From the blog

Expanding HPC in the Cloud to the Nordics

Overview and Key Comments

EDRMedeso is a Premier Reseller Partner of Rescale and an Elite Channel Partner of ANSYS

In this webinar, we dive into the maturity of the Cloud HPC market in the Nordics region. Frode Halvorsen, VP of Innovation at EDRMedeso, joins Cameron Fillmore, Global Channel Partner Manager to discuss their success as a trusted Rescale Partner. EDRMedeso has over 30 years of experience managing accounts in the Nordics ranging from Large Enterprises to Small to Medium Sized Businesses (SMBs). Each has a unique approach for how they are handling the innovative change from on-premise HPC to running HPC in the Cloud. Below are key topics tackled in the webinar and Frode’s responses.

Why cloud suits simulation market in the Nordics

“Simulation market is quite mature, a lot of the large corporations have been using simulation in their product life cycle. These corporations are explicitly saying we will have our last cluster that will be on-premise, moving forward we are investing only in investing in HPC in the cloud”

“I think we are definitely at a tipping point, a lot of the corporations have the cloud first. For the SMB market is much different, and suddenly they can gain access to an unlimited amount of compute. Typically have a powerful workstation at their desk that they use for simulation.”

The reality of DIY cloud versus partnering with Rescale

“At first seemed very reasonable to partner with cloud providers like Microsoft and AWS, and connect the tools together to create a trusted solution. We started developing on it, and we quickly found out that building a solution that is reliable, scalable, secure, and user friendly was a lot bigger task than we anticipated. As we looked at our roadmap for developing an HPC solution, we realized that where we wanted to be in a number of years, Rescale is already there today! We decided not to reinvent the wheel and selected Rescale as our HPC Cloud Solution.”

“The partner onboarding process was very smooth, from the time we signed our agreement, till the time we closed the first deal was only a couple of months!”

“My advice to other companies interested in partnering with Rescale is to move fast! Very often you need to aggressively market a new solution to generate business within your market, but with Cloud, there is more of a pull from the market. Companies are coming to us looking for a cloud solution. When we have a good product like Rescale to fit into our customer’s cloud migration strategy, it makes selling HPC in the Cloud very easy.”

Maintaining excellence at core business competencies

“I have been working in the simulation industry for 20 years and there have always been hardware restrictions with any type of simulation workload. Before Rescale came along, we would need to create a set of jobs that would be solved in an hour, and you would need to create another set of jobs that would solve over-night, and finally, you needed to create the larger jobs that could solve in a week or two. You would always need to create your simulation models to fit within a couple of different time limitations which were due in part by the limited availability of hardware.”

“With Cloud Computing, those older limitations are now removed. If you have a complex simulation that is quite large, it would typically take hours to complete, but now you can instantaneously submit this equation to Rescale and get your results in a fraction of time. This wasn’t an option five years ago. The Rescale solution has really opened up a lot of new opportunities for our customers. The thing with selling Rescale, I don’t feel like we are selling HPC because we are just selling and enabling a solution that allows our customers to do what they need to do.”

Related articles

Deep Learning with Multiple GPUs on Rescale: TensorFlow Tutorial

In a previous post, we showed examples of using multiple GPUs to train a deep neural network (DNN) using the Torch machine learning library. In this post, we will focus on performing multi-GPU training using TensorFlow. In particular, we will […]

read more »

The End of the On-Premise HPC

For a CIO, the ultimate goal is creating a lean and agile IT structure that both meets the current and unanticipated future needs of the various internal teams–without creating a cost structure that is unsustainable or disconnected from justifiable activities. […]

read more »

Leave a Reply

Your email address will not be published. Required fields are marked *