One of the factors that you need to consider when moving workloads into Azure is the VM size. Not all sizes fit all workloads. If you look at this chart, you'll see the different types of VM sizes in a tight. So we have the general purpose, they provide balanced CPU to memory ratio. This type of VM is ideal for testing and determining small to medium databases and low to medium traffic Web servers. Av2 series VMs can be deployed on variety of hardware and types of processors. Compute optimized VM sizes have a CPU to memory ratio, and these sizes are good in medium traffic Web servers, network appliances, batch processes, and application servers. The Fv2 series are based on Intel Xeon platinum 8168 processors. This is a best value in a price performance and Azure portal based on Azure Compute Units, which are ACUs. Remember the optimized VM sizes offer a high memory to CPU ratio that are great for relational database servers, medium and large caches and in memory analytics. Dv2 and DSv2 series are ideal for applications demand faster, these CPUs better storage performance and have higher memory demands. They offer powerful combinations for many of your enterprise grade applications. Storage optimized VM series offers high disk throughput in I/O and are ideal for big data, SQL, NoSQL database, database warehouses, and large transactional databases and this includes Cassandra, MongoDB, Cloudera, and also Redis. CPU may optimize VMs series or specialized VMs available with single or multiple NVIDIA GPUs. These sizes are designed for compute insensitive and visualization workloads. Then you have the H series VMs, these are designed for delivering leadership class performance. They have API, MPI or scalability, cost effective for a variety of real-world HPC workloads. These are used for Azure batch and Microsoft HPC pack. They offer the fastest VMs available and they also offer high throughput using RDMA Remote Direct Memory Access capableness and for ban cards. Azure Managed Disks or block-level storage volumes are managed by Azure in use with Azure VMs. Now they're like the physical disks that we have on-premises servers but virtualized. With managed disk, all you have to do is specify the disk size, this type and provision the disk. When you provision the disk, Azure handles the rest. The available types of disks are ultra disks, you have premium SSDs, standard SSDs and also we have standard hard-disk drives. Now, what are the benefits of using managed disks? Well, managed disks are designed for 5,9 availability, so 99.999 percent. They achieve this by providing you with three replicas of your data, allowing you for high durability. So in the event that one of the two replicas experiences an issue, the remaining replica helps ensure persistence over data and high tolerance against failure. This has helped Azure that are consistently deliver enterprise grade durability for Azure disk with an industry leading zero percent annualized failure rate. Using managed disk, you can create up to 50,000 VM disk of a type and a subscription per region. Now this is allowing you to create thousands of VMs on a single subscription. This also increases the scalability of the VM scale sets by allowing up to 1000 VMs in a scale set using a marketplace images. Managed disks are also integrated with availability sets. They'll insure all the VMs and availability centers are sufficiently isolated from each other, this avoids the single point of failure. If the disks are automatically deployed to separate storage scale units, they're called stamps. Now disks protects VMs when it's stamp fails only the VMs or an instance where the disc and the stamp fail, the other VMs will continue to function. Managed disks also support availability zones. We learned about these in previous lessons. This is a high availability offering that protects against failures of a data center in a region. Each region is made up of one or more data centers with the autonomous power. To ensure resiliency there's a minimum of three separate zones in all enabled regions and then what this does is provides with 99.99 percent VM SLA. To protect against region disasters, we also have Azure backup and this can be used to create a backup job, time-based backups, backup retention policies. This allows you to perform easy VM restorations. Azure backup can also support up to four terabyte disks. We have our back integration and this assign granular permissions for a managed disk to one or more of the users. They can control the actions that a person needs so they can do whatever they need for their job. You can also upload your own VHDs from on-premises environment. Right now, it needs to be a VHD format because VHDXs are not in preview right now. Managed disks support Bitlocker and also DM-CRISP for encryption of the disc. Also the disks are protected at rest using server-side encryption. Using managed disks also allows you to go over the 20,000 IOPS limit that's set with unmanaged disks,
