Check the InfiniBand placement boundary before troubleshooting MPI reachability
Are the HPC VMs positioned to use Azure's RDMA fabric rather than only sharing ordinary Ethernet connectivity?
Read the checklistDSE security knowledge hubNetworks & Infrastructure
Networks & Infrastructure
Knowledge
Page 4 of the DSE Networks & Infrastructure knowledge center, with source-backed guidance and practical next steps.
DSE-authoredOfficial sourcesReviewed guidance
Explore this topic
Start with the cornerstone guide Networks & Infrastructure knowledge center
Guidance for resilient networks, Wi-Fi, firewalls, cabling, power, exposure reduction, and connected-device planning.
DSE post stream
1,200 articlesNetworks & Infrastructure
Discover the HPC fabric partition key instead of copying an MPI example value
Which partition key should an MPI job use after deployment into a different Azure HPC tenant boundary?
Read the guide
Account for fault-domain alignment when reattaching Premium SSD v2 disks
Why can repeated disk moves fail while Azure is aligning Premium SSD v2 storage with an availability-set VM?
Read the guide
Package Linux application dependencies for the VM's actual repository access
Will a VM Application package install when its target Linux VM cannot download missing dependencies?
Read the checklist
Separate durable HPC storage from the layer that accelerates a job
Which part of an HPC storage proposal is the dataset's permanent home, and is a separate accelerator justified?
Read the guide
Apply Azure VM DNS changes through network configuration and DHCP renewal
Review Azure network-level DNS settings and their delivery to running VMs before treating a guest-side change as persistent.
Read the guide
Preserve Virtual WAN route configuration before enabling routing intent
Prepare an explicit restoration plan because removing routing intent does not restore the previous hub configuration.
Read the guide
Budget Application Gateway v2 subnet addresses for peak instance count
Include Azure reservations, gateway instances, and private frontend addresses in subnet capacity planning.
Read the guide
Check an active metric alert before changing repeat-notification behavior
Why can a continuing metric breach produce no new alert, and what changes if the rule becomes stateless?
Read the explainer
Validate the specific availability test beyond its shared source addresses
Why is allowing Application Insights availability-test addresses insufficient to identify your own test?
Read the guide