation is really good. They walk you through what you have to do to secure a server, what you have to do to run a patch, or whatever.

They've got all those sorts of knowledge bases of information, which I think is invaluable, especially for clients who are uneducated in these things.

It's extremely important to me that Linode offers worldwide coverage via multiple data centers, for various reasons. One is that because we live in this global world, our customers are everywhere. Secondly, for people who need geo-redundancy, with for example a server in China and one elsewhere as a backup, it's great.

It's also nice because if they were just US-based, I wouldn't be able to use them because I would need to go through a whole process of trying to certify the data integrity in other regions. I'm sure that most people wouldn't bother with this because of all of the EU laws and the UK laws around data privacy.

The US's data privacy laws are far more relaxed than what they are on my side of the world. The fact that I can have a server in London means that I don't have to bother with all of that. My physical location of that server is in London and to me, it is really important.

When you compare Amazon, they claim to have infrastructure all over the place but I think that the bulk is centered in Germany. Even if it is in a few different places, everything gets backed up to the US, which is a problem for a lot of people.

What needs improvement?

Before they changed the dashboard, I found some of the more granular options easier to find. However, it was just a matter of getting used to the new interface.

For how long have I used the solution?

I have been working with Linode for more than 10 years, since 2009.

I've got two use cases. The first one is that I use Linode to host my own personal servers that host programs and software. Some of the applications are things like firewalls, et cetera. Any offsite technology infrastructure that I need, I use Linode for.

My second use case is that when I provide my customers with solutions that are cloud-based, where it is customized software, websites, or something else that they want control over, I spin up a Linode for them and then hand it over to the customer. I'm constantly giving Linode new customers as I provide people with solutions. I normally spin the Linode up myself and then pass it on to the client. Once my relationship with that customer is over and my side is fulfilled, they continue a relationship with Linode as their infrastructure provider.

That said, the main objective is not to resell their product. What I'm doing is I'm including Linode as the hosting infrastructure in the solutions that I sell. This allows people to have a virtual machine in the cloud at a very reasonable price. For example, a lot of people use that for their bespoke websites, membership sites, or for hosting other services.

Our company makes some bespoke software for the music industry, for example, whereby they can manage contracts and things like that. When people purchase this, instead of giving them a physical machine in the office, I give them a Linode machine and my software is on that, which is how Linode gets bundled along with everything else.

I don't make any money on top of the Linode service. Rather, I just say to the customer, "You've got to pay Linode $20 or whatever it is a month to carry on using Linode." At that point, they say, "Absolutely," because they've got backups and all of the good stuff without any of the downsides, for literally a tenth, or in some cases a hundredth of the price compared to having to do it on their own site.

What do I think about the stability of the solution?

The overall stability is brilliant. I don't think I've ever had downtime that I wasn't in control of, for upgrades as an example. I don't think I've ever had an outage. 

There have been some DNS problems once or twice, but not a single one of my clients phoned me and said, "Hey, this is not working."

I use Linode daily. The machines serve all of my websites, some of my client's websites, some of my user applications, and some of my knowledge base stuff. I've got customers who've got critical data, such as their financial data stored on Linode. To me, Linode is like having your coffee in the morning or breathing. It is integral.

What do I think about the scalability of the solution?

I think that if I had to do something massive, it would be easy. Scalability-wise, if you've got the money, the sky's the limit with them. I'm guessing that if you were to try and buy some of their massive offerings, GPUs, which are $4,000 a month, I'm guessing that'll take a little bit of time to set up, but I may be wrong because I've never done that.

Basically, if I look at the options that they have, you can move from anything from a tiny one-gig configuration to something that's got 125 gigabytes or 200 gigabytes of memory. That's ridiculous. There is more RAM than you've got hard-disk storage in some cases.

There are also a lot of articles that help people. If you want redundancy, for example, you're going to go and have a look and see, "Okay, that's how I do that.", and then you've learned something. That's the beauty of it, is that they're not like all these proprietary people, everything's under a hat and you don't know until you pay your money. They're quite open about everything.

I have customers who use Linode, but I am the only person in my organization who uses it at the moment. I have one other software developer who has access to it. At the moment I've got one machine on Linode, which I've just recently made slightly smaller. It runs about 40 or 50 websites and web applications for myself and for some of my clients.

Every time I get a new project, I fire up a new Linode. I use it for as long as I need it for the project and then I kill it again. So as soon as I get my next customer, I'll be adding another Linode to my account.

My current Linode that's running now, I recently down-scaled because a lot of customers have moved away, and also, we're trying to save money wherever we can. I didn't have to go to Linode and explain that I was suffering because COVID has taken away some of the customers. Rather, I went onto my dashboard, selected to make it smaller, and they warned me to ensure that I had taken care of three steps first. I went ahead and took care of the one that I hadn't done, and the other two were not applicable to me.

