Module 7

Overview of high availability and

disaster recovery
Module Overview

• Defining levels of availability

• Planning high availability and disaster recovery
solutions with Hyper-V virtual machines
• Backing up and restoring by using Windows
Server Backup
• High Availability with failover clustering in
Windows Server 2016
Lesson 1: Defining levels of availability

• What is high availability?

• What is continuous availability?
• What is business continuity?
• Creating a disaster recovery plan
• Highly available networking
• Highly available storage
• Highly available compute or hardware functions
What is high availability?

All parts of an application and the infrastructure it

relies on must be highly available:

• Data center infrastructure • Network infrastructure

• Power, HVAC, Location • Switches, Routers, configurations

• Server hardware • Internet

• NIC, Power • ISPs

• Storage • Network services

• RAID • AD DS, DNS, DHCP
What is continuous availability?

To provide continuous availability:

• Perform business impact analysis
• Perform risk analysis
• Perform application specific analysis
• Create different continuous availability strategies for
different applications
What is business continuity?

Requirements for business continuity planning

should include:
• SLAs for the IT systems
• Contact info and technical background of personnel
assigned to recovery
• A secondary site
• Workaround solutions
• Maximum outages allowed for applications
Business continuity technologies and data collection

You can collect business continuity data from:

• Business impact analysis
• Risk analysis

Technologies for business continuity include:

• NLB
• Failover clustering on physical or virtual machines
• Application-aware high availability
• Conventional data backups
• Online backups
• Virtual machine backups
Creating a disaster recovery plan

• Developing a recovery plan includes:

• Performing a risk analysis
• Define what data should be recovered
• Define where data should be recovered
• Define when to recover data

• Recovery plan should be tested on a regular

basis

• Recovery plan should be evaluated on a regular

basis
SLA components

SLA components can include:

• Hours of operation
• Service availability
• Recovery point objective (RPO)
• Recovery time objective (RTO)
• Retention objectives
• System performance
 SLA components

AVAILABILITY IN MINUTES

Minutes 85.000 90.000 95.000 97.000 99.000 99.900 99.990 99.999

YEAR (365d) 525600 78840 52560 26280 15768 5256 525.6 52.56 5.256

MONTH (30d) 43200 6480 4320 2160 1296 432 43.2 4.32 0.432

BI-WEEKLY 20160 3024 2016 1008 604.8 201.6 20.16 2.016 0.2016

WEEKLY 10080 1512 1008 504 302.4 100.8 10.08 1.008 0.1008

DAILY 1440 216 144 72 43.2 14.4 1.44 0.144 0.0144

HOURLY 60 9 6 3 1.8 0.6 0.06 0.006 0.0006

MINS 1 0.15 0.1 0.05 0.03 0.01 0.001 1E-04 1E-05

Highly available networking

Planning for high availability in networking should

include redundancy for:
• Network adapters
• Multipath I/O
• Local Area Network
• Wide Area Network
• Internet connectivity
Highly available storage

When planning high availability for storage,

consider following technologies:
• RAID
• DAS
• NAS
• SAN
• Cloud services
Highly available compute or hardware functions

• Consider using the high availability features that

are built in to the operating system:
• Failover clustering
• Network Load Balancing
• RAID

• Follow the best practice guidelines and

recommendations for the specific application
Lesson 2: Planning high availability and disaster
recovery solutions with Hyper-V virtual machines
• High availability considerations with Hyper-V virtual
machines
• Overview of Live Migration
• Live migration requirements
• Demonstration: Configuring live migration (optional)
• Providing high availability with storage migration
• Demonstration: Configuring storage migration (optional)
• Overview of Hyper-V Replica
• Planning for Hyper-V Replica
• Implementing Hyper-V Replica
• Demonstration: Implementing Hyper-V Replica (optional)
High availability considerations with Hyper-V
virtual machines

High availability options Description

Host clustering • Virtual machines are highly
available
• Does not require virtual
machine operating system or
application to be cluster aware
Guest clustering • Virtual machines are failover
cluster nodes
• Virtual machine applications
must be cluster aware
• Requires iSCSI or virtual Fibre
Channel interface for shared
storage connections
NLB • Virtual machines are NLB cluster
nodes
• Use for web-based applications
Virtual machine migration options

Available options for moving virtual machines are:

• Virtual machine and storage migration
• Quick Migration
• Live Migration
• Hyper-V Replica
• Export or import of a virtual machine
Overview of live migration

• Migration setup:
• A TCP connection is established with the target physical host.
• A temporary virtual machine is created on the target physical host
• Guest-memory transfer:
• Guest memory is transferred to the target host
• Hyper-V monitors and tracks as the system modifies memory pages
• Once the transfer gets to 10 dirty pages (old memory pages) copying
stops
• State transfer:
• Hyper-V stops the source partition
• Transfers the state of the virtual machine, including dirty memory pages
• The virtual machine is restored on the target host
• Cleanup:
• Tears down the virtual machine on the source host
• Terminating the existing threads
Overview of live migration

Cluster
Storage

Node 1 Node 2
Live migration requirements

Live migration requirements include:

• Live migration enabled
• Host computers processor requirements
• Host computers domain membership and user
accounts configured
• Hyper-V roles and management tools installed
• Host computers authentication configured
• Host computers performance, network, and bandwidth
configured
Demonstration: Configuring live migration (optional)

In this demonstration, you will see how to enable

and configure live migration
Providing high availability with storage migration

• Virtual Machine and Storage Migration technology

enables you to move a virtual machine and its storage
to another location without downtime
• During migration, the virtual machine hard disk is copied
from one location to another
• Changes are written to both the source and destination
drive
• You can move virtual machine storage to the same host,
another host, or an SMB share
• Storage and virtual machine configuration can be in
different locations
Demonstration: Configuring storage migration
(optional)

