Containers want to get into the enterprise, and the feeling is mutual on the enterprise side. But somewhere between the wanting and the having are a number of hitches preventing this desirable union.
This is nothing that can’t be overcome by continued development, of course, but it does mean that containers won’t emerge as a true enterprise solution until they prove themselves in production environments.
According to Lars Herrmann, GM of Integrated Solutions at Red Hat, containers fail to meet enterprise requirements in five key areas at the moment. Speaking to CIO.com, he says containers still lack the ability to integrate smoothly into legacy environments, which is a red flag for organizations trying to remove data silos elsewhere in the stack. As well, issues like poor VM management and orchestration only add to the complexity that containers bring to the data environment, while poor native scalability makes them unsuitable for the kinds of microservice-based workloads they are supposed to support.
But things are not all bad, says eWeek’s Darryl K. Taft, particularly now that leading data frameworks and operating environments are taking steps to incorporate containers on their end. A case in point is OpenStack, which in recent months has launched a number of projects intended to address the very concerns that Herrmann and others are raising. To combat VM sprawl, for example, OpenStack is working to incorporate third-party tools like Docker Swarm, Kubernetes and Mesos into a single Container Orchestration Engine that enables automated provisioning across multiple compute instances. At the same time, the organization is incorporating ways to group VMs or bare-metal instances in isolation to address security concerns in multi-tenant environments.
And of course, container developers know they will never conquer the enterprise until they figure out how to ingratiate themselves with leading enterprise platforms. This is why Docker, for one, is funneling development dollars toward Windows Server. According to InfoWorld, the latest beta of the Docker Engine for WS 2016 incorporates tools like push-pull commands for Windows Images on the Docker Hub dev/test module, making it easier to tie containers to Windows-specific images. In addition, the company is working on ways to encapsulate multiple binary images into a single container, which should allow images to cut across multiplatform builds on both the OS and the CPU level.
All the while, various professional groups are looking to add container functionality in pursuit of their broader architectural goals. The Cloud Native Computing Foundation, for instance, is looking to adopt containers as a means to package cloud-native applications into a scalable development and operations framework, says EnterpriseTech’s George Leopold. This should enable dynamic configuration and deployment of cloud-facing services and microservices, even at Google-scale volumes of billions of containers per week. As CNCF head Chris Aniszczyk notes, this is similar to the Sematech movement in the 2000s that propelled semiconductor development to new levels.
A container-based data environment should prove invaluable as emerging high-volume, multi-threaded applications like Big Data and the IoT start to work their wonders on the business process. The challenges in pushing the technology into real-world production environments are no greater than those that confronted initial virtualization and cloud development, but it will take creativity and ingenuity to carve out architectures that leverage containers that support enhanced productivity without simply pushing complexity somewhere else in the data stack.
Development of container technology is crucial to meet these twin goals, but so is the reconfiguration of the broader data ecosystem.
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.