Once that was complete, I pressed the button and my Linode was half the cost in less than 10 minutes. I have got control of everything that I do. If at the end of next month things are back to normal, then I'll press the button and double my usage again. I expect that I'll be adding customers to it after that. In the meantime, I have the freedom to do what I want and I'm not captured by any sort of contract. I trust the people at Linode explicitly.

How are customer service and technical support?

Linode has 24/7, no-tiered human customer support, and its flexibility and overall responsiveness are why I don't use the other major providers. They are very responsive and are quite happy to listen to what you have to say, rather than trying to rush you through the process. You get a very real and very human interface, even with their email systems and through their ticketing systems. All of that appears to be very considered and it appears to be unique to you. You're not just getting a knowledge base thrown at you and told, "Go and sort through that and figure out what's going on."

Which solution did I use previously and why did I switch?

I also have some experience with Amazon AWS, and it is more complex than Linode. I know a little bit about the AWS pricing, where they bill according to time, at perhaps the millisecond level. I did use them in the early days but I got burned a few times where I suddenly had a bill of $1,000 when I didn't even realize that the machine was on. Amazon was quite unforgiving, at least in those days, about such things.

Linode, on the other hand, has a maximum cap that you can pay. If you use less, you pay less. I think that they're quite smart in that they worked out that cap quite well because you never really pay much less than that. It's a fixed price and whoever did those calculations has done them well.

I have no idea how much I have saved over the years by choosing Linode, but just from a management point of view, no one's time is spent having to double-check how much your bill's going to be all the time. I would estimate that would be one or two hours, of someone who knows what they're doing, a month having to do that. If you take a support person, earning, for example, $5,000 to $8,000 per month, if you consider what two hours of the month would cost, that's probably the cost of the savings. If you manage these things properly, obviously, you can keep the costs the same. Honestly, I haven't put much thought into it or wondered if anybody had better pricing because Linode does everything that I need, nicely.

Generally speaking, as I have been a customer of Linode for a long time, I don't have a lot of experience with other cloud providers. I don't even bother with them. But, I can tell you from seeing customers that have had other providers that I've got absolutely no problems with response time, in comparison.

I don't ever run anything that's mission-critical to the point where I need support in five minutes with someone. I don't have anything like that. None of my customers do because we build our systems so that there's redundancy. If something goes down, it's much like having a backup for your electricity. It doesn't really matter if it's down 5% or 2%, because you've got a backup.

That's the same with the way that we design solutions for our customers and for ourselves. If Linode does go down, normally, they're not down for long. When I say not for long, it's minutes. They communicate well and for the maintenance they do, they let you know months in advance. So, it's only your own fault if you suddenly find yourself caught in a situation where you're not prepared. Ultimately, they've got a very good balance and I think that they have got a very good future.

An example of this is that one of my machines needed to be moved to Tucson. It was a virtual machine that needed to be moved to some other hardware, which is a process that they said could take up to two hours. They said to me, if you do nothing, we're going to move it in April or whenever it was, but you can click here and move it at your convenience. I know when the US is not so busy because I'm half awake when they're still sleeping, so I pressed the button one morning and it took less than 20 minutes to move.

My machine was back up and running. Nobody knew, in terms of my users, that it had even happened. So, their response times and their options, because they allow you to manage so much yourself, are great. It's a very good balance between moving forward and balancing that with the needs of the customer.

How was the initial setup?

Linode is a virtual machine, so I'm often setting them up for the first time. I have images that I set up to be Linode in a particular format, and I can run them by hitting a button that says Deploy New Linode. It asks me what size, where, I press a button and it deploys that machine that I had deployed two years ago, exactly the same way. It means that if you're doing repetitive stuff, you shouldn't have to be doing it from scratch every time. You can just make an image and deploy your Linode from it. They've really made it quite easy.

Setting up a Linode from scratch is child's play. It's literally easier than updating your iPhone. But, it's what you have to do after that you need to consider. Depending on your choice of what you want to do, such as installing Ubuntu or some other software, that's where the complexity comes in. However, that's not Linode's issue. That's an issue with whatever it is you're doing.

The actual acquisition of a machine takes just minutes. I've got six questions to answer and when I press that button that says create, it takes approximately a minute and then you've got a machine that has an operating system installed. It's a basic operating system, whatever you choose, and a root password. It is at this point when the complications start because that's when you then have to do all the stuff that you do to set up a machine, but that's not a Linode function. That's a user function.

For what I do, if I start from scratch, without my pre-installed stuff, it could take me anything from 15 minutes for a basic system to about four hours for a complicated one. But, that's after the Linode set up. For example, setting up a Laravel dev environment probably takes me about three minutes, because I just pop an image and I say, "Deploy this image," and it just goes and does it. If you take a look at their marketplace, they've got apps that you can use to set up various machines.

