Network Orchestration is a Key to Success in Meeting Application Synergies for Combined On-Prem and Edge Workloads

By: Simon McCormack, Senior Manager of Product Management at Aruba, a Hewlett Packard Enterprise company

By the end of next year, the majority of enterprise network infrastructure teams will have completely changed their operating models to support applications, people, and IoT devices in entirely new ways. Network orchestration is a key technology that can help achieve success for tying everything together as the rapid shift to a “centers-of-data” model accelerates.

Don’t be left on the tarmac as this jet departs.

According to Gartner, “The IT we have known is changing forever. Infrastructures of the future will not be architected based on existing topologies, rather they will be deployed on a global scale, driven by business requirements and unspecific IT vendors. The end result will be an environment that is focused on enabling the rapid deployment of business services (by the business) and deploying workloads to the right locations, for the right reasons, at the right price. The drivers of this change will not be technology in many cases, but business, markets and customer requirements.¹

Simultaneously creating new operating models for the COVID pandemic, network teams are now quickly creating new strategies and processes for connecting on-premises and edge locations. Gartner predicts that “By 2022, 60% of enterprise IT infrastructures will focus on centers of data, rather than traditional data centers.”²

Network orchestration tools play a huge role by automating manual processes that are cumbersome, time-consuming, and error-prone.

According to SearchNetworking, “Network orchestration is a policy-driven approach to network automation that coordinates the hardware and software components a software application or service requires to run. An important goal of orchestration is to automate the way network requests are carried out and minimize the human intervention required to deliver an application or service. For example, if a cloud storage provider receives an order for 2TB of storage through its customer-facing website, the provider’s orchestration platform can translate the order’s requirements into configuration tasks for network devices to execute.”

Because applications now are spread across many points, remote locations and their respective networks need to be managed differently. Factoring in cloud-based architectures and application mobility requirements further justifies the need for considering automation and software-defined technologies.

Making matters worse, building out enterprise data center infrastructure to simultaneously support on-premises and edge locations present a new set of complex challenges in planning, integrating, validating, and managing all the required network infrastructure components. Even when all the components come together in a predefined composable offering, there remains the problem of evolving legacy operations and processes. This includes building out data center infrastructure across an entire network fabric, spanning edge to cloud.

At a high level, network orchestration tools become the foundational glue linking the trio of compute, storage, and application resources. They prevent fabric bottlenecks, on-premises or at remote locations, from negatively impacting the ability to deliver and manage applications efficiently downstream.

Shedding additional light onto this challenge, edges of siloed compute and storage are often connected via disjointed network architectures, with operating models that hinder the crucial elements of centralized management, orchestration, security, and visibility.

The infrastructure, staffing, and processes for supporting edge-connected users, applications, and a fast-growing population of IoT devices has created complexity, stretching the overburdened capabilities of networking and IT teams like never before. Increasingly, they must now manage workloads that are processed closer to where data is created, often across many edge locations.

While the interconnections across edges are increasingly solved with technologies like SD-WAN and SASE, overall IT operational simplicity remains a major problem for some organizations to tie everything together. 

A Single Networking Fabric for Centers of Data

In response to addressing this inevitable reality, intelligent software-defined network orchestration solutions can help reduce the broader IT struggles with manual, siloed IT provisioning and operational models across the entire datacenter paradigm, as the traditional data center transforms into polymorphic “centers of data.” By orchestrating a discrete set of switches as a single networking fabric topology, administrators can significantly simplify day-to-day operations to reduce operational costs and time to deploy new services.

Furthermore, IT generalists may not have deep networking expertise and also must increasingly contend with disjointed IT service provisioning. In response, the ability to provision and manage operations all from a single console can provide a powerful and simplified operating model.

Network orchestration establishes end-to-end network visibility as a single fabric, enabling simultaneous connectivity across hosts, virtual machines, VLANs, services, and workloads. This simplifies and reduces the time requirements for troubleshooting connectivity issues and identifies performance problems before applications, users, or the business is impacted.

Workflow visibility

Orchestrating network fabrics can also provide a greater understanding of the physical elements and the workloads that traverse across the network. It can also help gather the necessary information from third-party management tools and other repositories by leveraging RESTful APIs. As a result, network teams have a complete picture of the compute/storage/network resource ecosystem. This delivers a more accurate and meaningful global view of how and where virtual demands are impacting the physical network.

Workload Optimization

Software-defined orchestration and automation tools also can also help optimize data paths to satisfy the needs of critical application workloads. Policies created around explicit workload demands and SLAs, such as latency, are met by arranging the network to satisfy those needs first.

Orchestration allows the network to support workload needs by delivering deterministic and predictable performance, ensuring the right network resources are applied to the right workloads. This, for example, can translate to faster storage replication and mirroring or quicker database transaction responses.

Now it’s time to board the Orchestration jet to realize these benefits.

Taken together, network teams will increasingly recognize the need and the benefits that a unified networking fabric can provide for individual users, all the way through to the broader functionality of the business, from provisioning to workload management and optimization.

For more about the Aruba software-defined automation and orchestration solution, Aruba Fabric Composer, visit https://www.arubanetworks.com/products/switches/core-and-data-center/fabric-composer/

¹Gartner, Infrastructure is Everywhere: The Evolution of Data Centers, David Cappuccio, Bob Gill, January 18, 2021, Pg. 2

²Ibid.