Virtual Machines (VMs) in Azure come in predefined sizes that are called families or series. An individual VM is often referred to as an instance. Different VM families are designed for common use-cases and are comprised of certain amounts of CPU cores and GB of RAM. It’s not possible to arbitrarily mix and match CPU cores and GB of RAM as can be done with Hyper-V and VMware. Therefore, it is important to understand the specific VM Series you wish to deploy when thinking through the specific IT environment you are deploying in Microsoft Azure.
Here, we will focus on five of the most common Azure VM Series, how they compare to each other, and when to use each. You can also check out our YouTube video.
Azure VMs: A Series
Entry-level economical VMs for dev/test *
These are generally used in small deployments where cost consciousness outweighs performance. You should only use these in rare instance implementations with non-customer facing VMs.
Azure VMs: B Series
Economical burstable VMs *
B-series are economical virtual machines that provide a low-cost option for workloads that typically run at a low-to-moderate baseline CPU utilization, but sometimes need to burst to significantly higher CPU utilization when the demand rises.
We like B series for AD servers, servers that are 24×7 and non-customer facing servers.
When idle, the B series will bank credits and use those credits later when CPU utilization needs to burst past the baseline. When rebooted, B series will lose their credits which can take hours to rebuild without too much impact. When booted without any banked credits, the B series will only have access to allow for a small fraction of total CPU utilization until a bank of credits can be built up.
Azure VMs: D Series
General purpose compute *
D-series VMs feature fast CPUs and optimal CPU-to-memory configuration, making them suitable for most production workloads. DSv3-series instances carry more powerful CPUs and the same memory and disk configurations as the D-series. However, these CPU cores are hyper-threaded, meaning that a single physical CPU core is behind each two CPU cores in a Dsv3 VM.
Ds2-64 v3 instances (ie. Ds4v3 – 4 core x 16GB RAM) are the latest hyper-threaded generation of general-purpose instances and are based on the 2.4 GHz Intel Xeon® E5-2673 v3 (Haswell) processor or the latest 2.3 GHz Intel XEON ® E5-2673 v4 (Broadwell). They can achieve higher speeds with Intel Turbo Boost Technology 2.0. The Ds-series supports Standard and Premium SSD storage.
We like D series for performance as they pair well with premium SSD – a must for solutions or presentation layers that are customer-facing. There is not a lot of difference in price with v3 instances (as opposed to v2), which makes them ideal.
Example use-cases include many enterprise-grade applications, relational databases, in-memory caching, and analytics. The latest generations are ideal for applications that demand faster CPUs, better local disk performance or higher memories.
Azure VMs: E Series
Optimized for in-memory hyper-threaded applications*
The E-series family of Azure virtual machines are optimized for heavy in-memory applications such as SAP HANA. These VMs are set up with high memory-to-core ratios, which makes them well-suited for relational database servers, with medium to large caches, and in-memory analytics. The E-series VMs range from 2 to 64 vCPUs and 16-432 GiB RAM, respectively. The Es-series supports Azure Premium SSDs.
We like the E series for session hosts and collections where the line of business applications consumer a higher than normal memory footprint per user. Software that is optimized for multi-user session hosts can often consume large footprints of memory.
Example use cases include SAP HANA, SAP S/4 HANA, SQL Hekaton and other large in-memory business critical workloads.
Azure VMs: N Series
GPU enabled virtual machines *
The N-series is a family of Azure Virtual Machines with GPU capabilities. GPUs are ideal for compute and graphics-intensive workloads, helping customers to fuel innovation through scenarios like high-end remote visualization, deep learning, and predictive analytics.
The N-series has three different offerings aimed at specific workloads:
- The NC-series is focused on high-performance computing and machine learning workloads. The latest version—NCv3—features NVIDIA’s Tesla V100 GPU.
- The ND-series is focused on training and inference scenarios for deep learning. It uses the NVIDIA Tesla P40 GPUs. The latest version – NDv2 – features the NVIDIA Tesla V100 GPUs.
- The NV-series enables powerful remote visualization workloads and other graphics-intensive applications backed by the NVIDIA Tesla M60 GPU.
We like NV series for graphic needs related to engineering and 3D modeling. The footprint is quite large which makes them good shared machines. Assigning 2-4 users per N series as an RDS Session Host is an ideal for cost distribution and performance requirements. NV series will be the primary N series instance. Approach other N series with a very specific need or application requirement as they are not a good fit for GPU enabled virtual desktops.
Example use-cases include simulation, deep learning, graphics rendering, video editing, gaming, and remote visualization.