At CenterGrid, we go beyond just providing infrastructure; we offer a complete turnkey managed IT service. Our reputation is built on quickly responding to unexpected IT requests and managing projects with strict adherence to timelines. This requires a flexible and robust infrastructure, especially for our Media and Entertainment (M&E) customers, who demand rapid workload standup and high-powered GPUs for Visual Effects (VFX), Computer-Generated Imagery (CGI), and AI.
At-A-Glance
Our legacy infrastructure software posed significant technical and financial challenges. It had a restrictive list of supported hardware, a substantial virtualization tax consuming up to 20% of CPU resources, and lacked support for the GPUs crucial to our M&E clients. Financially, the per CPU/core licensing was a significant part of our IT budget, making it difficult to invest in ultra-dense servers to save on data center space and reduce power and cooling costs.
We embraced the challenge of finding a new infrastructure software solution, building an extensive lab and a detailed test suite that mimicked our customer workloads. The objective – to reduce costs and simplify operations without compromising our customers’ experience.
After extensive testing of various hypervisors and open-source solutions, we discovered VergeIO. Unlike others, VergeIO’s UltraConverged Infrastructure (UCI) integrates networking, virtualization, and storage into a single operating platform called VergeOS, with a server-based licensing model that allowed us to build dense servers without increasing costs. (VergeOS is licensed by the physical server, not by the number of CPUs or cores or the amount of RAM or storage capacity. VergeIO’s stance on licensing enables CenterGrid to build highly dense servers into its infrastructure without raising software licensing costs.)
As part of the operations testing, the scope included how VergeOS responded to various failure conditions, such as drive loss or server downtime. The workload continued running in each case and was appropriately moved to surviving nodes. We also tested adding new nodes to the VergeOS instance and found it to be the most straightforward process of any product we’ve tested, including so-called turnkey solutions.
VergeOS further impressed us with its seamless migration capabilities and better hardware support, especially for GPUs. It outperformed all other solutions in our benchmarks and was the easiest to set up and operate. The interface was highly responsive, even in large environments, significantly improving our operations team’s efficiency.
Our proof of concept with VergeOS turned into a full production rollout within a week, setting up high-performance NVMe storage and GPUs for a public real-time rendering website. The ease of node addition and the overall reliability of VergeOS have been game-changers for us.
Today, we continue to expand our VergeOS instance, taking full advantage of its automation capabilities with Terraform and Ansible. The intuitive and responsive user interface saves us hours daily, and the support from VergeIO’s Customer Success team has been exceptional.
In conclusion, thanks to VergeOS, CenterGrid’s infrastructure is now precisely where it needs to be. We can focus on servicing our customers and attracting new ones without worrying about infrastructure limitations.
“One of the things that shocked us was how responsive the interface was even in a large environment with dozens of virtual machines. Screen refreshes that can take minutes in our legacy hypervisor happen instantly in VergeOS.”
Chris Beard | COO Centergrid
