Forward-looking Enterprises already operate their IT in a hybrid/multi-cloud state of affairs, there is a need to manage multiple environments – some on-premises and some in the public cloud, with each supporting different types of applications- from legacy, bare metal, VMs, containers, serverless, and more. This situation is only going to get exponentially more complex as the industry slowly evolves to a Software-Defined Infrastructure (SDI) model. So, boiling everything down to bare essentials – what are the most important fundamental truths that IT executives should keep in mind as they build their IT infrastructure to disarm competition?
Photo by Robert McLay on Unsplash
While that is the vision, the average enterprise data center presents a decidedly mixed picture. The vast portion of industry lags in understanding the business possibilities enabled by an agile cloud. No-fault of their own as your average IT infrastructure person is burdened with Day 1, Day 2 and Day 2+ operations. Secondly, DevOps practices are fairly prevalent across enlightened development teams who are gradually evolving from monolithic architectures to microservices & serverless paradigms.
The Five Axioms of Software-Defined infrastructure
While individual approaches to the SDI or highly fluid infrastructure will vary, there are five essential truths to multi-cloud success, to allow organizations to bring their diverse infrastructure under unified management:
Axiom# 1 Keep Cloud Management Simple (or) Don’t turn your CMP into an ERP
You want to simplify and accelerating time to value at complex hybrid cloud management and operations tasks. You need a single pane of glass that works across any cloud provider and abstracts their custom IaaS APIs with a standards-based open-source API. CMPs should be easy to setup & shouldn’t be turned an ERP-type project, but rather just work out of the box and enable you to turn any infrastructure into a cloud instantly. Beware of vendors bundling months of professional services or custom integration work to implement the CMP. Those implementations will fail.
Axiom #2 Question anything proprietary, build on Open Source & Beware of vendors bearing ecosystems
to ensure standardization and portability between environments/clouds, avoid lock-in, and have an open API layer that is consistent, for dev and Ops, regardless of where the app is running. This also ensures that you future-proof your cloud management solution for whatever new technology comes next, be easily extensible and is flexible to support new integrations, services, and specific use cases, as well as benefit from the open-source economics and savings vs. high licensing fees of proprietary solutions.
Axiom#3 Simplify Day 1, Day 2 & Day 2+ Operations
The most difficult thing about running a true hybrid cloud is the setup, installation, configuration of the underlying clouds, Day 2 operations – upgrades, application installs, monitoring & Day 2+ operations – configuration management and orchestration. Public clouds have already set the bar for ease of use. Extending that multi-cloud solutions should “just work” out of the box in a similar fashion. What that implies is a smooth developer experience, easy setup, easy integrations, and automated operations. The brains of the operation – the control plane or management layer should be able to perform Day1, Day2 and Day2+ Operations using a SaaS-based delivery model. No more unnecessary manual work, heavy lifting on the Operations side, or taxing management overhead. Multi-cloud management should be delivered as SaaS, so it is up and running in a day or two. Virtually every large hybrid cloud project failed because vendors were not able to provide a smooth operational experience
Axiom #4 Provide your Developers with a Catalog, Single Click Provisioning that can deploy Reference Architectures
Developers need to be provided with four things –
- A catalog of precertified runtimes that are ready to be deployed onto VMs or Containers. Examples include Tomcat, Kafka, NGINIX, Spark et al
- A way to describe their applications that are composed of the above runtimes as a Reference Architecture
- A declarative manner to quickly deploy the architecture into the underlying cloud
- Finally, support for an unopinionated yet end to end Continuous Integration, Deployment and Delivery pipeline
Axiom #5 Provide your Developers, Cloud Administrators & SREs a Unified API to manage, configure and orchestrate both infrastructure or applications
Cloud Administrators needs a unified experience across four areas – no matter what the underlying cloud or hypervisor:
A single view of all cloud regions, all types of infrastructure: servers, VMs, Containers, storage, and network – across these clouds, the tenants across these regions.
- b) single way for Site Reliability Engineers (SREs) to administer hybrid infrastructure across critical areas such as security & identity management
- c) Unified & open API for operations to perform lifecycle management and easy integrations with point tools or management processed)
- d) Continuous monitoring across all of the different cloud regions and environments
The Vision (or how it should all work in a nearly perfect world)
Cloud implementation really stems from a vision that enterprises need to harness powerful macro technology forces in support of their business goals. Powerful technology forces that are going to overwhelm their business if they do not invest in and harness these paradigms. We see that the day is not far off when the average datacenter becomes highly fluid. What does that mean? Infrastructure (compute, network, storage) and the applications that operate on them will easily traverse across a hybrid cloud based on business policies or chasing cost criteria (spot pricing, et al).
There is a way to get hybrid clouds and multi-cloud right. Enterprises need to be able to instantly turn any infrastructure into a cloud, and benefit from a unified cloud experience on ANY infrastructure, for ANY application – to be able to consistently manage VMs, Kubernetes, and Serverless – running on-premises or in the public cloud. This may represent nirvana to a range of CIOs who are dealing with huge data center challenges. Their data centers have every technology under the sun from both a compute, network and storage standpoint. On top of that, all these technologies offer islands or silos of management and provisioning – which have created huge technical debt over the decades.
Developers complain about a lack of agility and self-service. Let alone provision running applications from a click of a button, it takes a few weeks to get a running VM to provision applications on.
These are exactly the challenges Platform9 is focused on and a large part of the reason four VMware veterans left the company to start Platform9. Cloud flexibility is one of the key advantages of our platform. Using an Open API, enterprises can create a hybrid cloud in hours. A hybrid cloud as a logical entity that can incorporate any underlying cloud – VMware., KVM, Containers on Bare metal, OpenStack, AWS, Azure, GCP, etc. Our open platform can work with all of these clouds from both a developer experience and infrastructure scalability standpoint. Further, our managed service provides four nine’s SLA thus freeing customers from having to manage 3-4 different kinds of clouds.