Cloud computing for business

Once the exclusive purview of large blue-chip organisations with seemingly bottomless budgets, remote data hosting has become increasingly affordable and available, helping to level the playing field and enable smaller businesses to compete with larger enterprises in their industry. But if you're a little hazy on whether your business should be in the cloud, on the edge, or fog networking, we're here to offer some clarity, and to help you decide whether an off-premises, decentralised infrastructure is the next step for you.

• 14 September 2022

Cloud, edge, or fog: what's the difference

Before looking at the benefits of decentralised architecture, we should cover the different terms applied to different types of decentralised approaches. Cloud, edge, and fog computing are all types of decentralised data management solutions. While there may be some overlap between all three, their applications and use cases are distinct, and deciding which ones you opt for will be dependent on the nature of your business and what you want to achieve.

Cloud computing

Cloud computing can best be defined as a model that stores, processes, and manages data in an offsite data centre, with all applications and services being accessible over the internet. Vendors like Salesforce, Dropbox, and Vooware provide Software as a Service (SaaS), which is also referred to as 'cloud application services'. SaaS enables customers to access services directly from a web browser with no need to install any software locally, and with vendors managing all aspects of maintenance, upgrades, and security patching.

Greater flexibility and control are offered by Infrastructure as a Service (IaaS) models, which create cloud-based virtual data centres through which clients can access their servers and storage solutions via an API or dashboard. This gives clients complete control over their infrastructure, but also places responsibility for the management of applications, data, middleware and other architectural components with them. Vendors will still manage servers, storage, virtualisation, and other aspects of the architecture. IaaS is a highly flexible and scalable cloud computing model, with clients able to purchase additional hardware and other resources as and when they are needed. IaaS providers include: Amazon Web Services (AWS), Rackspace, and Microsoft Azure.

Clouds can be public, private, or a combination of public and private. These distinctions are referred to as cloud deployment models, and affect how users access data. Public clouds are subscription-based services for multiple customers, managed and maintained by the service provider. Examples of public cloud providers include: Amazon Web Services; Microsoft Azure; and Google Cloud Platform (GCP). Private clouds are owned and used by a single entity. Servers might be on-site and owned by the entity making use of the cloud, or they could be offsite and managed by a third party. The key difference is that no other entities will be using the servers or resources in a private cloud. A private cloud is like renting a house, as opposed to the public cloud, which is more like renting a flat.

Hybrid cloud solutions combine private and public cloud storage and management, and are one of the most popular ways for businesses to manage their data and services, as they enable organisations to continue using onsite servers they have invested in, while simultaneously taking advantage of the benefits offered by the public cloud. Hybrid cloud solutions will always incorporate at least one private cloud solution, and at least one public cloud solution. If the solution is comprised of more than one public cloud solution (as well as the private cloud solution) it is considered to have a hybrid cloud and multi-cloud (see below) architecture. This can help maintain compliance by giving you control over where your application sits, and where hybrid computing occurs, as well as enabling you to expand on your existing setup with minimal data centre costs.

The term 'multi-cloud' is often used interchangeably with 'hybrid cloud', but there are distinct differences. Multi-cloud solutions will always employ more than one public cloud. For example: a website front end might be hosted by AWS, while the back end is managed by Microsoft Azure. Multi-cloud solutions may or may not incorporate a private cloud as well as multiple public clouds. If they do, they would be considered to have a multi-cloud and hybrid cloud architecture. This enables businesses to choose the best providers for specific components of their architecture, thus optimising their applications.

Edge computing

Edge computing is the term for the processing of data near where it is being generated, rather than passing it to the cloud for processing. This is useful in applications where real-time data is important, and is commonly used by the energy sector on oil rigs, where data is gathered by sensors on the rig and processed at that point to avoid having to pass it back to onshore nodes, which could introduce unacceptable levels of latency. This is how edge computing gets its name, with data being processed either by IoT devices or remotely located servers on the 'edge' of the network, with all data being passed back to the central nodes for storage or further processing. The chances of interception or malicious attacks on data are very low with edge computing, due to its remote and real-time processing. Autonomous vehicles also employ edge computing to process environmental data at its point of capture. Fractions of a second in response time can be a matter of life and death where autonomous vehicles are concerned, so mitigating against latency in this way is vital.

Fog computing

Fog computing is an architectural layer between edge computing and the cloud. Fog nodes are located closer to the cloud than edge nodes, and act as a filter for the vast amount of data received by edge nodes. Edge nodes pass all data back to the cloud, regardless of its importance. With a fog layer, the fog nodes analyse the data first, assess it for importance, and only pass important data back to the cloud. Unimportant data is either deleted, or stored in the fog node for further analysis and processing.

Fog computing tends to be more scalable than edge computing, with operational costs running lower. However, the power requirements of fog nodes tend to be higher than that of edge nodes.

Why migrate to the cloud?

With a better understanding of what cloud computing is, and how it and its various architectural extensions work, it is easier to assess whether migrating your business to a decentralised architecture would be beneficial.

The pandemic saw a surge in businesses migrating to the cloud as a means to facilitate remote working for their staff. For many, this revealed the range of other benefits that cloud architecture can offer, which has resulted in 94% of enterprises using some kind of cloud service.

