Tech vendors are building out service infrastructure to host cloud solutions at all levels. And businesses in industries like manufacturing where cloud messaging have taken a little more time to get through are picking up on the benefits.
Organizations are consciously looking at entrusting the digitizing of areas like customer experience, workforce engagement, supplier management and the Internet of Things (IoT) through external cloud services.
In Panorama’s 2018 ERP report, there was a major increase in the number of respondents (37%) that implemented Enterprise Resource Planning (ERP) through cloud or Software as a Service (SaaS). It’s good practice for businesses, particularly large companies where digital transformation is an overarching aim and significant amounts of money need to be spent – to look thoroughly at what SaaS, software delivered from a cloud environment and where all data resides with the service provider, means for them.
With SaaS, businesses generally choose a subscription-based model where the software is hosted in the cloud and accessed via the internet. Well-known examples of this include Office 365, Google G Suite, Salesforce, and Netflix. Instead of paying a big chunk up-front, users will subscribe to the software, often monthly basis, ending the subscription when they no longer want it.
On an enterprise-level, businesses have two options – single-tenant or multi-tenant SaaS.
What does this mean, and what are the positives and negatives of each type of SaaS architecture?
If a business is hosting their own private cloud, or using a third-party private cloud, they’re likely using single-tenancy SaaS. This is where business has for itself one virtualized cloud environment using hardware that provides computing power, storage power, and networking power.
None of these elements are shared – it’s all for that one customer.
Single-tenancy systems are popular with enterprise businesses as it’s impossible to affect the systems of other companies. Because there is only one business using an instance, it’s impossible that they’ll suffer from performance issues resulting from the tenancy of another business.
There’s also much more room for customization – the business can do what it wants and not be at the mercy of bigger firms that might be using the same instance. For major systems, you don’t have the threat of forced upgrades at a vendor’s discretion.
Businesses with size and scale (50+ users) may want to choose single-tenancy, as they might be running mission-critical applications that have database-driven technologies in the background.
“Multi-tenancy is like a bus journey. Single tenant, on the other hand, is more akin to a taxi.” – Dean McGlone, V1
They won’t want to run the risk of being impacted by security breaches and bad code that could bring the architecture down or grind to a halt.
A single-tenant system can be more expensive than a multi-tenant system, as it’s less efficient in terms of resources if the system not fully loaded. If might be a good option for businesses that are totally confident they’re getting everything they need with the product.
Many manufacturers might want to use single-tenant SaaS as it offers the management of its data and internal processes at a maximum level of security and privacy.
Of course, this could be open to integration with public cloud components that allow data to be collected outside the enterprise perimeter – such as the Internet of Things (IoT) data collected by research partners.
Single-tenancy might also be good for businesses involved in eCommerce taking on financial transactions, clearing payments or storing credit card information, again because of security. Single-tenancy is often recommended for PCI-accredited hosting, as vendors can guarantee with 100% assurance that data will not be shared with other tenants.
Many public cloud services work through a multi-tenant architecture, where multiple companies will share the same cloud environment to store their data. It will be segregated to prevent different businesses from accessing each other’s data, secured at a software level.
A good way of visualizing this is to think of a tower block – the floor plans are pretty much set in stone, with individual flats all separated into different compartments. Individually, though, you can decorate and furnish these flats how you wish.
Multi-tenant systems are generally cheaper than single-tenant – and a more efficient use of resources. Because you’re pooling resources, there can be significant savings in the use of power and hardware.
It also forces the SaaS provider to get their updates seamless and right the first time – the SaaS system needs to be updated perfectly for each customer, or you’ll get into huge difficulties.
Multi-tenancy might be a good option for a business building a web application, which wouldn’t want to invest in a more expensive single-tenancy environment. The right cloud environment will be quicker to spin up and deploy, and it could be more scalable in terms of what a company might want to do in terms of computing, storage, and networking.
One major problem with a multi-tenant architecture is a lack of customization. Because multiple businesses will be basically running off the same SaaS code and database, they will not be able to adapt it to their specific needs.
Your business might also be at the mercy of bigger tenants as you are working off the same codebase – they’ll get priority when it comes to future features and functionality.
Which is better for large businesses? Single-tenant vs. multi-tenant architecture (SaaS)
Adam Binks is the Chief Executive Officer of SysGroup, a fast-growing managed IT and cloud hosting provider that provides both single and multi-tenancy SaaS options. He said that large businesses often preferred single-tenancy as they feel a lot more re-assured with its capabilities.
He says, “If you think what’s happening in the market right now, everything’s becoming a lot more complex in terms of compliance and governance – take GDPR as an example. There’s an onus on businesses to make sure data is not only used in the correct fashion but secure from breaches.
“There’s still a lot of nervousness that exists from a lot of C-Suite type people about taking infrastructure and mission-critical data from the on-premise structure and into the cloud. However, if you bed them into the cloud, a single-tenancy architecture gives them the security guarantees that makes it a lot more convincing. They feel a lot more comfortable in making that transition from on-premise into the cloud.”
Binks would advise large businesses to run cloud ERP systems in a single-tenant architecture.
He said, “The business may have evolved over a number of years with added features that they need, and typically need big computing power for running back-end databases. Running ERP much more suited to single-tenancy because you can give it that guaranteed computing resource and scale that it needs.
“The biggest thing that we find when we deploy ERP in the cloud is that it needs really high intense storage. You need solid-state performance for your database read and writes. We always advise single-tenancy for ERP with businesses that have anything more than 100 users.”
A business needs to make some important choices before adopting the cloud for their needs. At the enterprise level, they need to ask the right questions that which may be specific to the industry they’re in.
Businesses must ask the cloud provider about what the scale of the architecture looks like and what the disaster recovery options are like. In the event of a data center loss, has the cloud provider got the presence and connectivity through another data center, to keep the business operating?
Interested in using the cloud? 5 qualifying questions
A business needs to make some important choices before adopting the cloud for its needs. Here are five basic questions to ask when making the decision.
- Can your cloud provider demonstrate successful similar deployments? Many vendors will have sound solutions on paper. However, there’s no substitute for hard evidence that they have implemented your specific type of configuration. Try to find this through proof points, return on investment and third-party confirmation and advice.
- What are disaster recovery options like? In the event of a data center loss, does the cloud provider have presence and connectivity through another data center to keep the business operating? Does it test its disaster recovery plan on a regular basis? Does it have evidence that it works?
- Does your cloud provider meet security and compliance requirements? In industries such as food and beverage, there may be international regulations to deal with, as well as accountability to partners and customers? What are they doing to protect your data? Does it meet industry-specific standards you might have such as those established by the payment card industry (PCI)?
- Is the cloud solution customizable or configurable to your needs? Large businesses have unique needs that might require configuration or customization. There isn’t a silver bullet here – every individual organization will need to look at how using the cloud will positively affect their bottom line.
- Does the cloud solution have strong service-level performance? You should check carefully on whether any service-level agreement (SLA) you make meets your needs. Your business may have specific demands when it comes to availability, storage, and performance, and the cloud provider should satisfy your maintenance and support demands to keep your subscription.