Thursday, September 19, 2024

Cloud vs. On-Premise Software: the Cloud Debate

Datamation content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.

These days, whenever enterprises are deploying new software, they face a dilemma: should they use a cloud-based software as a service (SaaS) solution or should they deploy the application on-premises in their own data centers?

Cloud Storage and Backup Benefits

Protecting your company’s data is critical. Cloud storage with automated backup is scalable, flexible and provides peace of mind. Cobalt Iron’s enterprise-grade backup and recovery solution is known for its hands-free automation and reliability, at a lower cost. Cloud backup that just works.

SCHEDULE FREE CONSULT/DEMO

For an increasing number of organizations, cloud has become a viable option — at least for some needs. In the RightScale State of the Cloud 2018 report, 96 percent of the respondents said that they use some form of cloud computing.

And according to a recent IDC forecast, SaaS “will be the largest cloud computing category, capturing nearly two thirds of all public cloud spending in 2018.” The firm further predicts that enterprise applications, such as enterprise resource planning (ERP), customer relationship management (CRM), collaboration and content management applications will be the biggest categories for SaaS spending.

But the fact that organizations are spending a lot of money on cloud software doesn’t mean it is the best option for every situation. And in fact, many experts estimate that organizations only run about 10 to 20 percent of their workloads in the cloud — the rest are still in-house.

Cloud and on-premise software each have distinct advantages and disadvantages that organizations should carefully consider before choosing one over the other. Often it comes down to weighing how important it is for an organization to maintain control over an application versus its needs for speed and scalability.

Cloud vs. On-Premise: Cloud Software

Cost: Many organizations choose cloud-based applications because they hope to see financial benefits. Instead of paying a big licensing fee upfront, customers pay SaaS subscription fees on a recurring (often monthly) basis. That allows them to convert some costs from capital expenditures (Capex) to operational expenditures (Opex), which might have tax benefits and/or be more attractive to stockholders. Those subscription fees generally include maintenance and support as well, so customers are spared from these add-on expenses that they might experience with on-premise software.

In addition, SaaS eliminates the need for organizations to buy, house, run and maintain hardware — the vendor takes care of all that for them. They have lower utility bills, and they probably won’t need as much data center space.

On the other hand, companies that run the numbers sometimes find that their total cost of ownership (TCO) over five or ten years is actually higher with cloud software than with on-premise software. Also, because it is so easy to sign up and start using SaaS, organizations sometimes find that their employees have cloud subscriptions that they aren’t actually using. In fact, in the RightScale survey, 76 percent of organizations said that managing their cloud spend was a key challenge of cloud computing. And organizations with several years of experience with the cloud were more likely to point to cost management as a problem. Large enterprises might need to invest time and money in audits and/or cost optimization tools to ensure that they are paying only for the services they are actually using.

Flexibility: One of the reasons why SaaS vendors can offer low subscription fees is that they don’t offer a lot of customization opportunities. Some have marketplaces full of add-ons that can add more functionality, and some have third-party service providers that can do some customization work. In general, however, customers have to make do with cloud software as is.

Deployment: Ease of deployment is one of cloud software’s strongest advantages. Customers can usually get started using SaaS applications in just minutes. That can be a huge competitive advantage for companies in fast-changing markets.

The downside here is that employees may sign up for cloud software without management or IT knowledge or approval. This “shadow IT” can represent a significant security threat, as well as leading to compliance problems and higher costs. Enterprises need to make sure that they have appropriate governance and controls in place to manage which SaaS applications employees are using.

Security: Security has long been the biggest concern of organizations that are considering using the cloud. In the RightScale survey, 77 percent of respondents said cloud security is a challenge. However, organizations with several years’ experience in the cloud are somewhat less likely to be concerned about security, compared with those who have less experience.

Many cloud experts say the cloud is actually more secure than on-premise data centers. That’s because the large cloud vendors have the money to pay for big teams of security personnel and the latest and greatest security tools.

Then again, those big cloud vendors might be a more attractive target to some attackers, which increases risk. And not all data protection solutions and other security tools can handle data from SaaS applications, so organizations might need to re-think their security strategy when deploying cloud software.

Often, it comes down to the question of trust. Does the customer trust the cloud vendor to handle security or is the IT team more comfortable staying in control of their own protection?

Compliance: Closely related to the issue of security is the issue of compliance. This is often the biggest disadvantage of cloud apps. The regulations placed on some types of businesses may make it all but impossible for them to use SaaS solutions. In fact, in some highly regulated markets like defense contracting or financial services, it may actually be impossible to use some types of cloud software.

