Containers on AWS: a quick guide



Containerisation allows development teams to move quickly and deploy more efficiently


Instead of virtualising the hardware stack (as you would with virtual machines), containers run on top of the OS kernel, virtualising at the OS level.

Here are the most popular container formats available:




In 2010, a company known as Docker helped transform cloud containerisation. This new way of architecting paved the way for the DevOps movement. But what made containers so popular? Thanks to the huge improvements in virtualisation and the rapid increase of cloud computing, containers can allow for isolated workloads based on an OS, exposing and accessing only what is necessary.

Within just a few years, Amazon Elastic Container Service (ECS) was introduced in November 13, 2014 and was the primary way to run containers in the public cloud. ECS is a container management service that allows you to run Docker containers on a cluster.




Google released Kubernetes in June 2014, which was later released to the Cloud Native Computing Foundation (CNCF) community the following year. The Google Cloud Platform and Microsoft Azure were early adopters to Kubernetes, but with GCP being the only public cloud provider to have a working service called Google Kubernetes Engine (GKE). GKE was launched in 2015 and Azure Kubernetes Service (AKS) was released in the Fall of 2017 into preview mode.



Amazon EKS

Amazon Elastic Container Service for Kubernetes (EKS) is a fully managed service that makes it easy for you to use kubernetes on EKS runs upstream Kubernetes so you can connect to it with kubectl just like a self managed Kubernetes. AWS Introduced EKS at re:Invent 2017 and claims to upstream Kubernetes by using countless AWS growing services.



AWS Fargate

AWS has a hidden service that neither GCP or Azure have. AWS Fargate is a new service for running containers without needing to manage the underlying infrastructure. Fargate supports ECS and EKS but is also often closely compared with Lambda. You pay per computing second used without having to worry about the EC2 instances.

Managing Kubernetes can be complicated and usually requires a deep understanding of how to schedule, manage your masters, pods, services, and additional orchestration of architecture on top of the virtualisation that was already abstracted from you.

Fargate takes all of this away by streamlining deployments. The game-changer is that you do not need to start with Fargate, but that you can use EKS or ECS then migrate your workloads to Fargate when your program has matured further.





KOPS was the go to method of deploying Kubernetes on ECS via EC2 instances or on EC2 instances. KOPS is an open sourced project that makes running kubernetes easy. KOPS is built using EC2 instances. KOPS provides a multitude of controls on deployments and good support for high availability.


Containers are not just a hype, but they could be the future for at least the next few years. With AWS finally joining the Kubernetes club, and Fargate being a strong game-changer, anything is possible. However, there is is still a lot of unanswered questions that we hope will be addressed.

EKS and Fargate are currently limited in Ohio and Virginia regions, but you should see a big push to use these services as more regions get rolled out.


What do we do in the meantime? I’m reminded of this quote:


“All we have to decide is what to do with the time that is given us.”


Until then, I believe KOPS will be the best method to use.


What containers do you use on AWS and are you waiting to explore with AWS EKS or Fargate? Let us know by contacting us here.

Check also my previous blog post on Container security here


Get in Touch.

Let’s discuss how we can help with your cloud journey. Our experts are standing by to talk about your migration, modernisation, development and skills challenges.

    AWS Fargate – Bringing Serverless to Microservice



    Microservices architecture

    Microservices architecture has been a key focus for a lot of organisations in the past few years. Organisations around the world are changing from the traditional monolithic architecture – to a faster time-to-market, automated, and deployable microservices architecture. Microservices architecture approach has its number of benefits, but the two that come up the most are how the software is deployed and how it is managed throughout its lifecycle.

    Pokémon Go & Kubernetes

    Let’s look at a real world scenario, Pokémon Go. We wouldn’t have Pokémon Go, if it wasn’t for Niantic Labs and Google’s Kubernetes. For those of you who played this once addictive game back in the summer of 2016, you know all about the technical issues they had. It was the microservice approach of using Kubernetes that allowed Pokémon Go to fix technical issues in a matter of hours, rather than weeks. This was due to the fact that each microservice was able to be updated with a new patch, and thousands of containers to be created during peak times within seconds.

    When it comes to microservices and using the popular container engine like docker with a container orchestration software like Kubernetes (K8’s), with a microservice architecture everything in the website server is broken down into its own individual API’s. Giving microservices more agility, flexible scaling, and the freedom to pick what programming language or version is used for that one API instead of all of them.

    It is can be defined more ways than one, but it is commonly used to deploy well-defined API’s, and to help make delivery and deployment streamlined.


    Serverless the next big thing

    Some experts believe that serverless will be the next big thing. Serverless doesn’t mean there is no servers, but it does mean that the management and capacity planning are hidden from the DevOps teams. Maybe you have heard about FaaS (Functions as a Service) or AWS Lambda. FaaS is not for everyone, but what if we could bring some of the serverless architecture along with the microservice architecture.


