Considering Hyperconverged Infrastructure for Mission-Critical Workloads? Read This First.

HCI promises greater agility, but can it deliver the consistent performance you need for mission-critical workloads? Enterprise Strategy Group tested a Cisco HCI solution to find out.

contenth
Intel

Fifty-seven percent of organizations like yours already use or have plans to use hyperconverged infrastructure (HCI)[1].

Why the push? It’s all about agility. To remain competitive, enterprises today must be extremely agile and flexible—able to quickly add applications and virtual machines (VMs) to mission-critical production environments.

However, it’s difficult to achieve this level of agility with silos of compute, network, and storage gear that are static and require individual management. By contrast, HCI offers a centrally managed solution with software-defined compute, network, and storage that is flexible, scalable, and easy to deploy.

But what should you look for in an HCI solution to ensure the kind of consistency and performance you need for mission-critical workloads? Testing by Enterprise Strategy Group provides a clue.

Improve Performance for Enterprise Workloads

HCI solutions like Cisco HyperFlex promise greater flexibility for enterprise workloads. Breaking down the silos between compute, storage, and networking is the hallmark of HCI solutions—but traditionally for Tier-2 workloads. With the inclusion of All-NVMe nodes with Intel® Optane™ DC SSDs, Cisco HyperFlex opens the door to powering mission-critical, Tier-1 workloads on HCI.

Compared with previous-generation HyperFlex All Flash deployments, HyperFlex All NVMe with Intel Optane DC SDD provides strong improvements to core measures of HCI performance: IOPS and read, write, and total latency.

What does this mean in real-world applications? Testing undertaken by Enterprise Strategy Group audited HyperFlex All NVMe with Intel Optane DC SDD performance against HyperFlex All Flash running OLTP workloads designed to emulate Oracle and SQL Server workloads. For Oracle workloads, the results included:

71graphic Intel
39graphic Intel

SQL Server workloads saw:

  • 57 percent increase in aggregated IOPS
  • 29 to 39 percent reduction in latency

ESG also evaluated HyperFlex All NVMe with Intel Optane DC SDD using a mixed workload designed to emulate a virtualized environment with multiple VMs running different applications. The results were impressive:

  • 63 percent increase in aggregated IOPs
  • 32 to 40 percent reduction in latency

ESG’s testing shows that Cisco HyperFlex All NVMe with Intel Optane DC SSDs supplies the flexibility of HCI that organizations like yours need while also providing the consistent high performance and low latency that virtualized mission-critical workloads demand. There’s much more to the full story, though, so check out the complete testing results from ESG here.

[1] According to a survey by Enterprise Strategy Group