Where to host Docker Containers on Azure (AKS, ASE, or ASF)?

Azure Kubernetes Service (AKS) service Azure App Service Environment (ASE) Azure Service Fabric (ASF) Comparison

Scenario:

So, your team recently has been tasked with developing a new application and running it. The team made the decision to take a microservices based approach to the application. Your team also has decided to utilize Docker containers and Azure as a cloud platform. Great, now it’s time to move forward right? Not so fast. There is no question that Docker containers will be used, but what is in question is where you will run the containers. In Azure containers can run on Azure’s managed Kubernetes (AKS) service, an App Service Plan on Azure App Service Environment (ASE), or Azure Service Fabric (ASF). Let’s look at each one of these Azure services including an overview, pro’s, cons, and pricing.

This Azure Kubernetes Service (AKS) Pros and Cons chart is clickable.
This Azure App Service Environment (ASE) Pros and Cons chart is clickable.
This Azure Service Fabric (ASF) Pros and Cons chart is clickable.

Conclusion:

Choose Azure Kubernetes Service if you need more control, want to avoid vendor lock-in (can run on Azure, AWS, GCP, on-prem), need features of a full orchestration system, flexibility of auto scale configurations, need deeper monitoring, flexibility with networking, public IP’s, DNS, SSL, need a rich ecosystem of addons, will have many multi-container deployments, and plan to run a large number of containers. Also, this is a low cost.

Choose Azure App Service Environment if don’t need as much control, want a dedicated SLA, don’t need deep monitoring or control of the underlying server infrastructure, want to leverage features such as deployment slots, green/blue deployments, will have simple and a low number of multi-container deployments via Docker compose, and plan to run a smaller number of containers. Regarding cost, running a containerized application in an App Service Plan in ASE tends to be more expensive compared to running in AKS or Service Fabric. The higher cost of running containers on ASE is because with an App Service Plan on ASE, you are paying costs for a combination of resources and the managed service. With AKS and ASF you are only paying for the resources used.

Choose Service Fabric if you want a full micros services platform, need flexibility now or in the future to run in cloud and or on-premises, will run native code in addition to containers, want automatic load balancing, low cost.

A huge thanks to my colleague Sunny Singh (@sunnys101) for giving his input and reviewing this post. Thanks for reading and check back for more Azure and container contents soon.

Read more

Azure Management Groups

If your company is like most organizations that are using the cloud, then you have many subscriptions floating around. This is often due to “shadow IT”. However, sometimes organizations simply use many subscriptions as a way to put boundaries around cloud services for departments, teams or other reasons.

Microsoft has built a new service in Azure to help with the governance of your cloud. This new service is called Management Groups. Management Groups is still in preview but it is something I highly recommend you start trying out or using now as it is going to be as big for cloud as group policy was for on-premises AD based environments.

Management Groups sit above subscriptions. This allows Management Groups to be at the highest level in the chain so they can be used to effectively manage access, policies, and compliance for any subscriptions that belong to your organization. Within Management Groups you can set access controls (RBAC) and Azure policy to be applied to subscriptions. Subscriptions are organized in logical containers and the containers are the “management groups”. Your governance conditions are then applied to the management groups. This is the much-needed enterprise level type of management that has been needed in Azure for a while.

Management Groups will eventually become the starting point of governance when organizations embark on the cloud. Management Groups also can be used for organizations that are already in the cloud. I am going to dive into Management Groups giving you a high-level tour but first I need to give some more background on the components of Management Groups.

Each directory has a “root management group”. This root management group is at the top level of the management group hierarchy. All other management groups and subscriptions fold up to the root management group. Access and policies can be applied at the directory level via this root management group.

A couple of other things to note about management groups are that you can only have up to 10,000 management groups in a single directory, a management group tree can go six levels deep not including the root management group, and each management group can have multiple children management groups but only one parent management group.

Now let’s explore how I have structured my management groups to give some examples of how this works. Note that all the examples I show in this blog post are for my Azure environments but yours will be different based on many factors such as your organizational structure of departments, teams, etc.

