The answer to a successful multi-cloud strategy lies in the network

Evidencing this new reliance on cloud, Gartner projects that the Desktop-as-a-Service (DaaS) sector will experience vast growth in 2020. Fuelled by the remote working situation, worldwide spending is expected to increase by 95 per cent year on year to $1.2bn1, up from $616m in 2019. Given the importance of cloud, many organisations are now adopting multi-cloud strategies to leverage multiple cloud platforms to support an expanding range of Software-as-a-Service (SaaS) and corporate workloads, each with unique performance and service level requirements.

With so much relying on the power of the cloud now, businesses need to know they’re achieving a multiplier effect from their cloud investments. Picking and choosing which provider to use for which service, based on that provider’s particular strong suits across the different aspects of the cloud environment, is a natural move to ensuring businesses are getting the most from their money. For instance, a company may be looking for the Google Cloud Platform’s strength in machine learning or Microsoft Azure’s seamless integration with Office 365 and Microsoft databases – each platform has its advantages, so why not use each where it is best suited.

However, it’s not simply a matter of putting various cloud services and applications into a variety of different baskets. Essential to the success in guaranteeing optimisation and return on investment while adopting multi-cloud strategies lies in several variables, but the most important of these is arguably the underlying network infrastructure upon which the cloud is supported.

Cloud demands advance beyond the capacity of legacy infrastructures

As a result of the accelerating adoption of cloud and multi-cloud strategies, enterprises are experiencing additional demands being placed upon their wide area network (WAN) infrastructure. However, many companies still rely on traditional, router-centric WAN approaches that are increasingly becoming outmoded. Indeed, in the face of multi-cloud strategies, even basic SD-WAN solutions struggle to cope.

These WAN solutions were never designed for the cloud era and can force compromises on organisations when it comes to choosing which cloud strategies to employ. Fundamentally, this is because network traffic patterns have shifted dramatically since the inception of cloud. Applications are more demanding upon the network, and security needs are of a greater importance now that everything is open and connected to cloud, increasing the potential attack surface.

Traditional, router-centric networks send all traffic from branch offices on a circuitous route back to the data centre instead of straight to DaaS, SaaS and Infrastructure-as-a-Service (IaaS) applications

and instances. The requirement to backhaul is due either to an inflexible architecture and/or security requirements that dictate advanced inspection that conventional routers lack the ability to perform.

This long-winded path, sometimes called the trombone effect, adds significant latency when connecting user to cloud applications and can cause significant impairment to performance and business productivity. Before the pandemic, this may have represented nothing more than an inconvenience, but with remote workforces now highly dependent on the cloud, these issues can cause significant disruption to company operations.

To better enable the practice of multi-cloud strategies, it falls upon advanced software-defined WAN (SD-WAN) networking solutions to provide the best in service option as networking infrastructure to support complicated cloud demands. SD-WAN technology enables enterprises to connect users directly to applications by using any underlying transport option available on the network. This might take the form of broadband internet, multiprotocol label switching (MPLS), or even 4G and 5G.

SD-WAN solutions offer an alternative to router-centric infrastructure as it increasingly slides into redundancy; however, not all solutions are best-suited to supporting multi-cloud environments and there are a number of criteria that have to be met to optimise its benefits.

Guaranteeing multi-cloud strategies

When selecting an SD-WAN solution to deliver on the promises of multi-cloud, enterprises must ensure that cloud-hosted applications can perform well over any underlying network, without compromising end user experience. Before examining the SD-WAN requirements in supporting multi-cloud, it is important to identify the key drivers as to why organisations put multi-cloud in place, as there are the ambitions that need to be fulfilled.

First and foremost, enterprises are seeking a means to assure optimal application availability and reliability, so that even if one WAN transport service experiences an outage, applications will remain available by failing over to the remaining service.

Following this insurance in maintaining operations, enterprises are seeking to support regulatory and compliance requirements, thereby allowing organisations to maintain the integrity of application data in a specific geographic region in accordance with local legal mandates. This is especially important for large enterprises whose operations span worldwide jurisdictions.

Lastly, in adopting multi-cloud strategies, enterprises are looking to reduce IT spend on IT infrastructure CAPEX and to increase agility to on-board and connect enterprise users to new applications. And through automation and centralised management, IT organisations can shift their focus from network maintenance toward the deployment of new cloud applications and services.

The answer to multi-cloud lies in a business-driven SD-WAN

Not all SD-WAN solutions can fulfil the criteria to meet the ambitions that lead enterprises to adopt multi-cloud strategies. The answer to optimised multi-cloud lies in a “business-driven” SD-WAN solution – WAN infrastructure that acts as an enabler to the company, rather than a constraint.

In the context of multi-cloud, this means the WAN must ensure that end users accessing cloud applications always have an excellent, consistent quality of experience whether they are working from a remote office, branch or corporate campus site.

To ensure this, there are three key SD-WAN platform requirements that enable IT to realise their multi-cloud deployment objectives and support their end users’ ability to access these multi-cloud applications directly and securely.

First, the SD-WAN platform must automate branch-multi-cloud connectivity to scale applications horizontally across multiple cloud providers with the flexibility to select the best cloud provider for the right application.

Second, the platform must also be able to identify and assure the consistent performance of business applications by applying granular quality of service (QoS) and security policies based on the end user and group profiles prioritised by application.

These two provisions allow for the smooth running of business-critical applications across the multi-cloud upon which company operations depend. Thirdly, to truly guarantee multi-cloud performance, these technological criteria must be met in conjunction with the ability to support applications across the leading public cloud providers.

Looking towards a multi-cloud future

The pandemic has accelerated the adoption of multi-cloud strategies, but this trend was well underway even before the turn of the year. Such technology offers soaring promises of digital transformation and the option to use the best cloud service when and where it is required. However, as cloud adoption itself increases demand on the network, this is even more so the case for multi-cloud. If companies want to truly make the most of the cloud and all it has to offer, their first target lies in employing a business-driven SD-WAN.