NVIDIA GPU Server Options for Inference and Training
This idea was imported from Canny. Originally created by: William Hector. The current owner is: Ravinder Braich. GPU enabled server platform for Workload Optimized configs.
This idea was imported from Canny. Originally created by: Justin Beck. The current owner is: Unassigned. The ability to support double tagged or stacked tags to match Equinix Fabric network design. Some service provider use cases require the abili...
Support User Collaborator Role with No Pricing Visibility
This idea was imported from Canny. Originally created by: Zachary Smith. The current owner is: Unassigned. Some collaborators may need access to provision and access services, but not be privy to pricing details, invoices or MTD charges.
This idea was imported from Canny. Originally created by: Charlie Lane. The current owner is: Unassigned. Offer Next Gen Altra Mt. Snow - Single socket 80 core CPU, and/or Mt. Jade Dual socket 80 core CPUs on our server lineup.
Potential Build op...
This idea was imported from Canny. Originally created by: Zachary Smith. The current owner is: Unassigned. Support mixed Layer2 + Layer3 mode while keeping bonded ports for High Availability.
Support for both L3 (Internet VLAN) and L2 (private VL...
This idea was imported from Canny. Originally created by: Benjamin Jenkins. The current owner is: Unassigned. e=edge. It would be amazing to have a true edge device to run router OS software and bridge L3 and L2 with 4 physical ports (2 visible bo...
This idea was imported from Canny. Originally created by: Zachary Smith. The current owner is: Unassigned. Connect to Equinix Metal from ECX, allowing you to create connections between your bare metal compute infrastructure in a Layer 2 VLAN and o...
This idea was imported from Canny. Originally created by: Enkel Prifti. The current owner is: Unassigned. It would be great to have the workflow for creating shared Fabric connections automated similar to how it works on AWS.
The proposed workflow...
High Memory Configuration for Hyper-converged or In Memory Workloads
This idea was imported from Canny. Originally created by: Damian Sieczkowski. The current owner is: Unassigned. We should consider a premium server offering with 1.5TB+ of RAM to meet hyper-converged and in-memory server workloads.
This idea was imported from Canny. Originally created by: Andrew Rynhard. The current owner is: Unassigned. Support Talos as an official Equinix Metal Operating System.