The growth of cloud computing is changing the nature of systems integration. “Integration” once referred to the process of combining system devices, software, and other components to create a superior solution. In the cloud era, however, integration calls for combining entire functioning systems.
How does that work? Defining some terms will help us clarify.
How cloud changed computing
With cloud computing, enterprising integrators seized an opportunity to create productive economies of scale. Rather than incorporate hardware and software to accomplish tasks at their clients’ locations, they began building that infrastructure in their own data center and selling the outcome of the services that infrastructure provided — rather than selling the actual hardware and software.
That way, providers could then sell the same service, using the same infrastructure to produce it, to other customers. As a result, the cost of operations is paid by multiple customers that contribute additional profit along with the cost recapture.
A “cloud,” then, is really just a data center that sells compute-related services to clients. Any productive remote data center may be considered a cloud.
Open networks ease the cloud integration journey
IT directors are now charged with evaluating cloud service providers to select the ones they determine to be most appropriate — and many select different cloud services for different tasks. Once the services have been selected, they must be made to work well together.
This need has created the new role of cloud integrator, who is charged with selecting and integrating cloud services from multiple cloud vendors into superior solutions. This means aligning the various protocols at every level of the network between multiple services from multiple vendors.
Imagine trying to create effective high-performing connections between diverse systems in an environment in which each of those systems is proprietary. Closed systems are often difficult to configure, and that’s very much by design. .
Open platforms remain very popular among developers and system administrators because they provide maximum flexibility and complete transparency. All the application programming interfaces (APIs) that are used to facilitate inter-system connections are fully exposed where developers can get them. Data structures can be altered to accommodate differences in the data structure of other systems that are being incorporated. Code can be modified to improve interfaces.
Cloud integrators prefer open systems because they make it easier to do their jobs quickly, efficiently, and effectively.
Traveling the open (network) road
Once all the data entities and structures are aligned, the only thing left to do is to make sure that all of the network traffic travels along obeying the same rules of the road to avoid bottlenecks or other conflicts.
Adoption of standard transmission protocols will satisfy the base requirements, leaving the question of how to further optimize network traffic between diverse cloud-based components and on-premises systems to create maximum efficiency.
Many engineers faced with the challenge of making things work better “where the rubber meets the road” have elected to create a better road.
The same is true in the building and operation of data management systems. Once you’ve exhausted all the ways you can optimize the program code, the data structures, and the programmatic interfaces between them, the only way to create any further improvement is to modify the network traffic to optimize throughput and minimize bottlenecks.
Originally, sophisticated intelligence was built into network routers that system administrators could configure directly to achieve performance improvements. This, however, gave rise to increasingly closed systems, as each manufacturer tried to create competitive advantage for their network equipment by improving the software. They then had to protect that improved software.
Today, software-defined networking (SDN) is reversing that trend, creating a control layer where code can be developed in a completely transparent multi-vendor environment that will optimize the interfaces between various network components.
When the connections between various network communications provisions are as flexible and programmable as those used to create the software, developers have every imaginable tool at their disposal. This maximizes their ability to be fully creative and innovative in the development of superior interoperability.
Open networks create the information transference that truly optimizes the performance of open systems.