GPUs, CPUs, and VDI, oh my! GPUs get all the credit for rendering stunning displays for fancy graphics-intense applications. However, CPUs are the true unsung heroes of the high-performance computing (HPC) that solves some of the biggest, most complex problems of our time. By maximizing CPU usage using cloud-hosted or on-prem VDI, you can deliver a virtual HPC machine, anywhere, anytime, for a fraction of the cost.


We spend a lot of time talking about the display protocols that support graphic-intense workloads. But, before you render a single pixel, a different part of your workstation comes into play, the CPU. So, let’s play a little homage to high performance computing, or HPC!

 shutterstock_395065996.jpg

 

First, what is high performance computing?

If we go straight to the source, inside HPC says, “High Performance Computing most generally refers to the practice of aggregating computing power in a way that delivers much higher performance than one could get out of a typical desktop computer or workstation in order to solve large problems in science, engineering, or business.

HPC makes complex calculations possible, enabling scientists, engineers, and others to put order to our unorderly world. But, HPC is expensive! The top 100 or so supercomputers cost upwards of $20M each! However, these are machines that are solving problems like the origins of the universe, and curing cancer. What does HPC really mean to the average SMB or even the enterprise?

 

What can HPC do for you?

HPC nodes work together to solve tough problems that wouldn’t be solvable by an individual machine. These HPC desktops are invaluable in sectors such as scientific research, engineering, and even gaming. For the MSP, HPC virtual desktops present a unique opportunity to solve one of the most common pain points of HPC, maximizing resource utilization.

Because these machines are costly by nature, these highly specialized machines are often shared across multiple departments and research units and the wait time to access these machines can become impossibly long. By offering HPC desktops in a cloud or on-prem VDI environment using an advanced connection broker, CPU resources can be maximized for optimal utilization, increasing user efficiency and productivity.

 

HPC For VDI

Historically, it was thought virtualizing HPC simply couldn’t be done. However, advancements in hypervisor and virtualization technology have finally made vHPC a reality.

In a VDI environment, CPUs are easily mobilized and shared amongst workstations to deliver HPC power where it is needed most. Moreover, with increasing HPC compute offerings from the major public cloud vendors such as Amazon AWS and Microsoft Azure, it’s easy to spin up an HPC cloud environment without purchasing a single piece of hardware. HPC in a cloud environment can be accessed from any client device with connectivity, so HPC workstations are as mobile as the user.

GPUs take all the credit for rendering stunning visuals for graphics-intense applications, but ultimately the less flashy CPUs are solving some of the biggest, and most costly, challenges in computing. By leveraging VDI or DaaS, MSPs and IT professionals are presented with a unique opportunity to help solve some of the most challenging and costly pain points for the sectors which rely on HPC.