In the marketplace, there are apps available to help you set up a cloud server, an open VPN, Jenkins, LAMP, or anything. You just say, "This is what I want," and that script runs for you. Another example is Minecraft; if you want the Java edition of Minecraft, you can go and set that up and it'll take you five minutes. Any customer can do this and you don't have to be a tech guru.

What was our ROI?

It is tough to determine what my return on investment is. I think there's a cost benefit more than a return on investment. When you look at an organization in terms of infrastructure setup, I've had a much lower cost than I would have, had I used the traditional bare metal or hardwired approach.

Compared to a traditional method with a server, plus someone running it, and all of the administration that comes with it, my cost savings are literally a hundredfold or a thousandfold. Moreover, without Linode, I probably wouldn't have been able to offer some of the services to some of the markets that I have in the past because of the pricing.

What's my experience with pricing, setup cost, and licensing?

The monthly cost depends on your requirements.

The pricing is absolutely spot on. I think whoever thought about how their pricing should work and how their sizing should work, clearly understands the use case of their customers. I'm a tiny customer of theirs, but, I'm loyal to them because they constantly deliver and they do things that make sense.

I'm not paying any more now than I was paying in 2009. As a matter of fact, I think I'm probably paying less now than I was then, yet I've probably got 20 or 30 or 40 times larger resources available to me. Every time they upgrade their infrastructure, they don't say, "Oh, we've got new equipment, now we're going to charge you more." Instead, they say that they will carry on charging you the same or less, yet, we have better stuff for you. That is just a good business model, which has stood them in good stead.

What other advice do I have?

Because I am such a happy customer, it is difficult to point to an area that is in need of improvement. I've not had a use case, and none of my customers have said, "Oh, we would have loved to have used Linode, but they don't do X." What they do is more than adequate for what I need. In my view, you'd have to go a long way to find something. I don't use the API, although I've seen a lot of documentation on it. I'm guessing that the API is an area where people would want different accessibilities, but I don't know enough about that to be able to comment.

They changed the dashboard quite a while back, but there was a while when you could switch between the classic and new one. The classic one had some different granularity, which was nice, but I've now found that with the new one. It had looked like some of that granularity had gone away, but it's just in other places. Ultimately, it was just a matter of getting used to what it looked like.

I think the new interface is more modern-looking and probably a little more user-friendly. However, when you've used something for a long time and then it changes, you think, "Oh, what's going on?" But I've not found myself sitting and wondering what is happening in different parts of the solution. I would say that now, it's fine.

Quite a while ago, they moved away from one particular underlying technology that is used for the virtualization of machines. There are two types of virtualization, and they moved from one to the other. I'm not sure of the details but there was a massive improvement as a result. I could tell because I only moved some of my machines at the time, and left some of them on the first platform. There was a noticeable difference and it was big. Whatever they did in terms of the backend of their virtualization, when they moved from one of the main ones to a better one, that was a good move.

The biggest lesson that I have learned from using Linode is that you can be human and still provide a good service. I don't know anybody in the company and I don't particularly follow any of their leaders, they're not even on my radar. That said, every single interaction I've had with Linode has always had all of my core values in there. Integrity is an important one.

I find that often with tech companies, they lose humanness for the sake of efficiency, or other reasons. Banks have also lost it. A lot of people have lost that human touch and whilst I don't think Linode in any way have ever said, "Oh, we're the company with a human touch.", I can tell you that they definitely, you can feel the love and the fact that people know what they're doing and they care about what they're doing.

My advice for anybody who is thinking about using Linode is to know what you want, and if you don't know what you want, ask Linode because they'll know what you want. Often with these sorts of things, you tend to do your research first and then go to a vendor and ask them, based on your research, what you want. I would be very comfortable suggesting to any customer of mine that they ask Linode.

I am confident that they are not going to try and figure out how much you can afford and then nail you with that. They'll give you the options. They're very transparent. That way, you end up buying what you need as opposed to buying what someone's trying to sell you or what someone thinks you need.

In summary, Linode is a good product and I love them.

I would rate this solution a ten out of ten.

Disclosure: IT Central Station contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.
Flag as inappropriate
MT
Principal Consultant at a computer software company with 10,001+ employees
Consultant
This helps us meet multiple requirements other PaaS solutions do not but there is a lot of room for improvement

Pros and Cons

  • "It is a flexible solution that is straightforward to use."
  • "Stability can suffer in the context of a large architecture."

What is our primary use case?

I work with our enterprise architecture. In my network, there are almost 400 total applications. I have been working here for almost six months on a network migration and in those six months, I have been working with many of those applications that have been included with the involvement of Azure in the migration.   

We are migrating everything from the old network to a new architecture. There are multiple teams that I work with and people work with me throughout the organization. I review all the target architectures and the deployment and everything that comes along with the pieces of the migration that involve Azure. Any issues, large or small, I have to look into. These issues might be simple certificate issues or they may involve multiple interfaces that need to be used for a solution.  

