Definition of Azure
Microsoft Azure is an ever-expanding set of cloud services to help your organisation meet your business challenges. It’s the freedom to build, manage and deploy applications on a massive, global network using your favourite tools and frameworks.
Azure is a cloud service provider. Other cloud vendors to consider are:
There are more than 600 products available from Microsoft Azure, across 22 categories. These products include:
AI and machine learning
- Cognitive Services are a suite of AI capabilities that can be integrated into your apps through API calls. This allows developers without any AI or machine learning experience to build intelligent apps
- Machine Learning services allow developers to build, train and deploy machine learning models. The service supports a range of open source tools and framworks, including PyTorch, TensorFlow, and scikit-learn
- Azure Bot Service supports development of intelligent, enterprise-grade bots such as QnA bots and virtual assistants. The service integrates with Cognitive Services to add powerful AI capabilities, including natural language processing (NLP) and computer vision
- Azure Cognitive Search is a cloud search service with built-in AI capabilities such as NLP and computer vision. The service makes it easier to work with unstructured data (e.g. text and images) by organising and structuring it to make it searchable
- Azure Databricks is a collaborative data science environment based on Apache Spark, that is optimised to work with other Azure services. It provides a platform in which teams can collaborate to gather, analyse and model data. Interaction with data is possible through interactive notebooks that support a range of languages - including Python, R, Scala, and SQL - and deep learning frameworks
- Azure Analysis Service is a scalable and secure data analytics platform that can combine data from multiple sources to produce shareable insights that integrate with Power BI for visualisation
- Azure Synapse Analytics is a highly scalable solution to analyse large volumes of data, especially across distributed data sources like data lakes
Data storage and management
- Azure SQL Database, Azure Database for PostgreSQL, Azure Database for MySQL, and Azure Database for MariaDB are relational databases to support analytics or operational activities, saving the need to manage the underlying infrastructure or tasks like backups
- Azure Cosmos DB is Microsoft’s globally distributed, multi-model database service, which provides highly scalable storage for data that can be accessed using a variety of APIs including SQL, MongoDB, Cassandra, Tables, and Gremlin
- Table Storage is a high-latency NoSQL datastore for structured data
- Data Factory is a service that allows you to integrate data from several sources for transformation and analytics
- Azure Data Lake Storage is a highly scalable and cost-effective data lake solution for big data analytics. It is a centralised repository for vast quantities of structured and unstructured data
Security and Compliance
Microsoft are commited to security, privacy and compliance across their Azure products. The Azure Trust Center provides details of Microsoft’s foundational principles of trust: security, privacy, compliance and transparency.
Microsoft Azure features multi-layered, built-in security controls that can be managed via intuitive controls. Intelligent features helps to identify and protect against potential threats. Security products from Azure include Security Center and Key Vault.
Read more about Azure security here.
The Microsoft Privacy Standard enforces strict protections on Microsoft business processes. Using Azure products, you have a clear view of where your data is stored, how it is secured, and who can access it. You can choose the geographical location of your data centre when you set up your Azure products.
Read more about Azure data privacy here.
Microsoft Azure holds more than 90 compliance certifications, and features products and tools that help you to ensure compliance. Tools such as Compliance Manager allow you to track your compliance with GDPR and other privacy regulations. Azure Security and Compliance Blueprints can be used to create, deploy and update compliant environments for your data.
Read more about Azure compliance here.
Microsoft Azure has a competitive, scalable pricing structure with several options to support businesses as they grow. From the dashboard you can access at-a-glance cost overviews to manage spending, and the Azure pricing calculator helps you to budget and estimate costs of using Azure services.
Creating an account on Microsoft Azure provides:
- 12 months of free access to select services, including data storage and some AI services
- £150 credit to use on any Azure service for 30 days
- access to more than 25 always-free services, including Security Center and Search
Beyond the free services, Microsoft Azure offers the choice between a pay-as-you-go (PAYG) model and reserved pricing. With either option, you are billed monthly for the services you use or reserve.
Pay as you go
A great option for scalability, Azure products are charged by time, events and storage space.
For example, Azure Machine Learning services are charged by hour of use by the virtual machine, Azure Bot Service is charged per thousand messages, and pricing for Azure SQL database is calculated based on hours of available access and storage space, in incremenets of 32GB.
If you are confident of the computing power and storage space you will need for your application in advance, the reserved pricing model can save you 20-30% of costs. In this model, you reserve the time and storage space required for the next year or two years, and pay for it monthly.
Azure Dev/Test Pricing
Microsoft Azure offers discounted rates to support ongoing development and testing. The discounts are available for the following services:
- Windows Virtual Machines*
- BizTalk Virtual Machines*
- Azure SQL Database*
- SQL Server Virtual Machines
- Logic Apps Enterprise Connection
- App Service Instances
- Cloud Service Instances
- HD Insight Instances
*These services also remain eligible for reserved instance pricing in addition to dev/test discounts.
You can find out more about Dev/Text Pricing here.
Purchase Azure services through a Cloud Solution Provider (CSP)
As well as having the option to buy directly from Microsoft Azure, you can go through a CSP who will provision, manage and support your subscriptions. This is a good option if you already purchase Microsoft solutions through a CSP, as they will consolidate your bills and ensure you are not using duplicate services. A CSP provides local support and expertise to help you get started with cloud services and manage costs, and will often offer bundle deals that can save you money. You can find a CSP here.
All Azure accounts come with free customer service, documentation, and tutorials to help you get started. Additional technical support becomes available when you purchase a support plan, starting at just over £20 per month. Learn more about the support plans here.
Implementing data science and AI in your business requires vast data storage capabilities and expensive infrastructure. A cloud computing solution such as Microsoft Azure provides pay-as-you-go access to these systems and allows you to scale the costs as you grow, avoiding expensive set-up costs.
Microsoft Azure helps businesses:
adopt AI without expensive set-up costs
develop high-quality applications that collect and leverage data to maximise revenue
Business function leader view
Microsoft Azure helps teams adopt AI solutions that can drive growth and improve performance, as well as providing solutions to host applications and build improved functionality that will drive sales. Microsoft Azure is a cloud computing platform, which allows pay-as-you-go access to storage and computing services without expensive infrastructure set-up fees.
Signs your department should invest in this are:
you are developing an application and need a hosting solution that is cheap to implement and will scale as you grow
you are planning to adopt AI in your team and need access to data storage and machine learning tools at a price that will scale with your needs
KPIs you should consider measuring for this are:
improved sales when AI features are implemented
savings on infrastructure investment
improved efficiency of resource management
improved product performance