Hydradyne came to hyper-converged technology relatively early as an alternative to a traditional storage, server and hypervisor setup.
The Fort Worth, Texas-based company was considering purchasing a Dell server, VMware software and EMC storage in 2012 before discovering the Scale Computing HC3 platform.
Mike O'Neil, director of IT at Hydradyne, said the one-box approach to compute, storage and virtualization software was a compelling concept, because it eliminated the licensing costs he would have incurred if he had gone with VMware. He found it much easier and cost-effective to cluster hyper-converged appliances.
"We are a Dell shop and I started down the path straight to VMware. But, at the same time, I was at a trade show in New Orleans and I saw Scale Computing," O'Neil said. "It turned out to be a 50% savings in cost. It was up and running out of the box, and we wouldn't have had to bring in a VMware expert. [The Scale Computing HC3 platform] has all the software built in to it. There's been no looking back since."
Hyper-convergence brings storage, servers, virtualization and management together in a single, unified system. Because Scale Computing HC3 uses an embedded KVM-based hypervisor, customers don't need a virtualization license.
Hydradyne is a fluid power sales and services company in the industrial automation sector, with about 36 locations worldwide and 515 employees. The company started with a four-node cluster of the Scale Computing HC3 2000 hyper-converged appliance that supported 20 virtual servers in 2012. Each node held 2.4 TB of storage, 32 GB of RAM and a single CPU with four cores.
Now, Hydradyne is up to an eight-node cluster of Scale Computing's highest-end HC4000, supporting 50 virtual servers. Each HC4000 node has 128 GB to 256 GB of RAM, 12-core dual-socket CPUs and 9.6 TB of storage. The hyper-converged systems support their Windows Server 2012, ERP database, financial applications and in-house help desk, while email is supported through Microsoft Office 365.
"That's the best part of the platform," O'Neil said. "You can add more nodes, and add them into the cluster and add a pool of available resources. That's what sold me in the end -- the scalability of it."
Scale's HyperCore software provides unified management and the systems have high-availability functionality. Workloads are migrated live across the HC3 appliance during node upgrades. The workloads are returned to the original node when the upgrade is complete.
Scale offers virtual machine-level snapshots, VM thin cloning, and users can export VMs to an external server for a full VM backup and archiving with a single click.
"You have a GUI that you use to ask for more storage RAM and CPU," O'Neil said. "There is nothing you have to do but buy the cluster, load the software and spin it up. There is no virtualization layer above and beyond the actual node itself."
Hyper-convergence means less assembly required
More vendors joined the hyper-converged market in 2015
How hyper-converged appliances change the job of the IT professional
Dig Deeper on Data Storage Solutions for a Virtual Environment
Sonia Lelii asks:
What's the largest hyper-converged cluster you've built?
0 ResponsesJoin the Discussion