Because we have a very complex system, it is not easy to complete the migration. The landscape also has a mixture of different technologies and platforms. If I have to customize, I just get a Terraform script or ARM template from a developer who is assigned to that task. I review all that stuff that they give to me.  

When we went to the version of Azure that we use now, there are certain solutions that we created. If we had trouble, we worked with Microsoft to create that solution for our organization and the problems that needed to be solved.  

We define our own solutions with Microsoft that are not available in the open market. Because of the way we have used Azure, we do not really have a very focused end-product. It is a highly customized product that we have built using many tools.  

Azure is now a mixture of solutions. There are certain applications, which are IaaS (Infrastructure as a Service) applications, where we just go and use them. Then there are certain applications that are a mixture of IaaS and PaaS (Platform as a Service). For certain parts, we use private clouds, public clouds, or hybrid clouds. We originally wanted to use more public clouds, but as we proceed, we are moving into more hybrid mechanisms. In the future, I don't know exactly what direction we will take because the technologies and the climate are changing so quickly.  

But right now, we are only using Azure with images being created from the existing architecture. For Azure, we use private cloud, public cloud, and mixed, or hybrid cloud as needed and all of these work together.  

In the future, we may go for some specific function-based services or even open-market APIs. We can use open APIs with Azure. API management is also possible. So there are a lot of permutations and combinations that go with each application based on sizing and NFR (Non-functional Requirements) validation.  

For Microsoft Azure, we use the product itself as a platform, I work mostly with their services. These can be PaaS services or DNS services, monitoring services, storage services — basically all the supporting services that are available to us with Azure. Anything that is not available, we try to build on PaaS. If the services we want are not available, I have to do a complete fabrication.  

So we use mostly PaaS services for most of the supporting services and then we work further in solution optimization, which is something we can accomplish through Azure. Ultimately all that depends on the budget. If a company is ready to spend on a cloud solution, an ROI (Return on Investment) model helps. The amount of customizations and the real need for a solution comes out of the realities of the ROI.  

Our contracts are based on supplying solutions for what the customer needs. If they have selected that a particular application will be available and make this a system mandate which we have to flow, then we have to keep those applications. Azure is one of the tools that we are using to help make these kinds of customizations and to meet their expectations after the migration.  

How has it helped my organization?

Azure gives us a different form of PaaS to work with during our migration and helps us to meet multiple requirements that current solutions do not provide in any one product. 

What is most valuable?

One of the most valuable things about Azure, I think, is that it is pretty straightforward. There are well-defined processes and it is not a bad product to work with. I only work on Azure right now most of the time. I cannot directly compare it with other solutions in the present situation because it is not always practical to consider every solution. Certain platforms on the market are very strong with other services. For example, Kubernetes on RedHat Openhift is better for working with AWS. But I have to ask from a usability, a complexity and a budget standpoint if that is really required.  

If I do my work and my applications are sorted out well in advance, I do not have any issues. From a user perspective — not from a cloud architect or enterprise architect perspective — my requirements are being met. As long as these requirements are met, I do not see anything as a showstopper. If there is a showstopper which I think I absolutely can not solve with Azure and I think another solution would handle, then possibly we may go into a multi-cloud scenario.  

That is also a limitation for our organization. The goal is never to seek complexity. Personally, I think there is no direct comparison between what solution is better and what solution is worse. There are only solutions that work or are capable of doing something and those solutions which can not do it, or were not designed to do it, or do not want their product to do it, et cetera.  

Part of my place in working with these solutions as part of my process is working with products I am comfortable with. So the more that I use Azure, the more comfortable I get with what it can do as a solution, and the more comfortable I am using it. If I started using AWS more, I would get more comfortable with AWS and maybe incorporate that more heavily in the solutions.  

What needs improvement?

There are some small things that could be done to improve Azure. I think they should actually do more to implement function as a service. It is a completely separate capability that they currently do not address. Function as a service can be a completely different scheme altogether than PaaS or IaaS which it does quite well.  

For an example of a FaaS, I think the Azure product can be stronger in terms of storage. I would like to see it have better management systems as a service specifically for managing documents. Right now they are handled as a more generalized object.  

Say Azure came out with Microsoft Document Management and it was very strong as a service. It would not have to be deployed as a complete infrastructure. I would be able to use that as a service inside my organization and it is a product that any organization can use.  

The question is what is the separate USP (Unique Selling Point) that Microsoft will provide to the user that would fit a unique need when making FaaS solutions available. Document management systems have already been proven to be very popular by Google. Microsoft Office uses OneDrive storage. There may be a better way to promote document management in a more general PaaS. Sometimes it is very useful to virtualize a platform or an infrastructure, but in the same way, it is sometimes valuable to virtualize a function. Applications may be a collection of functions.  

It is this type of branching out of services that Azure can do within the structure they already have.  