You can find management groups under All Services>>Management Groups.

When you first access Management Groups you will need to create a root MG. Note that the root MG cant deleted or moved. You can rename the root MG. In the following screenshot, I am showing the creation of a sub MG in my root MG. Also, notice on the left-hand side you can set Access controls (RBAC) on this MG.

In order to configure Azure Policies and apply it to a management group, you do that within the Azure Policy itself. You can see in the following screenshot that I have an Azure policy and I am scoping it to the Prod01 MG. Whatever subscription/s and resources in those subscriptions will inherit the policy unless an exclusion is set in the policy or I am breaking inheritance at the resource group level.

In the following screenshot, I am showing the addition of an existing resource. The resources you can add are other MG’s or subscriptions.

In the following screenshot, you can see that I am going to add one of my subscriptions to my Dev01 management group. After doing this I can configure development related access and development related policies to this subscription. I also can do the same thing with my production environments/subscriptions.

Here is what my Management Groups hierarchy looks like:

In my hierarchy I have 3 subscriptions I split into two for production and 1 for development. I have created a root management group and placed all other management groups in it. I created a parent management group for my prod subscriptions and 1 for my development subscriptions in case I add more in the future. I then created a prod01 and prod02 pulling a subscription into each one. Doing this allows me to have separate access and policies per subscription. One thing you could do is pull multiple subscriptions into a single management group.

Note that I also could apply access and policies at the root level or at one of my environment management groups i.e. Prod_Env/Dev_Env and the sub-management groups would inherit the access and policies that are set at the environment management group level.

Also if you need to you can move management groups to a new parent management groups.

Thanks for reading this post. As I mentioned at the beginning of this post Azure Management Groups are currently in preview but they are worth checking out and potentially using now as these are going to become a critical part of the Azure governance story.

Read more

Sys Admin to Cloud Admin…ITSM to CloudOps…On-Prem to Azure Stack/Azure

A while back I posted a blog titled “Surviving the future of IT as an IT pro”. In that blog post I set out to share my opinion on where IT is headed and what you should focus on as an IT pro going forward. I guess this post could be considered part 2 however in this post I will focus more on where things are heading as a whole.

So what is this blog really about? It is about “the Transition from ITOPS & ITSM to CloudOps via Azure Stack (Hybrid Cloud) powering DevOps and becoming core to the Digital Transformation of business” that is happening. Whew…..Ok, a lot was said in that previous sentence. J Let’s break it down.

Transition from ITOPS & ITSM to CloudOps

There has been this transition in IT for a while to increase the density in data centers. This was started with the wide adoption of the hypervisor (VMWare, Hyper-V, Citrix Xen etc…). The goal is to get more out of existing and less physical hardware. Think about 1 physical server hosting hundreds of virtual servers. Things have since accelerated at a fast pace. We now have containers, PaaS, and serverless. With these newer technologies, the density is even greater.

The real power behind cloud is software defined everything. With software, defined environments AKA cloud a new skillet and a different way of thinking about managing operations is needed. This new skillset and new way of thinking for the operationalization of cloud is known as CloudOps. IT Operations and IT Service Management do not go away with CloudOps. The evolution of ITOPS and ITSM become CloudOps. The best parts of ITOPS and ITSM (ITIL) funnel into CloudOps used for operating clouds.

Hybrid Cloud (Azure Stack)

Hybrid Cloud is going to be a huge part of cloud initiatives in many organizations for years to come. You can see this on the Gartner reports(http://www.gartner.com/newsroom/id/3354117), Right Scale reports (http://www.rightscale.com/blog/cloud-industry-insights/cloud-computing-trends-2017-state-cloud-survey) and based on the investments the major cloud players are making to build the best Hybrid Cloud solutions.

Hybrid Cloud Is the Preferred Enterprise Strategy, but Private Cloud Adoption Fell

From Rightscale “Cloud Computing Trends: 2017 State of the Cloud Survey” Report:

http://www.rightscale.com/blog/cloud-industry-insights/cloud-computing-trends-2017-state-cloud-survey#hybrid-cloud

Recently IBM and Red Hat announced their launch into the Hybrid Cloud space.