In this demonstration, you will see how to enable

and configure storage migration
Overview of Hyper-V Replica
Hyper-V Replica in Windows Server 2016 enables you to
replicate a single virtual machine over a WAN or LAN
network to another host. Hyper-V Replica components
include:
• Replication engine
• Change tracking
• Network module
• Hyper-V Replica Broker role

Replication traffic Recovery Replication traffic Recovery

CRM virtual machines virtual machines
File
IIS
SQL

WAN link WAN link

SAN\NAS SAN\NAS SAN\NAS

Primary site Replica site Extended Replica site

Planning for Hyper-V Replica

Use Hyper-V Replica features in Windows

Server 2016 to:
• Change the replication frequency to either 30 seconds,
5 minutes, or 15 minutes
• Extend replication to include a third host
Implementing Hyper-V Replica

Hyper-V Replica has the following prerequisites:

• The server hardware supports the Hyper-V role on
Windows Server 2016
• Sufficient storage exists on both the primary and
replica servers
• Network connectivity exists between the locations that
host the primary and replica servers
• Firewall rules are correctly configured to enable
replication between the primary and replica sites
(default is TCP port 80 or 443).
• An X.509v3 certificate exists to support Mutual
Authentication with certificates
Hyper-V Replica configuration steps

To configure Hyper-V Replica, you should:

1. Configure authentication and port options
2. Select replica servers
3. Select location for replica files and replication interval
4. Enable replication on virtual machine

Hyper-V Replica failover scenarios include:

• Test failover
• Manual failover
• Failover
Demonstration: Implementing Hyper-V Replica
(optional)

In this demonstration, you will see how to

implement Hyper-V Replica
Lesson 3: Backing up and restoring by using
Windows Server Backup

• Overview of Windows Server Backup

• Implementing backup and restore
Overview of Windows Server Backup

By using Windows Server Backup you can:

• Perform a full server backup and bare-metal restore
• Back up and restore system state
• Back up and restore individual files and folders
• Exclude selected files or file types
• Select from more storage locations
• Perform a Windows Azure Online Backup
Implementing backup and restore

Backup and restore operations include:

• Backing up and restoring Hyper-V hosts
• Backing up and restoring VMs
• Backing up and restoring AD DS, file servers, and web
servers
• Azure Site Recovery
• Each server requires an installation of Windows Server
Backup
• MS SQL, Exchange, and Skype for Business require
different best practices for backup
Lesson 4: High availability with failover clustering
in Windows Server 2016

• What is failover clustering?

• High availability with failover clustering
• Clustering terminology
• Clustering categories and types
• Failover clustering components
• Technology redundancy comparison
What is failover clustering?
Shared bus or Service or
iSCSI connection application

Cluster storage

Node 1 Node 2
A dedicated network
connects the failover
cluster nodes

A network connects Clients

the failover cluster and
clients
High availability with failover clustering

• Failover clustering provides high availability for

data, applications, and services
• Failover clustering considerations:
• Hardware prerequisites
• Software prerequisites
• Applications have specific failover clustering
configurations
• Applications must be cluster-aware
High availability with failover clustering

• Failover clustering in Windows Server operating system

provides high availability for the following applications
and features:
• DFS Namespace Server
• DHCP Server
• Distributed Transaction Coordinator (DTC)
• File Server
• Internet Storage Name Service (iSNS) Server
• Message Queuing
• Other Server
• Print Server
• Remote Desktop Connection Broker
• Virtual Machine
• WINS Server
Clustering terminology

Failover clustering terminology includes:

• Node
• A Windows Server 2016 computer that is part of a failover cluster
• Service or application
• A service that can be moved between cluster nodes
• Shared storage
• Storage that is accessible to all cluster nodes
• Quorum
• The number of elements that must be online for a cluster to continue
to run
• Witness
• “Tiebreaker” Server, disk or share
• Failover/Failback
• Clients
Clustering categories and types

• Type of application deployed:

• Failover clusters
• Network Load Balancing clusters

• Node location:
• Single site clusters
• Multisite clusters
• Nodes or witness server hosted in cloud environment

• Number of active servers:

• Active-Active clusters
• Active-Passive clusters
Failover clustering components

Failover clustering components include:

• Nodes
• Network
• A network across which cluster nodes can communicate
• Resource
• A node hosts a resource like shares, and applications
• Cluster storage
• A storage system that cluster nodes share
• Quorum
• Witness
• Service or application
• Clients
Technology redundancy comparison

Zero Hardware Site Data Automatic

Downtime Failures Failures deletion/ failover
corruption
Live Yes No No No No
Migration
Clustering Depends Yes Depends No Yes
on on
application application
Hyper-V No Yes Yes Depends No
Replica on
application
Windows No Yes Depends Yes No
Server on scenario
Backup
Lab: Planning and implementing a high
availability and disaster recovery solution

• Exercise 1: Determine the appropriate high

availability and disaster recovery solution
• Exercise 2: Implementing storage migration
• Exercise 3: Configuring Hyper-V Replicas
Logon Information
Virtual machines: 20740C-LON-DC1-B
20740C-LON-SVR1-B
Host machines: 20740C-LON-HOST1
20740C-LON-NVHOST2
User name: Adatum\Administrator
Password: Pa55w.rd

Estimated Time: 75 minutes

Lab Scenario

Adatum Corporation is looking to assess and

configure the new high availability features and
technologies that they can leverage. As the system
administrator, you have been tasked with
performing that assessment and implementation.
Lab Review

• How can you extend Hyper-V Replica in Windows

Server 2016?
• What is the difference between Live Migration and
Storage Migration?
Module Review and Takeaways

• Best Practices
• Common Issues and Troubleshooting Tips
• Review Question
• Tools