Shulou(Shulou.com)05/31 Report--

In this issue, Xiaobian will bring you about how to quickly build a PaaS layer remote disaster recovery platform through VMware disaster recovery technology. The article is rich in content and analyzes and narrates from a professional perspective. After reading this article, I hope you can gain something.

Taking VMware virtualization platform as the background, Xiaobian provides a small and fresh technical architecture from the software level, aiming to help everyone realize PaaS layer virtual machine remote disaster recovery solution.

I. Foreword

With the increasing popularity of virtualization technology applications, more and more important business systems are running on virtualization platforms. If the production room encounters earthquakes, fires, floods, power outages, etc., how to ensure that its business systems can recover quickly in a short time?

In addition to adopting higher-cost storage replication technology to realize remote disaster recovery backup, there are also less budget investment or free solutions to realize remote disaster recovery backup of business systems.

For virtualization platforms between different units, can business virtual machines backup each other at a lower cost?

Simple to deploy, easy to manage, and easy to maintain virtualized offsite disaster recovery solutions

The following is an introduction to VMware vSphere Replication technology. Through practical operations, you will learn how to quickly build a VMware virtualization platform for off-site disaster recovery and realize off-site backup of your business.

II. VMware Disaster Recovery Technology Implementation

1. VMware vSphere Replication Overview

VMware vSphere Replication is an extension of VMware vCenter Server that provides hypervisor-based virtual machine replication and recovery capabilities. vSphere Replication is an alternative to storage-based replication. It protects virtual machines from partial or entire site failures by replicating virtual machines between:

From source site to target site

Moving from cluster to cluster in one site

From multiple source sites to a shared remote destination site

vSphere Replication offers several benefits compared to storage-based replication:

Low data protection costs per VM.

Replication solutions allow flexible selection of storage vendors for source and target sites.

The overall cost per copy is low.

vSphere Replication can be used with vCenter Server Appliance or standard vCenter Server installations. vCenter Server Appliance can be installed at one site and standard vCenter Server installation at another site. vSphere Replication enables you to quickly and efficiently replicate virtual machines from a source data center to a target site.

Additional vSphere Replication servers can be deployed to meet load balancing requirements. Once the replication infrastructure is set up, you can select the virtual machines to replicate from different recovery point objectives (RPOs). You can enable a multi-point retention policy to store multiple instances of replicated virtual machines. After recovery, the retained instance can be used as a snapshot of the recovered virtual machine.

vSphere Replication does not have a separate license because it is a feature of a specific vSphere license version.

vSphere Essentials Plus

vSphere Standard

vSphere Enterprise

vSphere Enterprise Plus

vSphere Replication is free with a genuine vSphere license, and there is no limit to the number of VMs that can be replicated.

2. How VMware vSphere Replication Works

vSphere Replication allows you to configure replication for a virtual machine from the source site to the target site, monitor and manage replication status, and restore the virtual machine at the target site. When configuring a virtual machine for replication, vSphere Replication Agent sends changed blocks from the virtual machine disks from the source site to the target site. The changed blocks are applied to the copy of the virtual machine. This process is independent of the storage layer. vSphere Replication performs an initial full synchronization of the source virtual machine and its replica. Replication seeds can be used to reduce network traffic generated by data transmission during initial full synchronization.

During the configuration replication process, you can set a recovery point objective (RPO) and retain multiple point-in-time (MPIT) instances. As an administrator, you can monitor and manage the status of replication. You can view information about outbound and inbound replication, local and remote site status, replication problems, and warnings and errors. When you manually restore a virtual machine, vSphere Replication creates a replica of the virtual machine that connects to the replica disk, but does not connect any virtual NIC to the port group. You can view the recovery and status of replica virtual machines and connect them to the network. Virtual machines at different points in time, such as the last known consistent state, can be restored. vSphere Replication presents the retained instance as a normal VM snapshot to which the VM can be restored.

Virtual machines can be replicated between two sites. vSphere Replication is installed on both the source and target sites. Only one vSphere Replication appliance can be deployed on each vCenter Server. Additional vSphere Replication servers can be deployed to replicate between two sites.

3. VMware vSphere Replication bandwidth requirements

Before configuring replication, you need to determine vSphere Replication's storage and network bandwidth requirements for efficient replication of virtual machines. Storage and network bandwidth requirements increase when vSphere Replication is used. The following factors play a role in determining the amount of network bandwidth vSphere Replication requires to perform effective replication.

1)Network based storage

If all storage is network-based, network bandwidth requirements will increase because data operations between hosts and storage will also use the network. When developing deployment plans, keep in mind the following traffic levels:

Transfers between hosts running replication virtual machines and vSphere Replication servers.

Transfers between vSphere Replication servers and hosts that have access to the replication target datastore.

Transfer between host and storage.

Redo logs are transferred between storage and host during snapshots.

