High Scalability in Azure is the ability to increase your capacity based on the increasing demand for traffic, memory, and or computing power. This is especially true for high-growth companies where what worked for the infrastructure last month, very well may not be enough horsepower to meet demand this month. What does this mean? The company will need to Scale Up by getting bigger and better servers. Scaling up is not the only solution with Azure High Scalability so let’s look at the available options.
Azure Vertical Scaling
Vertical Scaling is also known as Scaling Up. Vertical Scaling means that to meet the demand placed on the cloud service, one upgrades to a bigger sized server. This is likely the most obvious type of scaling most people would think of. So once you upgrade to a bigger server, you have bigger hard drives, a faster central processing unit, and overall more horsepower.
Azure Horizontal Scaling
A second type of scaling is known as Horizontal Scaling. This method of scaling is also known as Scaling Out. When Scaling Out, you are not moving to bigger servers, you are simply adding more servers of the same size.
More About Vertical Scaling
The idea of vertical scaling can be thought of as adding or reducing the size of a virtual machine running in the cloud. If you were adding or subtracting entire virtual machines that would be horizontal scaling as we saw above. Vertical scaling keeps a single Virtual Machine up and running however the size and resource parameters of that VM may change. Vertical scaling is also different from reprovisioning, which is when a completely new VM is required for a fresh provision, whereas scaling only adjusts the existing configuration.
Vertical Scaling is a fantastic method to control costs. A public-facing web application that experiences slow usage on the weekends can scale down its resources during slower usage times to save on costs. Conversely, if that same application experiences a burst of demand during a holiday period or some other stimulus, then scaling up for a short time is the most cost effective.
More About Horizontal Scaling
Horizontal Scaling comes into play when you no longer have a bigger and better-sized server to upgrade to in Vertical Scaling. You might run into a limit with regard to how much memory, processing power, and disk size a given computer can support. This is when you turn to Horizontal Scaling. So when you’ve already upgraded to the biggest server available and you still need more power, you just start adding more servers.
Azure Scale Terms
Other terms that you may run across when talking about Azure High Scalability are Virtual Machine Scale Sets, Virtual Machine Configuration Management, Virtual Machine Storage Options, and Virtual Machine monitoring. A cloud engineer should be able to Scale VMs using scale sets and understand the difference between vertical and horizontal scaling, know the tools needed for VM configuration management, understand Powershell DSC and VM extensions, as well as deploying different types of storage options for Virtual Machines.
Learn More About Azure High Scalability
- Azure Architecture Framework Scalability Overview (docs.microsoft.com)
- Mysql On Azure Performance Benchmark Scalegrid Vs Azure Data (highscalability.com)
- Microsoft Azure Offers Scalability As A Service For High Volume Apps (computerworld.com)
- Azure Scalability Features And Functionality (vembu.com)
- Cloud Computing Scalability What Is It And Why Its Important (criticalcase.com)
- Stonefly.com Blog What Is Scalability In The Cloud (stonefly.com)
- Acloudguru.com Blog Engineering Scalability Cloud Computing (acloudguru.com)
- Cloud Elasticity Vs Cloud Scalability (blog.turbonomic.com)
- Cloud Scalability (vmware.com)
- Digital Cloud Cloud For High Scalability (blog.aspiresys.com)
- What Is Scalability In The Cloud (stonefly.com)
- High Availability Scalability Elasticity Agility Fault Tolerance And Disaster Recovery (coderjony.com)
- What Is High Scalability (quora.com)
- Azure Fundamentals Identify The Benefits And Considerations Of Using Cloud Services (manikanta.hashnode.dev)