(http://www.networkworld.com/article/3182989/cloud-computing/ibm-red-hat-an-open-source-hybrid-cloud.html)

A while back Amazon and VMWare announced their launch into the Hybrid Cloud space.

(http://www.businesswire.com/news/home/20161013006574/en/VMware-AWS-Announce-Hybrid-Cloud-Service-%E2%80%9CVMware>)

Microsoft was the first to jump into the Hybrid Cloud space and is the only company that has a 100% true Hybrid Cloud solution. Both VMWare/Amazon and IBM/Red Hat have solutions that run private cloud on public cloud. The private cloud solutions are being retrofitted to run in public cloud as the framework for their Hybrid Cloud solutions. These are not consistent cloud platforms running the same exact bits on bare metal on-premises and in the cloud like Microsoft’s Azure Stack solution. Azure Stack is the same bits in the public cloud and on-premises down to the bare metal.

IBM and Amazon jumping into the Hybrid Cloud space is more proof this will be a large area of growth in IT for years to come. I wonder if Google will decide to jump into the Hybrid Cloud space at some point and what their strategy will be.

DevOps powered by Azure Stack and CloudOps

Azure Stack serves as a catalyst to help move DevOps initiatives forward within organizations. With Azure Stack’s comes the native ability to run the environment using Infrastructure as code, continuous integration, continuous delivery, microservices, integration with source control systems, and more. All of the aforementioned are a part of DevOps.

Along with Azure Stack is the need to run the environment using a CloudOps model. Here is a list of concepts that drive CloudOps:

  • Extreme Hardware Standardization
  • Software Defined Everything
  • Extreme Automation
  • Focus on Zero Downtime
  • Self Service
  • Measured Service
  • Multitenancy

CloudOps is overall focused on business applications critical for running the business through the continuous operations of clouds. CloudOps leaves business unit projects to DevOps. CloudOps instead focuses on the delivery of the the cloud infrastructure to support self-service leveraged by DevOps teams.

David Armour of Microsoft often shares great information on CloudOps and what it means. You can follow him on twitter here: https://twitter.com/Darmour_MSFT

CloudOps supports DevOps and DevOps is core to Digital Transformation

Digital Transformation is the accelerating transformation of the way businesses do business from traditional ways often brick and mortar to the digital front through the use of digital technologies. Businesses are shifting to meet their customers and employees where they are today on digital platforms. In the business world, today it is well known that you must innovate and grow through the use of technology or become obsolete and left in the wake of disruptive companies that are leveraging technology to meet their customers on the digital front.

Examples of digitally transformed company’s vs non-digitally transformed companies are:

  • Netflix vs Blockbuster
  • Amazon vs Target, Best Buy, Macy’s
  • Airbnb vs Wyndom hotels
  • Uber vs Taxi Companies

Digital Transformation is critical to business and IT departments need to be a core driver to help organizations move forward on the digital transformation front. Digital Transformation is the new Industrial Revolution of business today with CloudOps/DevOps being the Assembly line that will bring innovation to the business.

Through DevOps businesses can bring digital services to the market at very fast rates and can pivot quickly as needed to beat and stay ahead of the competition meeting the customers’ demands in an agile way. CloudOps allows the scale and another point to pivot on at any time to redirect in a new direction as needed by the business in an agile manor.

Through a Hybrid Cloud solution like Azure Stack things IoT, Microservices, extreme automation, hyper-scale, and agility can be realized for the business empowering Digital Transformation from the core.

The transition of the IT Pro to Cloud Pro

Ok. That was a lot of information and background on CloudOps, DevOps, Digital Transformation and Hybrid Cloud. You may be asking yourself at this point where does the IT Pro fit into the picture? Let me answer that for you and take you on a tour of Azure Stack to prove why as an IT Pro you should start working with it today!

