Skip to main content

Kubernetes and cloud providers

In the microservices era, dominated by cloud providers the standard requirement that I receive on every strategic project is
The solution needs to independent from the hosting provider and needs to support on-premises solutions.
The standard context for this requirement is to support cloud providers - Microsoft Azure and AWS. Additional to this, with minimum effort (less than 5% of development cost), the solution shall be run on-premises.

The requirements are tricky even if at first glance you would say that Kubernetes is the solution. Two concerns need to be tackled. The first one is related to what kind of platform shall be used to host the Kubernetes solution. The second one related to how the solution shall use many native cloud services.

Kubernetes it is a powerful container-orchestration solution that allows us to scale our microservice solution. The downside of Kubernetes is that you need to install and configure many services besides Kubernetes itself if you want to have a working solution. From reverse proxy to package deployment solution, you will need to manage it by yourself.
AWS (EKS) and Microsoft Azure (Azure Kubernetes Service) are offering robust environments where you can run your Managed Kubernetes Environment. With low cost, you can spin-up your cluster and run solution inside such an environment. Even if it is super easy to spin-up your cluster, you need to know what you are doing there. Many times, you need to do the same things over and over again.
A solution that is starting to become more and more interesting for companies is OpenShift solution from Red Hat. The solution is constructed on top of Docker and managed using Kubernetes. It is not free but is providing a better experience by having more features included out of the box. Beside image management and deployment configuration, the so-called ‘Routes’ (reverse proxy) are integrated inside OpenShift much better than Ingress Resources and Ingress Controllers that need to be managed by yourself. OpenShift is supported on both AWS and Microsoft Azure ecosystem, and I notified that big organization prefers to pay for OpenShift costs to get them out of the box configuration that is already offered by default.
Things can become more interesting if you put on-premises on top of this. Managing the Kubernetes cluster by yourself, it is not an easy task and having the support of OpenShift can make the team life much more comfortable.

The second problem is the most sensible one because finding the right balance between native cloud services and build-by-yourself is hard. On one side you can optimize development and configuration cost by using native cloud services. This will force you in time to take technical decisions based on specific features available and sometimes forget about topics likes security or redundancy because you already have them by default.
When you build-by-yourself the solution, the dependency to a cloud provider is minimal, but the system looks more like on-premises system. The development and configuration effort are high, and you might ask yourself why I’m using a cloud provider.
A common approach for this situation is to build an API that isolated the cloud infrastructure from your application. In this way, you can control the area of your solution that has a direct dependency on the cloud provider. At least you know the code areas that need to be changed.
Another approach is to start to build components oriented to your functionality and design. The integration with cloud services shall be in only one location, but besides this, the API that is exposed by cloud integration components should take into account the features available cross-cloud provides. In this way, you can ensure that you have a solution that is compatible with other cloud providers too. For example, using Active Directory (AD) to control access to file storage can be a good solution when you are using Microsoft Azure, but you might have a problem if you would look for a similar solution on AWS.

Remember that no solution fits all and all the time is about tradeoffs. When you decide the approach that you want to use, take into account timeline and when you will need to integrate/use the second cloud provided.

Comments

Popular posts from this blog

Windows Docker Containers can make WIN32 API calls, use COM and ASP.NET WebForms

After the last post , I received two interesting questions related to Docker and Windows. People were interested if we do Win32 API calls from a Docker container and if there is support for COM. WIN32 Support To test calls to WIN32 API, let’s try to populate SYSTEM_INFO class. [StructLayout(LayoutKind.Sequential)] public struct SYSTEM_INFO { public uint dwOemId; public uint dwPageSize; public uint lpMinimumApplicationAddress; public uint lpMaximumApplicationAddress; public uint dwActiveProcessorMask; public uint dwNumberOfProcessors; public uint dwProcessorType; public uint dwAllocationGranularity; public uint dwProcessorLevel; public uint dwProcessorRevision; } ... [DllImport("kernel32")] static extern void GetSystemInfo(ref SYSTEM_INFO pSI); ... SYSTEM_INFO pSI = new SYSTEM_INFO(

Azure AD and AWS Cognito side-by-side

In the last few weeks, I was involved in multiple opportunities on Microsoft Azure and Amazon, where we had to analyse AWS Cognito, Azure AD and other solutions that are available on the market. I decided to consolidate in one post all features and differences that I identified for both of them that we should need to take into account. Take into account that Azure AD is an identity and access management services well integrated with Microsoft stack. In comparison, AWS Cognito is just a user sign-up, sign-in and access control and nothing more. The focus is not on the main features, is more on small things that can make a difference when you want to decide where we want to store and manage our users.  This information might be useful in the future when we need to decide where we want to keep and manage our users.  Feature Azure AD (B2C, B2C) AWS Cognito Access token lifetime Default 1h – the value is configurable 1h – cannot be modified

What to do when you hit the throughput limits of Azure Storage (Blobs)

In this post we will talk about how we can detect when we hit a throughput limit of Azure Storage and what we can do in that moment. Context If we take a look on Scalability Targets of Azure Storage ( https://azure.microsoft.com/en-us/documentation/articles/storage-scalability-targets/ ) we will observe that the limits are prety high. But, based on our business logic we can end up at this limits. If you create a system that is hitted by a high number of device, you can hit easily the total number of requests rate that can be done on a Storage Account. This limits on Azure is 20.000 IOPS (entities or messages per second) where (and this is very important) the size of the request is 1KB. Normally, if you make a load tests where 20.000 clients will hit different blobs storages from the same Azure Storage Account, this limits can be reached. How we can detect this problem? From client, we can detect that this limits was reached based on the HTTP error code that is returned by HTTP