The Latest Azure Updates: MFA; Cloud trends; Linux; Data storage; Form Recognizer

Compared to other times of year, July got off to a quieter start with fewer announcements from Microsoft due to the US Independence Day holiday. For security purposes, Microsoft is making multi-factor authentication mandatory for all cloud solution providers, although the change won't take effect completely for another few months.

A Spiceworks study, "Public Cloud Trends in 2019 and Beyond," indicated that service uptime, security and service compatibility are the key factors for a majority of companies in choosing a cloud service. Data suggested a slight lead in popularity for Azure. Many respondents indicated lingering doubts about the security of the cloud.

According to some reports, Linux is now the most widely used operating system on Azure, after increasing from 25 percent in 2015 and hitting 40 percent in 2017. Even at the end of 2018, 50 percent of VMs were Linux-based. Many native workloads like Azure SDN run on Linux and it is preferred by many enterprise developers.

The Azure team upgraded the Teradata connector for Azure Data Factory, adding a built-in driver to cut down on manual installation time, the ability to copy activity and resolutions to connection or query timeout. Users are recommended to adopt the upgraded connector, although old connectors will still work. Additionally, parameter support for Data Factory is being joined with the Mapping Data Flows public preview. This will support building  configurable data transformation logic. Pipeline expressions, Data Flow expressions or static values can be used to set values for passing data through a Data Factory pipeline.

During sprint 154 for Azure DevOps, the Azure team released Azure Pipelines for Jira. The link to the Atlassian-product adds Jira issues as work items, with new tracking and search functionalities.

Through the Retail feed, users will be able to access Azure Sphere 19.06 in preview. The key updates in this round are patching three common vulnerabilities. The Azure team also added Azure HDInsight support for Apache Kafka 2.1 and Spark 2.4.

Non-Azure group targeting for Azure Update Management will add non-Azure machines to update deployment cycles so that they get patched on time. Customers could use a Log Analytics saved search to group machines, with Azure VMs automatically filtered out of the deployment queue.

On the database front, Microsoft is adding up to 16TB of storage, equivalent to 20,000 IOPS in both Azure Database for MySQL and Azure Database for PostgreSQL. The release triples existing IOPS availability and increases storage by five-fold.  In a similar move, a higher scale standard tier is being rolled out for Azure Files, boosting from 5 TiB to 100 TiB and as much as 10,000 IOPS.

Azure Data Box Heavy is generally available, allowing users to copy data from standard object protocols and files, which is then secured with AES 256-bit encryption. The offering encompasses a petabyte of raw capacity and several 40 Gbps connectors. For now, Data Box Heavy is only available in the US and Europe, although Data Box is in Japan, Canada, and Australia while Data Box Disk includes Korea and Southeast Asia.

Nikhil Kamath, program manager for Azure, touted the new Azure AI Form Recognizer which offers machine learning to pull text, key-value pairs and tables from documents. The optical character recognition system is aimed at speeding up expense reporting. In fact, Microsoft is already using it internally, with MSExpense. The system sorts receipts into low, medium and high risk categories, spotting possible anomalies.