Demystifying Datastores: A Complete Guide to Storage Options in vSphere 7

For virtualized workloads, few architectural decisions are as foundational as choosing the storage infrastructure. TheWRONG datastore can cripple performance, availability and scalability regardless of how well you configure everything above it.

That’s why it’s essential to understand the core vSphere 7 datastore options and map their strengths to your specific application requirements.

In this comprehensive guide, we’ll explore:

  • VMFS
  • NFS
  • vSAN
  • vVols

By the end, you’ll have clarity on which datastore technologies align to different use cases — empowering you to make strategic investments that pay dividends across all your virtual workloads.

The Critical Role of Datastores in vSphere

Before diving into the individual datastore types, it’s useful to understand why datastore selection carries such weight in vSphere designs.

At a high level, vSphere datastores provide the storage volumes for virtual machine files as well as files necessary to manage the environment like ISO images. But the attributes of different datastores directly impact:

Performance

  • IOPS/throughput caps based on protocol and hardware
  • Latency profiles tied to resiliency mechanisms

Scalability

  • Per-volume limits on capacity and file count
  • Hypervisor and backend array bottlenecks

Availability

  • Data protection and redundancy models
  • Backup/replication support

Manageability

  • Provisioning and allocation efficiency
  • Monitoring, automation and maintenance overhead

Security

  • Access controls
  • Encryption capabilities
  • Vulnerabilities of components/protocols

Beyond this foundational vSphere integration, datastore selections have follow-on effects that ripple through other layers like cloud management platforms, backup tools and disaster recovery workflows.

Clearly, not all datastores are created equal. And the type best suited for your VM workloads depends entirely on technical requirements and operational constraints:

Datastore Selection Criteria

Now let’s explore how traditional and emerging vSphere 7 datastore technologies map to these decision criteria…

VMFS – Mature, Performant Local Storage

VMFS (Virtual Machine File System) provides fast, scalable shared storage for virtual machines by pooling local physical storage. It has long served as the staple datastore option for vSphere environments.

Technical Capabilities

Some key technical capabilities:

  • Block-based – Built on SAN/NAS devices surfaced as block storage logical units (LUNs)
  • Cluster shared – Can be simultaneously accessed by all hosts in a cluster
  • Scalable formats – VMFS6 extends capacity to 64TB with 1MB file block size
  • Fast performance – Designed for high IOPS with low latency
  • Resilient – Leverages distributed locking and redundancy mechanisms

VMFS is well-suited for read-intensive workloads with demanding latency requirements. Real-world benchmarks show VMFS datastores easily sustaining 30K+ IOPS with sub-millisecond response times.

Common Deployment Architectures

Typical VMFS deployment architectures include:

  • Fibre Channel SAN arrays
  • iSCSI SAN arrays
  • Shared SAS disk enclosures
  • Local SSDs, NVMe devices aggregated via VSAN

FC and FCoE offer the fastest and most consistent performance profiles. But most organizations opt for more cost-effective iSCSI-based arrays — accepting slightly higher latency for large $/GB savings.

When aggregated across a cluster, local NVMe SSDs can match or even outpace the IOPS of shared SANs. This makes VSAN on SSD an compelling option for extreme performance.

Scalability and Hardware Limits

VMFS scalability limits to consider:

  • 1 PB per LUN volume limit
  • 64 TB VMFS6 volume limit
  • VMFS metadata size caps that constrain file count

The vSphere High Availability mechanism can reduce storage capacity by 50-75% for replica VMs.

Stretch clusters are technically possible but perform poorly and add significant complexity.

When is VMFS the Right Choice?

VMFS aligns well to these common use cases:

  • Transactional databases – Low latency and high consistency critical for OLTP
  • High IOPS applications – Sustained drive performance realized by workloads
  • Localized clusters – Shared access limits need for complex replication

For high performance virtualized applications relying on low latency storage, VMFS leverages familiar Fibre Channel and iSCSI capability providing battle-tested reliability at scale.

Now let‘s explore how VMware modernized virtual storage to support more dispersed usage patterns…

NFS – Simple, Distributed File Sharing

Whereas VMFS relies on SAN-style block storage, VMware leveraged the ubiquitous and mature NFS protocol to unlock NAS benefits for vSphere users…