Availability and resiliency are top of mind for everyone involved in providing a platform for business services. VMware offers a great solution to increase resiliency and availability through the VMware vSAN stretched cluster functionality.
Over the past few years, new features and enhancements have been introduced to vSAN stretched clusters to increase availability and resiliency as well as improve operational efficiency.
In this session, Vivek and Duncan discuss a typical implementation, share best practices, and explain why vSAN Express Storage Architecture is preferred for VMware-based stretched cluster solutions.
Don’t have time to watch the full session? Here’s a rundown of key takeaways:
1. VMware is the clear market leader for hyper-converged infrastructure (HCI), with vSAN being the driving force behind it. With over $1 billion in revenue per year and more than 37,000 customers, vSAN has become the preferred choice for HCI deployments.
2. vSAN is an enterprise storage solution that is integrated into the hypervisor, allowing for the aggregation of storage devices into a single cluster-wide shared datastore. It is easily scalable and managed through storage policies in the vCenter Server, making it suitable for running both business-critical and cloud-native applications.
3. Data services in vSAN can be easily tailored to meet the needs of the organization. These services include space efficiency, data-at-rest encryption, data-in-transit encryption, iSCSI target service, and file services. They can be enabled, or disabled, based on organizational and application requirements.
4. vSAN provides granular storage policy-based management (SPBM), allowing administrators to define storage protection and performance outcomes for virtual machines (VMs) and their virtual disks (VMDKs). This enables the customization of storage policies based on the specific needs of each application.
5. A vSAN stretched cluster extends a single vSAN cluster across two physical sites, providing data availability during failures within a site or failures of an entire site. Resiliency settings are applied using storage policies, and the stretched cluster is integrated with vSphere HA and DRS for seamless failover and load balancing.
6. The topology of a vSAN stretched cluster heavily depends on the inter-site link (ISL) performance. The performance of VMs across the stretched cluster is only as good as the ISL used. It is important to consider the capabilities of the ISL and the host hardware and local network when designing a stretched cluster.
7. vSAN provides the ability to isolate witness traffic for stretched clusters, ensuring that it is fully separated from vSAN data traffic. This prevents data traffic from being routed via the vmknic for witness host traffic upon link failure. Different MTU sizes can be used for ISL used by vSAN data traffic and the link to the witness host.
8. Site affinity policy settings in vSAN allow for the specification of a single site location for a VM’s objects when site-level protection is unnecessary. This reduces network and storage requirements and is ideal for solutions that already have application redundancy in place.
9. The presenters emphasized the importance of testing failure scenarios in stretched cluster configurations. They recommended reading the vSAN stretched cluster documentation, bandwidth sizing documentation, and the vSAN failure scenario matrix to ensure a successful implementation.
Want more VMware Explore? Dive into our full video library for unlimited learning at your own pace. And stay up to date with the latest product announcements on the official vSAN blog.