They are targeting Azure into specific domains and not working as much with open-source as they could. That would be helpful. I think eventually this approach will just drive the competition away. If I have a product that is very good for manufacturing as a function — something like is being done with Edge — it might be beneficial for Azure to be able to tie in this FaaS and let manufacturing clients start working with the solution without having to reach outside of Azure. Right now that I do not see that happening and it is an opportunity that Microsoft is missing with Azure.  

For how long have I used the solution?

I am responsible for designing our migration, so I have to work with Azure to define the parts of that solution. I had previously been using AWS mostly for personal services so I was familiar with PaaS platforms, but I have now also been using Azure exclusively for the last six months to supplement the functionality we require.  

What do I think about the stability of the solution?

The product is stable. There are a few qualifications attached to that.  

I think the stability of Azure varies depending on the workloads. It is more stable from the perspective of how it behaves in a mid-size deployment. For a very, very large implementation, I have yet to see that same kind of inherent stability. I believe it is because of the complexity of the client's system or architecture.  

You may be able to say that if it is more of a Microsoft product landscape, then possibly it is more stable in general. The more that there is a mixture of technologies, then it will tend to be less stable. No application can be stable in every circumstance.  

As the project I am engaged in is very large, we have experienced some episodes of instability. We solve the stability problems as we go along to a great extent. But I think there are a lot of situations that have to be dealt with in real-time. Though we have direct contact with a Microsoft team architect, it is difficult for them at times to just jump in and solve an issue. You can not usually solve a problem instantly looking down at it from 55,000 feet when the situation on the ground is very, very complex.  

At first, they only have generalized solutions to your problem. I think they need an extension of the existing team. This would be like a core team to work with client organizations to do case studies to define patterns in what is causing instabilities.  

Because Azure is cloud technology and cloud comes with its own problems, these bleed over into Azure stability. All these patterns that contribute to instability have to come out in order to be solved. As Microsoft collects more case studies and more knowledge of where these problems tend to occur, this should enable them to stabilize the product against those issues.  

Overall, I would say Microsoft Azure is a stable solution, but even as a stable solution, it usually has some bugs or glitches.  

What do I think about the scalability of the solution?

As of today, we have almost 1,000 people using the solution. We have a very big migration project that will last for the next four to five years before it is completed. They have many applications and many users for those applications. If the volume of users or applications were to scale, that should not be a problem.  

How are customer service and technical support?

I do not really have much direct contact with the Azure or Microsoft support teams. We have a separate team for that. I have a great architect that I work with here (Sweeden). But if an issue comes up, the application team goes to work on it to support the resolution. It is their option to contact Azure to raise that issue or resolve it themselves.  

Which solution did I use previously and why did I switch?

I was using AWS before Azure, but I was using it mostly for my own personal needs. I was deploying my own applications. I used it for about two years but not from a company perspective. I deployed my own applications in the public cloud and loaded them there for use at a personal level.  

In the company right now, I am only using Microsoft Azure. The company itself is using everything, really. At this point, my experience in the company is specialization as the person who is helping to utilize Azure.  

How was the initial setup?

The initial setup was simple and it is simple for a simple application. If I want to build with a simple application, I simply go do that. But if I have a very heavy interface-based application, then the choices become more difficult and involved.  

If I have a WebSphere application, that is easy. A complex platform or a complex interface dependence becomes difficult to implement because of restrictions. If I can not simply go and deploy as it is, obviously it is more complex to deploy in the system.  

For a small company with a typical landscape of Microsoft technology, it becomes very easy to work with Azure. It is possible to go through that setup by yourself and test your servers and the entire functionality. 

After deployment, you will require maintenance. We can not simply have a production list and push everything out. You need pre-production, testing, and then deployment. All that has to be done on Azure.  

There are a lot of things you will have to work out with security certificates. Meanwhile, things keep on changing in the product itself. New upgrades keep on rolling out. If the old version does not support the new upgrade, then you will need to get involved with patching and other upgrades to take care of the issues that are introduced.  

We have a dedicated team for maintenance. We know we need to do testing and that is why we created tasks for that. But, generally, I think complexities in the setup depend upon what applications you are building. Simple applications and simple systems make for simple deployment.  

What about the implementation team?

We are working with the vendor directly. We also have contacts with Microsoft. Microsoft directly provides us all the tools and information we need for implementations.  

What's my experience with pricing, setup cost, and licensing?

The pricing of Azure depends on the build of what you prepare. You can optimize everything, and with Azure, you can optimize your utility and costs. For example, say you create a subscription and you want to do more backups and you want a private cloud for that. This will affect your cost differently than if you do not add the backups with Azure or if you add the services with a public or hybrid cloud.  

We have very good, large contracts with big organizations. We do very high-level analytics and modeling to predict outcomes. For example, we may show that a certain solution that we implement with Azure will be likely to reduce a company's cost from the current level to 50% over the next five years. That, to me, is important when considering the cost of a subscription. It is not just the cost perspective that is important, but the ROI as well.  

What other advice do I have?

