OUR PARTNERS:
VDI or Virtual Desktop Infrastructure has been around since 2010 when businesses started to move away from Terminal Services (TS) and Remote Desktop Services (RDS). Citrix and VMware at the time were the only real options before companies like Teradici started to make a name for themselves.
Fast forward to 2023 and there are a myriad of vendors that can provide VDI and Azure now has AVD (Azure Virtual Desktop). The best practice guidelines for designing and building any VDI deployment is to have the VDI infrastructure as close as possible to the Application, this is to mitigate the impact of latency.
Data Scientists and Data Engineers need high performance client machines so a business has two options. Firstly they can purchase high performance laptops or workstations from the likes of HP (Z) and Dell (Precision). These devices range between £2.5k to circa £6k depending on specifications. The devices will most likely be a capex overhead and then depreciated as an asset over 3-5 years.
The second option which is usually the preferred solution is to provide the team members with a high performance VDI which is either from an on premise or cloud based solution. Due to our expertise in design and Architecture, we have created a reference Architecture (Project Cerberus) where each node is designed from the ground up to be capable of running AI, ML, HPC & VDI, the only thing that changes on the node design is the GPU, memory and storage capacity.
This means we can provide ‘App Adjacent VDI’ in a localised cluster where Data Scientists and Data Engineers can consume a high performance VDI that is directly (racked) adjacent to the application nodes. These nodes can then be immersed in an Immersion Cooling tank such as GRC (Green Revolution Cooling) to provide a power efficient platform that meets sustainability goals.
OUR PARTNERS: