AZURE Updates 2026-11-19 | Cloud Provider News
Spot Smarter with Azure Database Hub’s Real-Time Dashboards
Performance problems and wasted resources often hide across databases. Database Hub in Microsoft Fabric now provides real-time performance dashboards for SQL Server, Azure Database for PostgreSQL Flexible Server, and Azure Cosmos DB.
Better cross-database observability helps teams spot waste, underutilization, and performance issues that drive cost. Having all of that in one place makes it easier to decide where to act first.
Troubleshoot Storage Bottlenecks Faster with Azure Boost NVMe Observability
I/O-heavy workloads often hide cost problems in plain sight. Azure Boost observability for NVMe disks adds disk I/O and latency metrics, plus disk-to-device mapping, inside compatible Azure VMs.
These additions make storage bottlenecks easier to identify, which improves both cost and performance troubleshooting. Once you can see where the bottleneck sits, you can fix it instead of overprovisioning to cover it up.
Use Cheaper Capacity More Reliably with Azure Compute Spot Plus
Interruption-tolerant workloads can benefit from a better Spot option. Azure Compute Spot plus is a new enhanced Spot tier that offers greater runtime consistency and earlier eviction signals.
That combination can help these workloads use cheaper capacity more effectively. Earlier warning before eviction gives your systems more time to respond, and greater consistency means fewer surprises.
Keep Throughput in Check with Cosmos DB Throughput Buckets
One busy workload shouldn’t be able to eat up a whole shared container. Throughput buckets in Azure Cosmos DB are now generally available, letting teams limit how much resource consumption a workload can use within a shared container.
This supports workload isolation and cost control by preventing one workload from consuming excessive throughput. It’s a handy way to keep shared environments predictable.
Save on Shared Storage with Azure Files NFS Shares on HDD
Linux workloads that need shared storage now have a lower-cost option. Azure Files now supports NFS shares on HDD in public preview, providing a managed storage option for Linux workloads at a lower cost.
This extends your cost-effective storage choices for shared datasets and backups. Since it’s currently in public preview, it’s worth testing before you rely on it for production work.
Match Compute to Demand with Azure Compute Fleet and IBM Spectrum LSF
HPC workloads can waste capacity when supply doesn’t match demand. Azure Compute Fleet now integrates with IBM Spectrum LSF to provision and release capacity based on demand.
Matching compute supply more closely to what your workloads actually need can improve cost efficiency. Capacity that scales with demand means you’re not paying for machines sitting idle between jobs.
Run Services on Fewer Cores with Cobalt 200 Virtual Machine Families
Newer hardware can mean doing more with less. Azure’s Cobalt 200 Arm-based Dplsv7, Dpsv7, Epsv7, Mpsv4, and Lpsv5-series Virtual Machines are now generally available, delivering major performance gains over Cobalt 100.
These new instance types can improve performance per core and reduce infrastructure cost for eligible workloads. That makes them a strong candidate when you’re reviewing compute choices.
Get Up to 50% Better Performance with Cobalt 200 VMs
Microsoft is pitching Cobalt 200 as a real efficiency upgrade. In a blog post about Azure Cobalt 200 VMs, the company says these VMs are now generally available and deliver up to 50% better performance than Cobalt 100.
Microsoft says the new generation can help customers run services with fewer cores. Fewer cores for the same work is a clear path to better efficiency and cost optimization.
Cut Database Costs with PostgreSQL’s Diskless v5 Tier
Not every database needs local disk, and paying for it when you don’t can add up. Azure Database for PostgreSQL Flexible Server now has v5 diskless compute options, which remove the local temporary disk from selected SKUs.
This can be relevant to cost optimization and architecture selection for PostgreSQL workloads that don’t need local disk. Choosing the right option for each workload keeps your spending aligned with what the database actually uses.
Run Managed PostgreSQL on Cobalt 200 for Better Performance per Dollar
More compute options mean more ways to fit your workload. Azure Database for PostgreSQL Flexible Server now supports Cobalt 200 ARM-based compute, broadening the instance portfolio for managed PostgreSQL.
This update may improve performance-per-dollar and workload efficiency. Having another instance option gives you more room to find the right balance between speed and cost.
Scale PostgreSQL Higher with v6 Compute and Temp Disk
Some PostgreSQL workloads need more power and local temporary storage. Azure Database for PostgreSQL Flexible Server now offers v6 compute tiers with local temporary disk, adding higher performance and scale options.
This instance update may improve workload efficiency and unit economics for PostgreSQL deployments. It’s a useful choice when your workload benefits from that local disk.