I would definitely recommend Azure as a solution because it is a popular product by a major brand and it is very easy to use. I think those people I would recommend it to should normally be those who understand the cloud and the advantages and disadvantages. I use it for a lot of things and I do not see any problems. I love it now as a solution so I would recommend it. But if I have a different experience with another very large migration project using a different product, I would have to compare Azure with that. I may get more comfortable with the other product for reasons I have not discovered yet.  

On a scale from one to ten where one is the worst and ten is the best, I would rate Microsoft Azure as a seven-out-of-ten. It is a good product and I love using it but it could do even more and has a lot of possibilities to grow as part of a relatively new technology. The future is more open than closed to the possibilities.  

Which deployment model are you using for this solution?

Hybrid Cloud
Disclosure: My company has a business relationship with this vendor other than being a customer: Partner
SD
Senior Manager at a consultancy with 1,001-5,000 employees
Real User
Helps to simplify your workload with automation

Pros and Cons

  • "This product is currently the leading product within this category."
  • "The product could use more capabilities in the areas of robotics, machine learning and artificial intelligence, but these improvements are already on the way."

What is our primary use case?

There could be many use cases for this product. The use cases will be different depending on the industry where it is employed. It could range from the manufacturing industry to the retail industry. The actual use case could be anything.  

The solution is deployed with two possible methods. One is the on-premises deployment (with AWS) and the second one is the cloud deployment. You follow the software development life cycle for deploying the solution. First, you do the project preparation followed by the solution design. This is followed by the actual realization of the design plan. In the final testing, you engage in three go-live activities. When this is successful, you follow it with an actual live deployment. With the go-live successful, there would be PGLS (Post Go-Live Support), which is nothing but the support aspect. So implementation typically follows the normal software deployment life cycle. The approach could be anything, like support could be a waterfall approach (project activities into linear sequential phases) or it could be a more aggressive approach.  

What is most valuable?

Because SAP is a part of ERP the features range from finance functions to manufacturing functions to quality control functions to procurement functions — anything you might need in development planning. So you can have tight integration between all the functions in which a manufacturing industry operates to develop market and distribute its products. The S/4HAHA platform provides insight into the processes which can become part of robotic process automation. S/4HANA as a platform is enabled for artificial intelligence and machine learning. So you can use it to help simplify your workload with automation.  

What needs improvement?

This is the latest offering by the SAP company. Of course, there are the pros and cons to on-premises deployment versus cloud deployment — especially as far as the infrastructure maintenance is concerned. There is not so much difference as far as activities are concerned. But having said that the cloud deployment is on AWS (Amazon Web Service). It is always the latest offering and the cloud is indeed today's emerging market.  

From what I understand, SAP is already at work focusing on improving the CE (Customer Experience) and the CRM (Customer Relationship Management) as far as S/4HANA is concerned. They are working with having better robotics, machine learning and artificial intelligence capabilities, which can work with this ERP. So there are already new concepts on the way. They are making a lot of use of Python technology, and they are making heavy investments to make it much more user-friendly and customer-friendly. Automation is at the center of their development focus.  

For how long have I used the solution?

We have had an S/4HANA implementation for some time.  

What do I think about the stability of the solution?

The stability of the solution is fine. We have no issues with that.  

What do I think about the scalability of the solution?

For scalability, the skill set required is very specific. You need someone who has a functional and business knowledge along with technical knowledge. A candidate or the consultant who has experience in the difference between the data technological concerns and the business trends can easily go through the entire deployment very easily. Having industry experience is very much required for deployment and scaling.  

How are customer service and technical support?

As a consultancy, we provide some of the technical support for SAP. The support we provide is very much function specific. I am from a finance background. For finance-related support, I could be the key person in our company. For materials management, it would be someone else. Materials and quality management can be handled by a production management expert. One team member may be the marketing, distribution and logistics expert. Each subject matter expert will be deployed on the support of a particular ERP based on their areas of expertise. It is not a one-man show. One person can not run the entire support services as an expert in every area. It is very much specific to the role the candidate needs to play in addressing the subject matter as an expert in that particular function.  

How was the initial setup?

As far as the deployment on AWS is concerned, it is the cloud deployment. It is a little complex. But once it is deployed, the post maintenance and the support services are very easy because the upgrades and maintenance are not handled manually. It is just like on smartphones where the patches get auto-upgraded. In the same way, the patches or the upgrades are automatically pushed to the ERP on cloud. So there is no manual maintenance required going forward. There is no downtime required. For these reasons, clients are attracted to going with cloud deployments. But others are still going with on-premises deployments for various reasons.  

Which other solutions did I evaluate?

