kantver - Fotolia
Clusters are certainly not a new idea, but have typically been constrained by the limitations of network bandwidth and latency. After all, cluster nodes must be kept in close synchronization, and network limitations have made it impractical to extend network nodes beyond the same building. This means clusters are often terrific protection against server failures, but not against building failures that can impact an entire building -- such as fires, earthquakes or other large-scale disasters.
What does VMware vMSC do? How does vMSC fit into overall workload availability?
VSphere has introduced the idea of stretched clusters, which can coordinate node communication and synchronization across campus- or metropolitan-scale physical distances. VMware calls this vSphere Metro Storage Cluster (vMSC) technology. By extending cluster nodes between buildings -- data centers -- located across a larger geographic area, the business can boost its disaster recovery by preventing a catastrophe in one facility from rendering a critical workload completely unavailable.
When properly implemented and configured, a vMSC deployment offers a fully active and workload-balanced environment, allowing downtime and disaster avoidance through non-disruptive live migrations between different data centers. Considering the relative immaturity of stretched clustering, vMSC is generally not recommended as a primary DR infrastructure, but rather as a means of managing data center performance and preventing or avoiding disasters.
VMSC takes advantage of vSphere High Availability (HA), vSphere Distributed Resource Scheduler (DRS) and vSphere Storage DRS to configure the constituent cluster nodes. HA checks node and data store heartbeats to determine available -- working -- cluster nodes, while DRS handles the load balancing within the metro cluster. This is important when it comes to disaster recovery.
VMSC supports Fibre Channel (FC), iSCSI, network file system (NFS) and Fibre Channel over Ethernet (FCoE) protocols, but vMSC links require low latency and high-bandwidth network connectivity for best operation. For example, the maximum latency between sites for ESXi management and synchronous storage replication is 10 milliseconds round trip time (RTT). Even though the maximum latency for vMotion and Storage vMotion is up to 150 milliseconds, a minimum bandwidth of 250 Mbps is required for vMotion.
How does vSphere Metro Storage Cluster differ from Site Recovery Manager?
How much do you know about vMotion requirements for network and storage?
Metro storage clusters manage data growth for IaaS providers
Dig Deeper on Backing up VMware host servers and guest OSes
Related Q&A from Stephen J. Bigelow
Learn how load balancing in the cloud differs from a traditional network traffic distribution, and explore services available from AWS, Google and ... Continue Reading
Access management is critical to securing the cloud. Understand the differences between AWS IAM roles and users to properly restrict access to AWS ... Continue Reading
Containers have rapidly come into focus as a popular option for deploying applications, but they have limitations and are fundamentally different ... Continue Reading