IBM is continuing to embrace Kubernetes as a core cloud service by making the container orchestration platform available across multiple availability zones. That extension is part of an expansion of the IBM Cloud service into 18 additional availability zones across North America, Europe and Asia Pacific.
Jason McGee, vice president and CTO for IBM Cloud Platform, says that, given the complexity associated with managing the underlying infrastructure needed to build, deploy and manage microservices based on containers, there’s been a significant increase in interest in managed services. Most IT organizations still don’t have enough internal expertise to manage these applications, he notes.
Being able to deploy instances of Kubernetes across multiple availability zones within a distributed public cloud is yet another example in which a managed service pulls together all the container infrastructure elements needed to drive modern application development, McGee says. Otherwise, IT organizations are forced to divert resources to self-assemble various infrastructure components, which can increase costs and compromise agility, notes McGee.
In fact, McGee says that as cloud computing continues to evolve, the key differentiator is likely to have more to do with the experience provided to developers rather than any advantage provided by a specific technology. All the cloud service providers essentially have access to the same underlying open source technologies, he says, so in some instances, it simply makes more sense for a cloud service provider to partner with a third-party vendors such as Sysdig to provide specific capabilities such as container monitoring.
Over time, McGee notes the quality of an IT experience enabled by containers will be much higher because containers expose a much richer set of metadata. IT management frameworks have yet to fully exploit those capabilities, but it’s only a matter of time before IT organizations begin to benefit from more advanced IT management capabilities.
McGee also notes that IBM is starting to see more containers being deployed on bare-metal servers rather than virtual machines, in part to improve application performance and reduce the overhead associated with supporting guest operating systems.
IBM and other cloud service providers are betting the transition to microservices based on containers will accelerate a trend toward outsourcing the management of IT infrastructure. Organizations are under more pressure than ever to build applications faster, which results in many of them seeking to reduce the number of personnel dedicated to managing IT infrastructure in favor of hiring more developers.
Another factor driving applications into the cloud is the need to infuse them with machine and deep learning algorithms to embed artificial intelligence capabilities. Those applications typically require access to massive amounts of data, which are cost-effectively stored in a public cloud.
It’s not apparent to what degree IT organizations may be willing to rely on external service providers to manage IT infrastructure. There’s no doubt many of them will be employing cloud infrastructures. But who actually gets to manage those environments is going to a much tougher conversation, be given all the stakeholders involved.