When you're phasing advanced analytics, machine learning, and artificial intelligence into your infrastructure, traditional configurations aren't necessarily up to the task. Applications related to AI can accumulate a large volume of data based on I/O requirements. You'll need to ensure that these attributes are part of your setup:

Microsoft Cloud Services, for example, utilize commodity hardware and scale virtually infinitely to handle AI workloads. By using commodity hardware, Microsoft is able to provide storage services over standard protocols like iSCSI, NFS, SMB, CIFS, etc. and more advanced features.

Commodity hardware is a growing trend when designing a system to manage large volumes of data. Big data infrastructures normally utilize commodity hardware to distribute terabytes of data throughout the network. Microsoft Azure storage provides an affordable solution for AI by supporting structured and unstructured data in a highly scalable platform. You also have the ability to implement storage-related automation routines using RESTful APIs.

AI applications and the loads that they can put on a system's resources will vary in volume and complexity. To support deep learning, problem-solving, reasoning, and learning, these applications need to the ability to analyze large volumes of data in various formats, then compile data and produce results in an optimal amount of time.

Here are two examples of how Microsoft is designing applications to build the infrastructure of AI:

Azure Machine Learning is a methodology designed by Microsoft to incorporate predictive analytics into applications. The Machine Learning Studio allows developers to use known programming techniques to design models for AI.

Microsoft Azure Cognitive Services is another offering that combines user experiences with machine-based intelligence. Cognitive Services empowers the user by helping deploy applications across platforms easily.

Cognitive Services Development covers the following areas:

Microsoft recently announced Azure Batch AI Training, as well. This is a space segmented within the Azure Cloud that developers can use to work on AI applications without concern of compromising infrastructure resources. By doing this type of modeling, developers will be able to determine what is needed for their infrastructure before the application is deployed.

Each of these tools can utilize the Azure Cloud for deployment, as well, so you can reap the benefits of test/dev and production in Azure.

AI infrastructure dependencies vary, and smart design is critical. Microsoft Azure, along with Microsoft's development and training tools, put a custom solution within reach.

Go here to see the original:

Exploring the infrastructure needs of AI - ZDNet