There are a lot of competitors coming into the market. The main organizations as far as IT or information technology companies are involved already. We have Infosys, there is Oracle, there is Siemens, there is Accenture, there is Cognizant — all of these are from the IT organizations' perspective. From consulting companies, we have got the Lloyd Group, KPMG International (Klynveld Peat Marwick Goerdeler), PricewaterhouseCoopers, and others who are all into this. There are other up and coming competitors and consultants like BDO Mauritius (Binder Dijker Otte). All these various kinds of organizations are good at deployment and the support of SAP solutions. The main difference is in their approaches to providing solutions that are on-premises or on the cloud.  

What other advice do I have?

The advice that I would give to people who want to use S/4HANA on AWS is that a specific study needs to be involved in the implementation process. It is what we call a discovery session. The idea is to discover all the independent needs for the use of the product. This includes things like the size of the organization, the size of the industry, the number of users expected, the revenue, et cetera. It should also include projections into the future aspects of organic and inorganic growth. The budget may be the main issue for many organizations. The deployment plan should be created based on keeping all these discovered factors in mind. This can affect the deployment model or choices as to whether to go for SAP S/4HANA on the cloud, on-premises or if another ERP may be a better choice. But in any case, all the factors need to be taken into consideration. It is not a black and white decision.  

On a scale from one to ten where one is the worst and ten is the best, I would rate SAP S/4HANA as very near to a ten. Maybe nine because there is no competitor like SAP in the market right now to compare the product to and maybe there is room for improvement. I think that Oracle and Microsoft are looking to close the gap. But, looking at the current market share of SAP, it is at the top so it is near to a ten.  

Which deployment model are you using for this solution?

Private Cloud

If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?

Amazon Web Services (AWS)
Disclosure: My company has a business relationship with this vendor other than being a customer: Partner
LH
Product Manager - SaaS at a tech services company with 51-200 employees
Real User
A cloud service that will please more technically-minded users which is making strides with ease-of-use

Pros and Cons

  • "This platform is popular with technical users because of the abilities for customization and fine-tuning performance."
  • "The ease-of-use could be improved for less technical users."

What is our primary use case?

We are using GoogleCloud for hosting a SaaS platform.  

What is most valuable?

The most valuable feature of the product is that it has a very powerful command line.  

What needs improvement?

It is difficult to say what should be improved because, obviously, they have made some major improvements in the organization of how you do things — such as the way you set up a server. Google has made a lot of effort to try and catch up to the competition in the area of ease-of-use because that would have been my one complaint: that you have got to be quite technical to understand some of the ways that things are done. Azure and Heroku are number one in ease-of-use and they make it very easy. Google has done a lot of work to alleviate that objection and to catch up with Heroku and Azure. But the people that have the most to say about the ease-of-use would be the guys using it. For the developers we have, they like the power and the control that Google gives them.  

So I can not actually answer what exactly has to improve for developers to be more satisfied at this point because they seem quite satisfied with it already. I do not get any complaints from the guys. They are the ones using it every day and I do not use it on a daily basis so I really can not comment in that sense.  

I could say that it can be easier to use for people that do not have the same level of technical skills, but even that has improved a lot with their upgrade to the user documentation.  

For how long have I used the solution?

I have experience with Google Cloud for about two years now.  

What do I think about the stability of the solution?

It is extremely stable. I have no negative feedback and no complaints.  

What do I think about the scalability of the solution?

Google Cloud is very scalable. The use of Docker and Kubernetes has really made it extremely scalable. Google's implementation of Kubernetes is excellent.  

At this stage mainly we have developers and dev-ops using the product and it is a team of about 25 guys. We could expand that at any time.  

How was the initial setup?

The initial setup is somewhat more complicated than competing products. I was not directly involved in that capacity so I can not provide details. But also complexity sometimes leads to opportunity as far as customizing performance. The people who are working with the product directly like the ability to fine-tune more than they want simplicity.  

Which other solutions did I evaluate?

Initially, I was just doing research. I was keeping an open mind and looking at all of the possibilities. I do not think it is quite right to call it "evaluate" when you just do a feasibility study. We did take a closer look at a few solutions like Azure, Heroku, and Google Cloud. We dabbled a little bit with some of them. We eliminated Azure because we were concerned about the support for Mongo DB which is part of our technology stack. Obviously Azure has changed a lot since.  

We trimmed that group to Heroku and Google Cloud which are technically both platforms of high-quality. Heroku is easier to deal with. Using Google Cloud, you have to build a bit of experience with the product because it is not easy if you do not understand how to do things. Heroku makes it a lot easier for you.  

The reason we went with Google Cloud had to do with two things. Number one was cost, and number two is that Google supported everything we use. We had to control the costs initially, so Heroku was pretty much out the door almost immediately. It was a competitive product but it was too expensive. An end-user would not know where a platform is hosted, and they would not care. For an end-user, they go on your website or on your SaaS platform and there is no difference in the experience whether you are hosted with Google or Heroku or AWS. It makes no sense for them to worry about that. But the cost ends up being an important component of the decision for the service company.  

