With time, some newer architectures are being practised which is an attempt to deliver workload mobility, availability and disaster recovery-all within one solution like Stretched Clusters that span sites as well as disaster recovery automation tools such as VMware vCenter Site Recovery Manager.
The reason to opt for VMware based virtualization for any organization is to use the advanced VMware HA features.
Very often, Stretched Cluster and SRM are compared with each other but this is not the correct placement of the tool as they provide solutions for different requirements.
Stretched Clusters is having the same cluster of vCenter stretched across 2 different geographical sites managing ESXi host from both sites from a single cluster and use the DRS function for load balancing and achieving better utilization of resources.
The decision to use Stretched cluster should be taken based on the tradeoff between Pros and Cons mentioned below
Pros: To enable both sites hardware to be utilized inefficient way being with an active-active model by using vSphere vMotion.
When enabling multi-site load balancing. This architecture is dependent on proactive management and vMotion.
Cons:
- VMware HA is not site aware due to which no guaranty can be provided for startup order sequencing.
- There is no ability to ensure or report free resources and capacity to provide a successful failover.
- Consumes more storage with Synchronous Replication in place, higher cost of ownership.
- Can impact the host affinity rules and VMware HA admission controls.
- It requires constant Operational management and change control.
Prerequisites for stretched clusters:
- Enterprise License for ESXi host.
- Sites must be closer than approximately 100 kilometres apart
- Round trip time between sites must be 5 ms or less connected through a redundant link of Minimum of 622 Mbps.
- Synchronous replication between sites which is going to consume an equal amount of storage on both sites.
Site Recovery Manager is to provide managed and consistent automated processes for restoring services after an outage, coupled with non-intrusive testing and reporting for routine confirmation of the capability of restored services.
Pros:
- Centralized and repeatedly tested recovery plans to ensure a successful recovery.
- Can be proactive or reactive after sustaining an outage.
Cons:
- Where the priority of recoverability is lower than the priority of increased local or site resilience.
- Per socket or OSI pack license required.
Note: Stretched Clusters and SRM provide different solutions, Where Stretch Cluster’s focus on Data Availability and service mobility and SRM focuses on Controlled and repeatable disaster recovery processes to recover from outages.
Conclusion:
A stretched cluster provides live migration of workloads with load balancing of data centers but with minimally controllable disaster recovery, while disaster recovery models provide controlled, repeatable, any-distance, orchestrated disaster recovery but require an outage and restart of all virtual machines and services being moved between sites.
Source: Stretched Clusters & VMware Site Recovery Manager.pdf
- How to Implement Hytrust KMS and enable vSAN encryption? - 16/10/2022
- How to delete Inaccessible objects in vSAN datastore? - 16/09/2022
- Availability options with vROPs 8. - 04/07/2022