Initial outlay on server equipment is mitigated. Ongoing electricity costs are also significantly reduced, and cloud computing can reduce operational costs by freeing up time your DevOps and sysadmin teams would need to spend on backups and hardware maintenance if using onsite servers. These teams can then be tasked with activities that are more valuable to your business, improving digital transformation and potentially discovering new ways to generate revenue. Budgeting is also more efficient with cloud architecture, as many vendors will offer pay-as-you-go pricing models, meaning you only ever pay for what you use. And if you underestimate or overestimate the computing power you need, cloud models enable you to scale up or down instantly, meaning less wasted cost, and fewer missed opportunities as you adapt with improved agility to unforeseen events or unexpected market shifts.

As well as cost savings, cloud models also make enterprise-level solutions available to businesses of any size. This can be a great way to gain a competitive edge over other companies in your industry, and helps any business compete with large, established players. Having world-class technologies at your fingertips can greatly reduce build times and increase efficiencies. It is not unrealistic for an organisation to reduce their development environment build times from several months to just a few minutes after migrating to the cloud.

Downtime is also greatly reduced as upgrades and system patches take place seamlessly with cloud models, usually with no need for applications to go offline. This means your systems are always up-to-date, giving you the latest and best available security. Top cloud solutions vendors ensure their data centres are protected by industry-leading protocols, which not only helps ensure your data is safe, but also ensures you will almost certainly be compliant with any security requirements your clients or partners may have regarding the processing and storage of their data with your systems. Data centres are also monitored 24/7 using a combination of automated detection and manual testing, so threats are usually pre-emptively prevented before they have a chance to manifest. Resilience to disaster events is also increased, with cloud service vendors having robust disaster recovery strategies and redundancy servers that would be prohibitively expensive for many businesses using onsite servers.

Cloud solutions also offer flexibility over location. Your staff can be working in the field, onsite with clients, from home, at an internet cafe, or anywhere they have an internet connection, from any device that is web enabled. This gives you the freedom to expand your operations and explore cost-saving strategies, such as 100% remote working. It also provides resilience for your business against uncontrollable events like the pandemic lockdowns, as the need for your staff to work onsite is mitigated, helping your business keep trading even if your offices have to close. Flexible accessibility also increases opportunities for collaboration, with teams located in different cities or even different countries able to work together on the same systems in real time. Studies suggest that cloud-based collaboration tools can increase a company's productivity by as much as 400%.

Migrating your business to the cloud

Migrating to the cloud doesn't have to be complex, but it should be carefully considered, with detailed plans and roadmaps. This might begin with capturing a rationale for movement to the cloud. As we have seen, there are many potential benefits, but the ones that will have the most impact on your business are the ones you should focus on, as they will inform the strategy for your migration. With a clear idea of what you want and expect to achieve from the migration, you will be able to set KPIs to monitor the progress of the migration, and establish metrics to measure its success.

A cost-benefit analysis, followed by investigation and comparison of cloud service providers to ascertain the specifics of their services, and the VM technologies they offer can also be a useful planning step.

An analysis of the types of data you want to migrate will help add detail to your plan. Assess what can be moved easily, what might need refactoring, and what the interdependencies are across all your data. This will help you plan the order in which your various data should be migrated, and make it easier to estimate how long the migration should take. It is at this stage you may want to enlist the help of cloud migration and digital transformation experts to partner with you on the project. They will be able to advise on the various types of data you have, how they should be migrated, and any potential risks.

Start with data that is easy to migrate, and not business critical. This will prove how robust your plan is, and give you confidence as you move on to migrating more complex or important data. Keep your old systems running in parallel as your data is migrated, and make sure everything is thoroughly tested in the new system before decommissioning the old one. A cloud migration partner will be able to help with the simultaneous running of two environments, the testing strategy, and the decommissioning of the old system. They will also advise on the most efficient migration strategy to help minimise the time it takes to safely migrate all of your data.

Do I need a professional cloud migration partner?

Data is one of the most valuable assets for any business, so its management and migration should be handled with care. There are a number of potential complications that can arise during data migration to the cloud. From ensuring your systems can still share new and old data correctly, to maintaining data integrity and protecting it from malicious actors or accidental corruption, an experienced professional cloud migration partner will consider all aspects and devise a migration strategy that that minimises the impact to your business continuity during the migration, and mitigates the impact of any unplanned events after migration.

A cloud migration partner will also advise on cloud models suitable for your business in relation to your available internet speed and network bandwidth, and the budget you want to assign to these. Cloud models typically require higher internet speeds, and making sure your connection is fast enough is critical to getting the most out of your migration, and avoiding latency issues which can significantly impact business productivity.

Other considerations include: volume of data movement, access control, and user authentication. Interdependencies between these processes, along with other factors, will need to be analysed to ascertain how they interact with each other and other areas of the systems. Migrating to the cloud does not make systems more agile by itself. Often, applications need to be redesigned to take full advantage of cloud native benefits. Existing strategies and processes for monitoring, configuration management, backup procedures, licencing, and maintenance of availability may not be appropriate or applicable once your business is hosted in the cloud. Expert cloud migration partners will be able to help you move away from traditional IT architectural approaches, and introduce changes that support a more agile, service-focused methodology.

If you're thinking about moving your business to the cloud and would like an expert partner to help you plan and execute a successful migration, drop us a message or give us a call on +44 (0) 8456 808 805 today.

It is not unrealistic for an organisation to reduce their development environment build times from several months to just a few minutes after migrating to the cloud.