Grounded in the Cloud
cancel
Showing results for 
Search instead for 
Did you mean: 

Why Containers are an essential component of the Cloud journey

Stephen_Spector

Author: Omri Gazitt, VP Engineering, HP Helion

 

Today, a coalition of industry partners including HP has announced the Open Container Project under the Linux Foundation, whose goal is to develop open standards for containers, and specifically an open, interoperable container image format.  Containers are an exciting new technology and the goal of this post is to orient container technology within the broader context of the Journey to the Cloud.

 

Cloud Platforms transform a set of physical resources – compute, memory, storage, and networking – into elastic utilities where heterogeneous workloads can share these resources dynamically.  The key technological enablers for the current generation of Cloud Platforms are virtualization technologies such as hypervisors, software-defined storage, and software-defined networking.  Hypervisors allow running multiple operating system images on the same physical machine, with the underlying hardware enforcing security and resource isolation.  The “unit of work” is a virtual machine image – the same as for traditional workloads – which helps provide an effective bridge between traditional workloads and the Cloud.  We call this use case “Cloud Core”.

 

While Cloud Core workloads are an essential part of the evolution to the Cloud, Enterprises won’t realize the full power of the Cloud until they start building Cloud-Native workloads.  The Cloud-Native architecture is designed for failure at the level of the physical resources by transitioning from the “Pets” architecture, where each virtual machine is mission critical and must be kept operational, to “Cattle” – a collection of stateless micro-services that can be scaled independently to provide fault-tolerance and automatically adapt to application load.  With the “Cattle” architecture, if one instance of a micro-service fails, another can automatically take its place, and more instances can be quickly provisioned to take up application load.

 

When applications are designed and built this way, a full virtual machine is no longer an efficient “unit of work”.  Because of this, many of the industry’s Cloud Platforms have employed container technologies as a way to provide lighter-weight units of work that use operating system primitives, such as Linux namespaces and cgroups, to provide security and resource isolation.  For Cloud-Native workloads, these lighter-weight isolation techniques can result in an order of magnitude improvement in application density per physical machine, and often translate to an order of magnitude improvement in container startup times versus virtual machines.  These factors make containers a natural fit for Cloud-Native applications: indeed, containers are the primary enabling technology for Cloud-Native workloads in the same way that hypervisor-based virtualization is the primary enabling technology for Cloud Core workloads.

 

While container technologies aren’t new (for example, most Platforms-as-a-Service, such as Cloud Foundry, have employed container technologies for years), the industry has seen explosive growth in their usage over the last couple of years as Docker created a de-facto standard image format and API for defining and interacting with containers.  The Helion Development Platform has built upon and taken advantage of Docker’s pioneering work by using it as a format and execution engine for our Cloud Foundry Application Lifecycle Service.

 

Docker has done a great job at making containers mainstream by showing leadership and innovation for the container ecosystem.  We are still early in the evolution of the container ecosystem with many choices emerging at every level of the stack – container-optimized operating systems, container engines, container clustering and scheduling systems, container orchestration systems, and container networking technologies, just to name a few.  Innovation is important at this stage of the container ecosystem, and much like with hypervisors, we believe that providing choice among the multiple implementations will best serve our customers.  At the same time, if each container “stack” is built on mutually incompatible formats, we will see confusion that will slow down the adoption of container technologies.  This is why we strongly support the Open Container Project as the vehicle for developing open standards for the emerging container ecosystem, and will actively participate in the creation of these standards while we continue bringing the best of the container technology ecosystem to the HP Helion Platform.

Senior Manager, Cloud Online Marketing
0 Kudos
About the Author

Stephen_Spector

I manage the HPE Helion social media and website teams promoting the enterprise cloud solutions at HPE for hybrid, public, and private clouds. I was previously at Dell promoting their Cloud solutions and was the open source community manager for OpenStack and Xen.org at Rackspace and Citrix Systems. While at Citrix Systems, I founded the Citrix Developer Network, developed global alliance and licensing programs, and even once added audio to the DOS ICA client with assembler. Follow me at @SpectorID

Events
28-30 November
Madrid, Spain
Discover 2017 Madrid
Join us for Hewlett Packard Enterprise Discover 2017 Madrid, taking place 28-30 November at the Feria de Madrid Convention Center
Read more
HPE at Worldwide IT Conferences and Events -  2017
Learn about IT conferences and events  where Hewlett Packard Enterprise has a presence
Read more
View all