The path for an IT Professional when moving from traditional IT into a Hybrid Cloud world consists of:

  A cloud administrator can configure and manage resource providers, tenant offers, plans, services, quotas, and pricing.
A tenant purchases (or acquires) services that the service administrator offers. Tenants can provision, monitor, and manage services that they have subscribed to, such as Web Apps, Storage, and Virtual Machines.

Those cloud roles fit in a new world of CloudOps including Cloud architect, engineer, and administrator. Being a part of CloudOps requires a different mindset. Think about dynamic shifts such as software defined everything and extreme standardization. More concepts and technologies that a cloud role requires an understanding of are:

  • IaaS
  • PaaS
  • Software Defined Data Center technologies
  • Automation
  • Source Control Systems
  • Business Intelligence (Showback/Chargeback)
  • High Availability technologies
  • Backup and Disaster Recovery
  • Scaling technologies
  • Containerization
  • Server less technologies
  • Cloud Security
  • Both Linux and Windows
  • Self-Service (Service Catalog)
  • Multitenancy technologies
  • Tenant administration
  • And more

Ok. Now let’s jump into some example of CloudOps tooling in Azure Stack. First off, we as a cloud admin you need to know how to perform management of tenants (customers). Here is an example of a dashboard for doing this in Azure Stack:

In Azure Stack, you will need to know and understand the administration of managing the cloud itself. This includes many things some of them being management of a region/s, resource providers that contain the services you can offer up to tenants, along with monitoring, high availability, and backup of the cloud. Below is an example of administration in Azure Stack at the cloud model layer of CloudOps.

We already mentioned monitoring. There is monitoring of the cloud environment itself but there also is a need to monitor the resources being consumed by the tenants. One of the great things about Azure and Azure Stack is the out of the box monitoring and health diagnostics of IaaS virtual machines. I am a SCOM guy and have done a lot of SCOM projects. SCOM works well and serves a purpose but the out of the box monitoring in Azure and Azure Stack is amazing in the ease of turning it on. Once turned on it just works and has very nice visuals to see and work with as shown in the following screenshot. As a cloud administrator, you need technology to be easy so that you can move away from complex setups and troubleshooting the monitoring solution and move to monitoring the resources.

One of the best benefits about Hybrid Cloud is the consistency between public and on-premises cloud. In the following screenshot news updates on Azure and Azure stack are the same. 🙂 Another huge point of consistency between Azure and Azure Stack is the ability to view, deploy and run items from the Azure marketplace in Azure Stack. This is called marketplace syndication.

 

Azure

 

Azure Stack

Azure Stack is set to release in 2017. I want to highlight some of the services already in Azure Stack and more coming to Azure Stack that can be offered in your Service Catalog to tenants.

Already in Azure Stack as of TP3:

  • SQL PaaS
  • MySQL PaaS
  • Web Apps PaaS
  • Computer IaaS
  • Virtual Machines (Linux or Windows)
  • VM Scale Sets
  • Storage
  • Networking
  • PaaS: Storage
  • Key Vault
  • Management of Azure Pack virtual machines
  • Marketplace Syndication

Coming to Azure Stack at some point:

  • Microservices
  • Service Fabric
  • Cloud Foundry
  • Blockchain
  • Container Service
  • IoT

Another big part of CloudOps is being able to measured services that are being consumed. Measured Service can translate to show back or charge back. Measured Service is the ability to track the usage of resources down to the individual resource level. With Azure and Azure Stack resource management (ARM) model resources are carved out and placed into resource groups. In ARM, each resource has an associated cost that is tracked via the usage. There is full role based access around resources and resource groups. Resources and resource groups can be tagged and each resource or resource group’s usage can be tracked and displayed on business intelligence reporting or a dashboard like shown in the following screenshot.

That concludes this blog post. I hope I was able to shed some light on the transition from IT Pro to cloud pro, from IT Ops/ITSM to CloudOps and showcase the power of Hybrid Cloud via Azure Stack. Stay tuned for more exciting stuff coming from Azure Stack.

Read more