    AWS Fargate

    This is why back in November at the AWS re:Invent 2017 (see the deep dive here), AWS announced a new service called AWS Fargate. AWS Fargate is a container service that allows you to provision containers without the need to worry about the underlying infrastructure (VM/Container/Nodes instances). AWS Fargate will control ECS (Elastic Container Service) and EKS (Elastic Kubernetes Service). Currently only available in the us-east-1 in Preview Mode.

    AWS Fargate, simplifies the complex management of microservices, by allowing developers to focus on the main task of creating API’s. You will still need to worry about the memory and CPU that is required for the API’s or application, but the beauty of AWS Fargate is that you never have to worry about provisioning servers or clusters. This is because AWS Fargate will autoscale for you. This is where microservices and Serverless meet.

    Get in Touch.

    Let’s discuss how we can help with your cloud journey. Our experts are standing by to talk about your migration, modernisation, development and skills challenges.

      Day 2 at Re:Invent – Builders & Musicians Come Together



      When Werner Vogels makes bold statements, expectations are set high. So when Vogel’s tweeted 15 minutes before the start of re:Invent’s day 2 keynote, we had to wonder what was coming.

      And how right we were. The close to 3 hours spent in the Venetian hotel in Las Vegas was an experience in itself.

      Andy Jassy opened the keynote with a long list of customers and partners, alongside the latest business figures. AWS are currently running at an 18 billion run rate with an incredible 42% YoY growth. With millions of active customers – defined as accounts that have used AWS in the last 30 days – the platform is by far the most used on the planet.

      As per Gartner’s 2016 Worldwide Market Segment Share analysis, the company (successfully led by Jassy), has achieved a 44.1% market share in 2016, up from 39% in 2015, more than everyone else combined. This became easily noticeable when AWS displayed an entire catalogue of new services throughout the keynote. The general stance Jassy took this year was that AWS are trying to serve their customers exactly what they asked for in terms of new products. The mission of AWS is nothing short of fixing the IT industry in favour of the end-users and customers.

      The first on stage was a live ‘house’ band, performing a segment of ‘Everything is Everything’ by Lauryn Hill, the chorus rhyming with ‘after winter must come spring’. Presumably, AWS was referring to the world of IT still being in a kind of eternal ‘winter’. The concept we also heard here was that AWS would not stop building their portfolio and that they want to offer all the tools their ‘builders’ and customers need.

      AWS used Jassy’s keynote for some big announcements (of course, set to music), with themes across the following areas:

      • Compute
      • Database
      • Data Analytics
      • Machine Learning and
      • IoT

      The Compute Revolution Goes On

      Starting in the compute services area, an overview of the vast number of compute instance types and families were shown, with special emphasis given to the Elastic GPU options. There were a few announcements also made on the Tuesday night, including Bare Metal InstancesStreamlined Access to Spot Capacity & Hibernationmaking it easier for you to get up to 90% of savings on normal pricing. There was also M5 instances which offer better-priced performance than their predecessors, and H1 instances offering fast and dense storage for Big Data applications.

      However, with the arrival of Kubernetes in the industry, it was the release of the Elastic Kubernetes that was the most eagerly anticipated. Not only have AWS recognised that their customers wanted Kubernetes on AWS, but they also realise that there’s a lot of manual labour involved in maintaining and managing the servers that run ECS & EKS.

      To solve this particular problem, AWS announced AWS Fargate, a fully managed service for both ECS & EKS meaning no more server management and therefore increasing the ROI in running containers on the platform. This is available for ECS now and will be available for EKS in early 2018.

      Having started with servers and containers, Jassy then moved on to the next logical evolution of infrastructure services: Serverless. With a 300% usage growth, it’s fair to say that if you’re not running something on Lambda yet, you will be soon. Jassy reiterated that AWS are building services that integrate with the rest of the AWS platform to ensure that builders don’t have to compromise. They want to make progress and get things done fast. Ultimately, this is what AWS compute will mean to the world: faster results. Look out for a dedicated EKS blog post coming soon!

      Database Freedom

      The next section of the keynote must have had some of AWS’s lawyers on the edge of their seats, and also the founder of a certain database vendor… AWS seem to have a clear goal to put an end to the historically painful ‘lock-in’ some customers experience, referring frequently to ‘database freedom’. There’s a lot of cool things happening with databases at the moment, and many of the great services and solutions shown at re:Invent are built using AWS database services. Out of all of these, Aurora is by far growing the fastest, and actually is the fastest growing service in the entire history of AWS.

      People love Aurora because it can scale out for millions of reads per second. It can also autoscale new read replicas and offers seamless recovery from reading replica failures. People want to be able to do this faster, which is why AWS launched a new Aurora features, Auto Multi-Master. This allows for zero application downtime due to any write node failure (previously, AWS suggested this took around 30 seconds), and zero downtime due to an availability zone failure. During 2018 AWS will also introduce the ability to have multi-region masters – this will allow customers to easily scale their applications across regions have a single, consistent data source.