Network based storage needs to be considered when replicating virtual machines in a single vCenter Server instance that shares the network for the traffic levels listed. If there are two sites, with one vCenter Server instance on each site, the link speed between the two sites is most important because it slows down replication of traffic between the two sites.

2), data change rate, and recovery point goals

Data change rates are affected by recovery point objectives (RPOs). To estimate the data transfer size for each replication, you must estimate the number of blocks to change for each VM in a given RPO. The rate of data change during this RPO period provides the total number of blocks transferred by vSphere Replication. This number may vary throughout the day, which changes the traffic generated by vSphere Replication at different times. vSphere Replication schedules transmission blocks according to RPO. If RPO is set to one hour, vSphere Replication transfers any blocks changed during that hour to meet the RPO criteria. When vSphere Replication creates a bundle of blocks to transfer, vSphere Replication transfers blocks only once in the current state. vSphere Replication only records that the block has changed during the corresponding RPO period, but does not record the number of changes. Average Daily Data Change Rate provides an estimate of the amount of data transferred by vSphere Replication or how often. If the RPO is changed, vSphere Replication transfers more or less data per replication to meet the requirements of the new RPO.

3)link speed

If you need to transfer bundles averaging 4GB in one hour, you must check the link speed to see if it meets the RPO requirements. If the link is 10Mb, then ideally it would take about an hour to transmit a 4 GB bundle on a fully dedicated link with little overhead. If RPO requirements are met, the 10Mb WAN connection will be saturated. Even under ideal conditions, where there is no overhead or limiting factors such as retransmission, shared traffic, or excessive bursts of data change rates, the connection is saturated. It can be assumed that only about 70% of the links are available for traffic replication. This means that on a 10Mb link, link speeds of approximately 3 GB per hour can be achieved. On a 100Mb link, link speeds of approximately 30GB per hour can be achieved.

III. VMware Disaster Recovery Operation

Environmental Description:

Production Environment vSphere Version: ESXi 6.0

Production VC Version: VMware vCenter 6.7 U2

Production VR: VMware vSphere Replication 8.1.0

Disaster Recovery Environment vSphere Version: ESXi 6.0

Disaster Recovery Environment VC Version: VMware vCenter 6.7 U2

Disaster Recovery Environment VR Version: VMware vSphere Replication 8.1.0

1. vSphere Replication installation

1)Install by importing OVA

2)Deployment of OVF templates

3), select storage, network, set password

4)Binding to vService

5)Installation complete

2. vSphere Replication Configuration

1). Visit https://ip:5480 port for configuration via WEB

2), configuration and VC integration, enter VC login account information

3)VR service startup normal state is running

4)Similarly, VR installation configuration is also required in the same steps at another disaster recovery site.

3. vSphere Replication** Disaster Recovery **

After the previous installation configuration, the exciting moment has arrived. Let's carry out the virtual machine remote disaster recovery operation!

No ads in the middle. Don't go away. Oh, J J J

1)First of all, let's go to VC.

2)Menu->Site Recovery

3)Create site pairs, i.e. production sites-> Disaster Recovery Site Pairing

4)Enter site psc host information

5)Default Next Step

6)2 sites paired

7)On the production side, open the paired site

8)Log in to the disaster recovery site as prompted.

9)Select Copy-New

10)Select replicated virtual machines

10)Select target site, default

11)Select Storage Disk

12), replication RPO settings, policy can be set according to demand

13)Configuration replication relationship completed

14)Start replication synchronization. Note: During synchronization, the virtual machine replicated on the production side needs to be in the boot state.

15), copy complete, log in to the disaster recovery site, select Copy-Reverse Copy

16)Select the virtual machine we have established replication relationship with and perform local registration recovery.

17)During recovery, cancel powering on the virtual machine (if NSX Vxlan is used to connect the two sites, because the mac address and IP information of the virtual machine at the disaster recovery site are the same as those at the production end, conflicts will occur)

18)Select corresponding recovery site resources

19), complete, perform VM recovery

20)After the recovery is completed, the recovered virtual machines can be seen at the disaster recovery site, and the network can be adjusted for rapid service recovery.

21)For qualified customers, VMware NSX VXlan technology is combined to realize the connection of two sites. Just shut down the virtual machine at the production site, select the corresponding VXlan network port group for the virtual machine at the disaster recovery site, and after the virtual machine is started, the network IP address remains unchanged, truly realizing the remote operation of the business.

IV. VMware Disaster Recovery Summary

1. VMware vSphere Replication realizes remote disaster recovery data protection of virtual machines at low cost and high reliability, and customizes RPO synchronization policies according to business requirements;

VMware vSphere Replication When replicating virtual machines between 2 sites, fully consider vSphere Replication storage type and network bandwidth to efficiently replicate virtual machines.

The above is how to quickly build a PaaS layer remote disaster recovery platform through VMware disaster recovery technology shared by Xiaobian. If you happen to have similar doubts, you may wish to refer to the above analysis for understanding. If you want to know more about it, please pay attention to the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.