Managed Azure services hotspots span optimization to AI

Demand for cost optimization, microservices

While customers express interest in cloud-based AI, service providers are finding work in somewhat more prosaic areas: managing cloud costs, retooling applications and keeping cloud services as resilient as possible.

One hot area of interest is infrastructure as a service for storage and data protection activities, O'Konski said. These trends, he added, give rise to the need for services such as platform as a service and a lot of managed services around cost optimization and cost reporting.

Thomas Burns

Thomas Burns, CTO at Green House Data, an MSP based in Cheyenne, Wyo., has also noticed the demand among his customers for cost optimization tools and services, which is a lucrative area of revenue generation for MSPs, CSPs and cloud consultants.

"Customers that already have a deployment in Azure, have seen significant costs sprawl over the time of their engagement, and they are looking for people to help them understand what that infrastructure looks like, why it is growing at the rate it is growing and how to help make optimization choices for future deployments however they streamline that infrastructure so they can control costs," Burns said.

Customers that already have a deployment in Azure, have seen significant costs sprawl over the time of their engagement, and they are looking for people to help them understand what that infrastructure looks like.

Helping customers design new or rationalize existing applications for Azure platforms is another in-demand service. Burns said many midmarket customers are having trouble moving applications built on old technology, such as IBM mainframes, to the cloud. Customers, he noted, are demanding that applications residing on older technology become cloud-first applications. And they want to take advantage of PaaS, microservices and other more advanced software development methodologies, using tools like Azure DevOps to get there, he said.

Gordon McKenna, CTO of public cloud at cloud service provider Ensono, said many customers are transforming from monolithic applications running on monolithic virtual machines to microservices and serverless computing. Such customers seek to take advantage of off-the-shelf application development platforms "that give a head start in developing new types of services that are more efficient, more feature rich and more user-friendly than ever before."

McKenna said he has seen increased demand for container technology such as Azure Kubernetes Service, as well as database technologies such as Azure Cosmos DB and Azure Database for PostgreSQL as alternatives to platforms like Oracle (see "Areas of demand for Microsoft cloud services").

Areas of demand for Microsoft cloud services

Chris Smith

Service providers cite several areas where they are seeing demand for Microsoft cloud services. Here are a few hotspots in addition to cost optimization, application development and high availability.

Microsoft cloud apps. "There is a very large demand for utilizing the Microsoft product suite, i.e. Windows 10, Microsoft 365 and the whole Microsoft cloud platform," said Chris Smith, vice president of cloud architecture at Unitas Global, a Los Angeles-based company that provides managed hybrid cloud solutions. "I would say that's probably No. 1, and then the other area of high demand is around analytics and artificial intelligence."

AI. "There is always a continuing interest in AI technologies, like Azure Machine Learning and Microsoft Cognitive Toolkit, as people look to make much smarter applications," McKenna said.

IoT management. Customers "need things like internet of things management at scale available to manage hundreds of thousands or tens of thousands of devices at scale on the back-end infrastructure," O'Konski said.

Accelerating cloud adoption. Managed Azure services that support cloud adoption stand out as the most profitable, McKenna suggested. "Any services where we can provide a professional services wrap to help the customer accelerate adoption -- Azure Service Environment [and] Azure Service Fabric and Functions are a few examples."

Architecting against failure

Limiting downtime is another business opportunity. While Microsoft, through efforts such as Project Tardigrade, has been taking steps to reduce Azure outages, some managed Azure services providers focus on boosting clients' uptime.

Gordon McKenna

Ensono, for example, architects for high availability whenever it designs and builds solutions on the Azure platform, McKenna said.

"What I mean by this is we never leave our customers' eggs in one basket," he said. "We never deploy a customer's infrastructure and applications to a single Azure Region. In fact, we ... deploy a minimum of two regions, preferably three within a country, and, where possible, we recommend these regions be duplicated in different geographic locations," McKenna said.

"We have customers in the South Central U.S. region, but we would typically have a twin region in say, Virginia," he continued. "Each region replicates its data to the secondary region, and in the event of an outage at the region level, we quickly failover the secondary region or even third."