Microsoft is betting that good security support will be key to keeping its enterprise customers from straying to rivals.
At the kickoff of the company's Ignite conference for IT professionals, Microsoft executives unveiled a number of advanced security services, and took jabs at competitor Google for not being as mindful of security.
"Google takes no responsibility to update their customers' devices, leaving end-users and businesses increasingly exposed every day they use their Android devices," said Terry Myerson, Microsoft's executive vice president of operating systems. "Google just ships a big pile of code, and then leaves you exposed with no commitments."
Microsoft is refining how it distributes security updates, starting with the upcoming releases of Windows 10 and Office 2016, Myerson said.
Right now, Microsoft sends out updates for Windows on the second Tuesday of each month, a routine called Patch Tuesday that touches over 858 million computers. Many consumer machines are configured to download and apply the patches automatically, and many enterprises control the update process using software such as Microsoft's System Center Configuration Manager.
With Windows 10, consumers can now get security updates as they are released from Microsoft, along with other updates and new features, resulting in a "steady stream of innovation every month," Myerson said.
Businesses will get a number of new options for handling security patches. They can opt into one of a set of "distribution rings." Some can choose to get their updates as soon as they arrive, or they can wait to see if any additional issues pop up with the patches, which has been a recurring problem with Microsoft patches of late.
Microsoft will also start offering the option to get only security updates, and not new feature updates, which can be handy for those mission-critical machines that organizations need to keep precisely configured.
Administrators can now specify when they want patches to be applied, so the patches aren't deployed during a busy time, or at night when some computers may be shut off. For those organisations with limited bandwidth, Windows 10 computers can share the updates with one another in a peer-to-peer network, rather than downloading patches for each machine.
In addition to updating the patching process, Microsoft also unveiled a number of new services to help better secure systems against data leakage and compromised identities.
A feature in Windows 10, Device Guard, limits the computer to running only those applications that have already been approved to run on that machine. This safeguard could prevent the user from unwittingly installing malware, thinking it came from an approved source.
Windows 10 is being outfitted with software to control the unauthorised copying of organisational data. A user can cut and paste information from an e-mail in the Outlook mail client into another organization-approved application, such as Word. But the user can not paste the data into an unapproved application, such as into a Twitter account.
Users can override the block, but only after they click through a dialogue box; Windows 10 will log all these unapproved copying actions.
A service called Azure Rights Management Services which can further guard against corporate data leakage, was also made generally available. This service provides the ability to protect access to files even after they leave the individual computer.
It introduces a concept called self-protecting files, or files that retain information about how they can be accessed. Before sending a file to someone, a user can specify what permissions that recipient has with that file. The sender can specify, for instance, if that file can be forwarded to additional parties. The sender can revoke access to the document even after it is sent out. "The file becomes self-protecting," said Brad Anderson, Microsoft corporate vice president for enterprise client and mobility.
The administrator can also get statistics on the usage of the file, such as how many people read the document, how many of those were authorized, and the names of those who tried to open the document but who weren't authorised to do so.
Those using the company's Azure-based Active Directory service can take advantage of a new service, now in preview mode, that can identify anomalous sign-ins, or those attempted log-ins to the organisation's system that probably don't originate from the employee.
If an employee signs into the network from Chicago at noon, and then again at 12.30pm from North Korea, it is probably a safe bet that the second sign-in is not a legitimate one, explained Anderson. The service uses machine learning to correlate the immense number of log files in order to find these matches.
Another machine-learning-based service, called Microsoft Advanced Threat Analytics, can provide organizations with an easy way to identify network intruders. The offering, available as a preview starting Monday, is based on technology developed by Israeli startup Aorato, which Microsoft acquired in November.
Microsoft Advanced Threat Analytics can provide a timeline view of a series of activities that make up a single attack as they unfold across different system resources. It can show, for instance, someone using a brute-force attack to compromise a user account, and when that account is breached, the service can then follow subsequent actions on different machines, such as accessing content.