Hybrid Cloud for Architects

Alok Shrivastwa

• Éditeur: Packt Publishing
• Année: 2018
• ISBN:
  9781788627986
• Édition: 1
• Sujet: Ciencia de la computación
• Sous-sujet: Computación en la nube

Build your own hybrid cloud strategy with this comprehensive learning guide.About This Book• Build a hybrid cloud strategy for your organization with AWS and OpenStack • Leverage Hybrid Cloud to design a complex deployment pipeline• Learn to implement security and monitoring best practices with real-world examplesWho This Book Is ForThis book is targeted at cloud architects, cloud solution providers, DevOps engineers, or any working stakeholder who wants to learn about the hybrid cloud architecture. A basic understanding of public and private cloud is desirable.What You Will Learn• Learn the demographics and definitions of Hybrid Cloud• Understand the different architecture and design of Hybrid Cloud• Explore multi-cloud strategy and use it with your hybrid cloud• Implement a Hybrid Cloud using CMP / Common API's • Implement a Hybrid Cloud using Containers• Overcome various challenges and issues while working with your Hybrid Cloud • Understand how to monitor your Hybrid Cloud• Discover the security implications in the Hybrid CloudIn DetailHybrid cloud is currently the buzz word in the cloud world. Organizations are planning to adopt hybrid cloud strategy due to its advantages such as untested workloads, cloud-bursting, cloud service brokering and so on. This book will help you understand the dynamics, design principles, and deployment strategies of a Hybrid Cloud. You will start by understanding the concepts of hybrid cloud and the problems it solves as compared to a stand-alone public and private cloud. You will be delving into the different architecture and design of hybrid cloud. The book will then cover advanced concepts such as building a deployment pipeline, containerization strategy, and data storage mechanism. Next up, you will be able to deploy an external CMP to run a Hybrid cloud and integrate it with your OpenStack and AWS environments. You will also understand the strategy for designing a Hybrid Cloud using containerization and work with pre-built solutions like vCloud Air, VMware for AWS, and Azure Stack. Finally, the book will cover security and monitoring related best practices that will help you secure your cloud infrastructure. By the end of the book, you will be in a position to build a hybrid cloud strategy for your organization.Style and approachThis book follows a step-by-step, practical approach to deploying and implementing hybrid cloud architectures using AWS and OpenStack.

Hybrid Cloud Deployment – Architecture and Preparation

It is finally the time to put all the theory we have learnt into practice, and create an environment that will be used to show the implementation of the different use cases possible in the hybrid cloud.

For this, we will need a private cloud environment and a public cloud environment connected to each other. The private cloud environment will be run from our data center, so, we can also assume that the traditional workloads are also there.

In order to set this up, we will use OpenStack as the private cloud and we will create an AWS VPC for a public cloud infrastructure. Since, we are using this as a lab or demo environment, we will simply use the IPSec VPN to connect between the two as no particular jitter-sensitive applications are going to be running in the environment.

In this chapter, we will also look at the use cases that allow the hybrid cloud to be more pragmatic in nature.

A hybrid cloud is said to be pragmatic when one end is the public cloud and the other end is holistic, by including the physical and virtual environments in the private data center in addition to the private cloud.

Getting started with the public cloud – AWS

So, regardless of the fact that we are integrating it with only a private cloud, or being realistic and integrating it just with a virtualized data center, we still need to set up the public cloud. In this case, we have chosen to set up AWS as our public cloud, however, the concepts will be similar for other cloud providers.

AWS is nothing short of a technological marvel. The scale they operate at is just mind-boggling. In order to give you an idea of the scale they operate at, read the following quote:

"In 2015, AWS almost deployed enough server capacity EVERYDAY to support Amazon in 2005, when it was an $8.49 Billion Enterprise"
- James Hamilton, VP & Distinguished Engineer, AWS, (Re:Invent 2016)

Now, we can assume that enough capacity is being added to support a Fortune 500 company, every day of the year.

To top it off, AWS also uses its own network equipment and its own undersea cables running about 100 waves of 10 GBps each on an optical fiber in full-redundancy.

If you want to hear the full presentation from James Hamilton, with details on the AWS infrastructure at scale, you could visit: https://www.youtube.com/watch?v=AyOAjFNPAbA

Before we start setting up our own little piece of this public cloud, let's understand some of the terminology.

AWS terminology

If you have worked with AWS, you would already know this. However, just to ensure that we are all on the same page, I am mentioning the terminology you need to know to get started with your AWS architecture. Please remember that this is not an exhaustive list.

If you want to know all the different terminology in use, please refer to the glossary at: http://docs.aws.amazon.com/general/latest/gr/glos-chap.html#A.

Account

The easiest way I can explain this is that an account is a billing/payment construct. It is quite literally an account that is allowed to use AWS services. You need to set an account up before we can spin anything up on AWS. It's normally associated with a credit card, which gets charged at the end of a billing cycle.

For enterprises however, the bill of all accounts can be charged to a master paying account.

Region

A region is a geographical tag where AWS resources are located. This is mostly on a city level. For example, North Virginia, Sydney, Mumbai, and so on are regions. Each region comprises multiple Availability Zones. The data transfer out of the regions will typically be carried by some undersea cables, and they are chargeable.

Availability zones (AZ)

An AZ is a collection of one or more data centers, a region will be made out of a few of them (minimum of two). The data centers in the availability zones are connected with high-speed fiber, with a single digit latency.

Virtual private cloud (VPC)

A VPC is a private network that you can create for your own cloud in AWS. In a way, the VPC can be considered your personal network space.

You start out by defining the major subnet, and then divide these into smaller pieces called subnets. VPCs also include gateways to reach the internet or the VPN (called Internet Gateways and VPN gateways respectively).

It is inside these VPCs that you create the resources that are needed. A VPC spans a region, but the subnets span only an AZ.

AWS services

AWS keeps adding services and features to services at such a rapid pace, that it is almost impossible to write about them in a book and expect it to still be relevant. It is for this reason, that I point you to their official website for the list of services that they have currently.

Refer https://www.amazonaws.cn/en/products/ to see the list of services that the cloud offers (at the time of writing, there were about 100 services being offered).

Let's take a look at some of the services:

Service name	Category	Description
Virtual public cloud (VPC)	Networking	Allows us to create private networking and connectivity in ...