Unikernels: The Next Generation of Cloud Technology
Big Data and the Internet of Things are the two major disruptors in enterprise circles these days. Organizations need to ramp up infrastructure quickly in order to leverage the value that is locked away in massive data stores, but it needs to be done in a way that supports rapid, even real-time, results without crushing the IT budget.
This is why many enterprises are turning to the cloud: The infrastructure is already there, it’s cheap, and it is usually more state-of-the art than what is available in the on-premises data center. But with data points, cloud providers, local infrastructure and other elements distributed across wide geographic areas, how realistic is it to expect actionable data in time to take advantage of rapidly changing circumstances on the ground?
Real-time performance, of course, requires careful coordination up and down the stack, and this is a particular problem in the cloud given the variety of platforms that exist within most deployments. This is why vendors as diverse as Intel and SAP are working together on end-to-end solutions that deliver near-real-time performance at the least. The companies have integrated the Intel IoT Gateway with SAP SQL Anywhere to enable seamless data ingestion and transfer across SAP HANA deployments residing on distributed architectures. In this way, the enterprise gains a pre-integrated hardware/software solution that provides scale for security, connectivity and other functions, as well as data sync and unified management across SaaS, remote and mobile environments.
Meanwhile, IBM is addressing real-time performance on its Bluemix cloud with a pair of new services that enable seamless app updates and the ability to integrate date from a variety of sources for rapid analytics. The Active Deploy module allows developers to increase the development cycle for cloud-based apps with functions like deployment configuration control, mid-stream acceptance testing and fast rollback if things don’t work out. The Event Hub module coordinates multiple web events onto a centralized service repository where they can be accessed by the Bluemix Message Hub for integration with multiple services like Salesforce.
Real-time performance is also the secret sauce behind targeted cloud services like Oracle’s Marketing Cloud, which was recently upgraded with new tools to provide more immediate response to customer behavior and activities. The idea is to capture as much relevant data offline as possible so that when a customer does engage the enterprise’s digital infrastructure, actions and services can then be delivered immediately across multiple sales and marketing channels. To that end, tools like Offline Data Append, Match Multiplier and Business Units for Content Marketing help ensure that information is up-to-date to provide a higher level of orchestration that in turn drives a mutually beneficial customer experience.
Real-time performance in the cloud is unlikely to have a significant impact unless it has a strong mobile component, however. This is the aim behind Mitel’s new strategy to enable real-time communications across cloud, enterprise and mobile footprints. The plan is to devise a cloud-enabled, mobile-first infrastructure in which access, tools and information are lifted off of physical resources so they can be seamlessly and instantly transferred between enterprise and mobile networks. Key elements in the system include real-time, cloud-based voice, video and collaboration services, broad use of 4G services like VoWifi, plus new mobile apps and cloud services focused on business and process acceleration.
Clearly, the biggest inhibitor of real-time performance in the cloud is the distance between data and processing points. Long-haul networks are lightning fast these days but, as in the case of financial transactions, even a few seconds can end up costing millions.
It stands to reason, then, that much of the analytics capabilities of Big Data will be pushed to the edge, closer to where the real action is and less subject to the bottlenecks that arise within the data center, even in software-defined architectures.
But as long as Big Data infrastructure is geared for speed as well as capacity, even in the cloud, the enterprise should have little trouble generating returns on its investment.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata and Carpathia. Follow Art on Twitter @acole602.
