Windows Azure Innovation: Caching Solution Preview

Caching is a hot topic both in http://www.aliyun.com/zixun/aggregation/13423.html "> Cloud computing platforms or in building a high-performance and highly scalable system in a non cloud environment." Windows Azure AppFabric Caching was launched in March 2011. It provides a as and distributed caching service on the cloud. In the updated version of June 2012, the cache team announced windowsazure a new Caching solution called Windows Azure Caching (preview). And the original Windows Azure AppFabric Caching was renamed Windows Azure Shared Caching.

What is caching (preview)

If you've been using shared caching you should know that it's made up of a lot of cache servers. Before using, you first need to create a cache account from the Developer portal and specify the size you want to use, that is, the amount of memory that can be used to store the data that needs to be cached. You can then add, retrieve, and remove the data through the code via the cache URL.

Shared caching is a multi-tenant system that hosts all cached elements in all users. So you don't know which server your data is stored in.

This caching model works well and can handle most situations. But it has some problems. The first problem is the performance problem. Because shared caching is a multi-tenant system, it means that all cache operations need to go through a shared caching gateway and then route to the server that stores the data you need. Even though there are some caches in shared caching systems, it will take some time from cloud services to cache services.

The second problem is that shared caching services work for developers as a black box. We only know the cache endpoint. That's enough for some people because they don't care about anything. But if you want to get more information and more control in the shared caching, that's impossible.

The final issue is price and cost-effectiveness. You need to pay for the cache you request each month. But when we host a web role or worker role, we rarely use all of the memory and CPU in the virtual machine (service instance). If you use shared caching we need to pay for the cache service but also waste some local memory and CPU.

Because of the above problems, Microsoft has provided us with a new caching mode, namely caching (preview). Caching (preview) leverages the memory and CPU of our cloud computing services (Web role and worker function) as cache clusters, instead of having separate cache services. Because caching (preview) is running on a virtual machine, hosting our cloud computing application or near it. No gateways and routes are required because it is located in the same data center and the same rack, rather than the shared
Caching provides higher performance.

Caching (preview) works side by side with our applications, initializes and runs as a Windows service in a virtual machine, invoked by our startup task in role, we are able to get more information and control them.

Because caching (preview) leverages the memory and CPU of our existing cloud computing services, it is free. We only need to pay for the calculation. Resources on each machine can be used more efficiently.

Enable caching (preview)

Enabling caching (previews) in cloud computing services is straightforward. Let's start with Visual Studio to create a new Windows Azure Cloud Project and then add a asp.net Web role. Then open the role setting interface and select Caching page. Here we can enable and configure caching (preview) on a role.

To enable Caching (preview), open the Enable Caching (Preview release) check box. Then we need to specify the caching cluster pattern that we want to use.

There are two types of caching modes, co-located and dedicate. Co-located mode means that we use memory that runs the Cloud computing Service (Web role or worker role) instance. By using this pattern we have to specify how much of the memory will be used as cache. The default value is 30%. So make sure it doesn't affect role business execution.