I think the point is that it is very difficult for companies these days to decide between Heroku, Azure and Google Cloud. They all have data centers in the right places in the UK. There is very little that differentiates any of them. Heroku obviously stands out because they are a very stable platform and they do all the hard work for you. If you do not have the expertise to go with a less expensive more labor-intensive solution, then you would go with Heroku and pay more.  

We have a development center in Manila with very experienced guys and they love Google Cloud. It gives them everything they need and everything that is required for a big, fast platform — like the ability to use clustering. I think all of the solutions support Java and ATC (Advanced Analytics Technology). But we have not had any issues since we started on Google Cloud, so we are happy with the direction we have taken.  

What other advice do I have?

The lesson I learned from adopting Google Cloud is that you should do more training before you commit to it.  

On a scale from one to ten where one is the worst and ten is the best, I would rate Google Cloud as an eight-out-of-ten. You can not have a perfect platform. There is always room for improvement and something to add.  

That rating is really because of the feedback I get from the team. I get good feedback from the guys. But it is not really fair of me to give any product a ten if I am not more intimate with it in daily use. I could just as well give it a one if I were totally ignorant of the product, but that would not mean anything.  

Which deployment model are you using for this solution?

Public Cloud

If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?

Google
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Ashutosh Malik
Subject Matter Expert - Data Network at Mphasis
Vendor
Top 20Leaderboard
Flexible, user-friendly, and competitively-priced

Pros and Cons

  • "One of the features that I really like about IBM Cloud is the flexibility where you can order your own hardware."
  • "There is not a lot of support for this solution, which is something that needs to be improved."

What is our primary use case?

We are a managed service provider and IBM Cloud is one of the products that we use to provide services for our clients. We have a large number of customers who are in different industries and some of them are moving from a traditional data center to the cloud.

How has it helped my organization?

There are several advantages to migrating to a cloud-based data center. Some of these are that you don't need 24/7 power or power backup units, local staff for maintenance, or to handle your own security.

There is also the important point that servers are available on-demand. For example, I may need 100 servers for a client for a short time, but after which they are no longer needed. This ability to scale up or scale down using the cloud is helpful. By contrast, in an on-premises solution, if you need more power then you have to purchase it.

What is most valuable?

One of the features that I really like about IBM Cloud is the flexibility where you can order your own hardware. Customers do not want to have a full data center but they can order virtual machines, use solutions like VMware vCenter and NSX, and they can design their own network. 

This solution is user-friendly and easy to manage.

What needs improvement?

There is not a lot of support for this solution, which is something that needs to be improved. 

For how long have I used the solution?

We have been using IBM Cloud for a few years.

What do I think about the stability of the solution?

IBM Cloud is stable and our customers haven't had any issues.

What do I think about the scalability of the solution?

This solution is very scalable. IBM took over a company called SoftLayer, which has resources across the globe. They are available in the UK, US, Australia, India, Asia, and everywhere.

How are customer service and technical support?

There is not a lot of support for IBM Cloud. The response from them and the ETA for resolving issues is too slow, and also, the SLA is too expensive. If you want to have premium customer support then you have to pay more.

Obviously, they will still make the best effort according to the SLA. Whatever engineer becomes available will give good support but it is a very slow process.

For example, something that is a priority-one issue for you will not necessarily be of the same urgency for their support. Rather, they will go as per the SLA, which is based on how much you pay for support. If you are a gold partner then you can just pick up the phone, tell them the problem, an engineer will be assigned and they will just start troubleshooting. For most organizations, it doesn't work this way.

Which solution did I use previously and why did I switch?

I have worked with AWS and Azure clouds. IBM Cloud is the cheapest of these three, and I found it to be more flexible in terms of third-party integration.

How was the initial setup?

The initial setup is straightforward and not very complex. The fact that you can make your own infrastructure is the best part.

What about the implementation team?

Being a network architect, I set up all of these things on my own. There are different vendors who can support you, based on your requirements.

What's my experience with pricing, setup cost, and licensing?

The price of IBM Cloud is very cheap compared to competitors AWS and Azure.

The licensing fees are billed on a monthly basis and some costs vary between cities. For example, data centers in smaller cities will cost less. The savings in operating costs of a remote data center are passed on to the customer.

Premium support is available for an additional cost.

What other advice do I have?

The suitability of this solution depends on your organizational requirements and how you want to function, as well as the policies. One example is that a lot of financial institutions do not want to migrate to the cloud for security reasons. Some organizations prefer one provider over another, such as AWS. There is no hard and fast rule that says you have to use only the IBM Cloud. The option is up to the customer.

Overall, I would stay that IBM Cloud is stable, straightforward, and is a good service. I recommend it. My main complaint is about support, which is tied to pricing.

I would rate this solution an eight out of ten.

Which deployment model are you using for this solution?

Public Cloud

If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?

IBM
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Get our free report covering Red Hat, Microsoft, SAP, and other competitors of Amazon AWS. Updated: September 2021.
540,884 professionals have used our research since 2012.