As innovative companies shape the landscapes of their industries with their growth, the need for even more computing power becomes undeniable. For Hashgraph, a pioneering blockchain company, this meant transitioning from the public cloud to dedicated servers. 

In case you're wondering, this led not only to stronger single-tenant bare metal servers hosting their workloads but also cost-efficient scaling.

Company Origins and Early Infrastructure

Hashgraph, originally known as Hedera, was founded to revolutionize distributed ledgers through their proprietary Hashgraph algorithm. The company's early infrastructure relied heavily on Google Cloud Platform (GCP) for both production and pre-production environments. Dan Alvizu, an early team member, worked closely with Nathan Klick from Swirlds, leading to a partnership that culminated in their 2023 rebranding to Hashgraph.

The Challenge: Unsustainable Cloud Costs

As Hashgraph's operations expanded, particularly in CI/CD testing and performance validation, their cloud costs became unsustainable. "Cloud costs had become a rising concern. That was somewhat natural for where we were as a company transitioning from an early-stage startup, mid-stage startup to a late-stage startup," Dan explained.

I shrunk our actual compute cost in GKE from $60k to $75k a month to $10k a month. I'm running the same amount of workloads, with some 20% overhead when we bought the hardware and Latitude. So that was a huge savings.

Nathan Klick, Sr. Engineer Manager

Finding the Right Partner

While exploring alternatives, Hashgraph encountered various challenges. Previous experiences with other providers were disappointing: "We started a container process and the whole system restarts—just like some hard kernel lockup we assume happened," Dan recalled. And support from other providers was minimal, as he would often hear things like: "Try slapping on more CPU base clock and then try again."

For Dan and Nathan, Latitude.sh’s consistent performance and API support became the foundation of their decision to move away from other cloud providers and embrace a more reliable, cost-efficient bare metal infrastructure. The transition to Latitude.sh proved to be different and Nathan praised our approach: "Your support team has always been very much customer-centric...They either come back and say, 'You know what, it was a bug on our end, we just fixed it, try again’ or in other situations when they wouldn't find any issues on your end, they would still help us find a fix to our problem.”

The Results

The partnership with Latitude.sh exceeded expectations in both performance and support.

The fact that we have a direct line to the account manager and other folks like the Network Engineering has been huge.

Nathan Klick, Sr. Engineer Manager

The support team proved to be particularly valuable during technical challenges, such as setting up BGP connections.

"When I say we get what we pay for, we definitely get what we pay for," Nathan emphasized, contrasting this with their GCP experience where benchmark results fell short of advertised performance. The move to Latitude.sh has positioned Hashgraph for continued growth, and increased performance while maintaining cost efficiency.

"You guys have gone the extra mile to really help us work out a solution to address it, which I think is going to be cost-effective and keep our costs where we need them to be," Nathan concluded.