      Lastly, and certainly not least, was the announcement of Aurora Serverless. which is an on-demand, auto-scaling, Serverless version of Aurora. The users pay by the second – an unbelievably powerful feature for many use cases.

      Finally, Jassy turned its focus point to DynamoDB service, which scaled to ~12.9 million requests per second at its peak during the last Amazon Prime Day. Just let that sink in for a moment! The DynamoDB service is used by a huge number of major global companies, powering mission-critical workloads of all kinds. The reason for this is, from our perspective, is the fact that it’s very easy to access and use as a service. What was announced today was the new feature DynamoDB Global Tables. This enables users to build high performance, globally distributed applications.

      The final database feature released for DynamoDB was managed back-up & restore, allowing for on-demand backups, point-in-time recovery (in the past 35 days), allowing backups for data archival or regulatory requirements to be taken of hundreds of TB with no interruption.

      Jassy wrapped up the database section of his keynote by announcing Amazon Neptune, a fully managed graph database which will make it easy to build and run applications that work with highly connected data sets.


      Next Jassy turned to Analytics, commenting that people want to be using S3 as their data lake. Athena allows for easy querying of structured data within S3, however, most analytics jobs involve processing only a subset of the data stored within S3 objects and Athena requires the whole object to the processed. To ease the pain, AWS released S3 Select – allowing for applications, (including Athena) to retrieve a subset of data from an S3 object using simple SQL expressions – AWS claim drastic performance increases – possibly up to 400% performance.

      Many of our customers are required by regulation to store logs for up to 7 years and as such ship them to Glacier to reduce the cost of storage. This becomes problematic if you need to query this data though. How great would it be if this could become part of your data lake? Jassy asked, before announcing Glacier Select. Glacier Select allows for queries to be run directly on data stored in Glacier, extending your data lake into Glacier while reducing your storage costs.

      Machine Learning

      The house band introduced Machine Learning with ‘Let it Rain’ from Eric Clapton. Dr Matt Woods made an appearance and highlighted how important machine learning is to Amazon itself. The company uses a lot of it, from personal recommendations on to Fulfillment automation & inventory in its warehouses.

      Jassy highlighted that AWS only invests in building technology that its customers need, (and, remember is a customer!) not because it is cool, or it is funky. Jassy described three tiers of Machine Learning: Frameworks and Interfaces, Platform Services & Application Services.

      At the Frameworks and Interfaces tier emphasis was placed on the broad range of frameworks that could be used on AWS, recognising that one shoe does not fit every foot and the best results come when using the correct tool for the job. Moving to the Platform Services tier, Jassy highlighted that most companies do not have to expect machine learning practitioners (yet) – it is after all a complex beast. To make this easy for developers, Amazon SageMaker was announced – a fully-managed service that enables data scientists and developers to quickly and easily build, train, and deploy machine learning models at any scale.

      Also at the platform tier, AWS launched DeepLens, a deep learning enabled wireless video camera designed to help developers grow their machine learning skills. This integrates directly with SageMaker giving developers an end-to-end solution to learn, develop and test machine learning applications. DeepLens will ship in early 2018, available on for $249.

      The machine learning announcements did not stop there! As Jassy moved into the Application Services tier AWS launched:


      Finally, Jassy turned to IoT – identifying five ‘frontiers’ each with its own release, either available now, or in early 2018:

      1. Getting into the game – IoT One Click (in Preview) will make it easy for simple devices to trigger AWS Lambda functions that execute a specific action.
      2. Device Management – AWS IoT Device Management will provide fleet management of connected devices, including the onboarding, organisation, monitor and remote management through a devices lifetime.
      3. IoT Security – AWS IoT Device Defender (early 2018) will provide security management to your fleet of IoT devices, including auditing to ensure your fleet meets best practice.
      4. IoT Analytics – AWS IoT Analytics, making it easy to cleanse, process, enrich, store, and analyze IoT data at scale.
      5. Smaller Devices – Amazon FreeRTOS, an operating system for microcontrollers.

      Over the next weeks and days, the Nordcloud team will be diving deeper into these new announcements, (including our first thoughts after getting our hands on the new releases) We’ll also publish our thoughts and how they can benefit you.

      It should be noted that, compared to previous years, AWS are announcing more outside the keynotes, in sessions and on their Twitch Channel and so there are many new releases which are not gaining the attention they might deserve. Examples include T2 UnlimitedInter-Region VPC Peering and Launch Templates for EC2 – as always the best place to keep up-to-date is the AWS ‘whats new‘ page.

      If you would like to discuss how any of today’s announcements could benefit your business, please get in touch.

      Get in Touch.

      Let’s discuss how we can help with your cloud journey. Our experts are standing by to talk about your migration, modernisation, development and skills challenges.