The good news is that cloud vendors are aware of their customers’ compliance needs, and many are taking steps to comply with more regulations and allow more companies to use their products.

Management: Companies that choose to deploy cloud software are often attracted to the idea that the vendor will handle all the management responsibilities for them. They don’t have to manage the hardware or the software — the vendor does it all.

The obvious pitfall here is that some customers may want to have some control over the management of their applications. They may want or need visibility into application performance log data or the data stored within the application, and with cloud apps, this often isn’t possible.

Availability: By their very nature, cloud apps are accessible from the browser of any Internet-connected device. That means people can access them when they want, where they want and from any device they want. Plus, many vendors have native mobile apps available as well.

In the cloud, the vendors are responsible for uptime and disaster recovery (DR), and most are able to offer service level agreements (SLAs) with availability guarantees that meet enterprise requirements.

However, organizations sometimes fail to consider a two availability issues related to Internet access. The first is latency. If your networking team is managing your LAN or WAN, you can probably guarantee fast connectivity to your own data centers. But if you are connecting to a cloud service over the public Internet, latency might become an issue, particularly is the cloud provider is a long way away geographically.

Second, some enterprise users might be in places with little or no connectivity. For example, workers on oil platforms, remote construction sites or in parts of the world where connections aren’t reliable might not be able to access cloud apps. Remember, too, that in the case of a natural disaster that knocks out the Internet, you won’t be able to access your cloud applications until both power and Internet service are restored.

Scalability: Scalability is a huge advantage for cloud software. Organizations can add more users as necessary without worrying at all about whether they have enough infrastructure to support more users. The only potential downside here is, as previously mentioned, the very real possibility that they will have users sign up for services that they aren’t actually using on a regular basis.

strong>Upgrades:As with security, compliance and management, upgrades are another area where the cloud vs. on-premise software debate comes down to the issue of control. In the cloud, the vendor controls when customers get new features. Improvements are rolled out all the time, and upgrades generally don’t disrupt day-to-day business.

However, organizations literally have no say over when their employees get upgrades. And that can lead to problems related to training, support and security; it can even cause some functionality to break if the vendor hasn’t tested the upgrade adequately or if the customer has done some customization.

Cloud vs. On-Premise: On-Premise Software

Cost: With on-premise software, organizations generally have high upfront expenses for licensing fees, and those capital expenses might not look as good on their balance sheet as operational expenses do. They may also have add-on charges for support and service.

However, some organizations find that their TCO is actually lower with on-premise software, especially when they extend their calculations out over several years.

Flexibility: On-premise software is generally much more customizable than cloud software. Vendors know that if customers go to the effort to customize software to meet their exact needs, they’ll be likely to stick with their vendor for a long time. But that customization can take a long time and lead to problems with upgrades.

On-premise software also gives organizations more freedom to choose the infrastructure that their application will run on, including the operating system, hypervisor and hardware. They also may find it easier to integrate the application with their other tools and hardware, including monitoring and security solutions.

Deployment:Deploying on-premise applications takes a long, long time, not to mention a lot of work. Organizations have to purchase, provision and configure their hardware and test the solution before rolling it out in production. That can take weeks, if not months, depending on the complexity of the application and the number of users it will serve. Vendors sometimes offer services that can speed deployment, but it is never as fast as with SaaS applications. The customer does have more control over the process than with cloud software, but deployment is one of the biggest drawbacks of on-premise software.

Security: Knowledgeable people have differing opinions on the security of on-premise applications vs. cloud software. In general, enterprises have smaller security staffs — and smaller security budgets — than the security vendors. But being in control of their own security sometimes gives organizations greater confidence. It’s also easier to integrate with other data protection tools if the application and data reside in-house than if they are in the cloud and outside the organization’s direct control.

Compliance: Compliance is often the one issue that trumps all others when organizations are choosing cloud vs. on-premise software. Only when an application is running on in-house infrastructure can the organization be completely sure that it is meeting its compliance obligations. And, as already mentioned, in some markets and industries, on-premise deployment is really the only option for some applications that utilize sensitive data.

Of course, the downside to having complete control over your own compliance is that you have complete control over your own compliance. Organizations must put it a lot of time and effort to ensure that their on-premise applications meet their regulatory responsibilities.

Availability: When organizations run applications on their own infrastructure, they are usually able to avoid latency issues. However, they often aren’t able to provide the same level of uptime and mobile support as SaaS vendors. In addition, they bear sole responsibility for disaster recovery, which adds to the management burden.

Scalability: This is another big drawback to on-premise applications. If the number of users increases, IT has to source and deploy more hardware to meet demand, which can take many days or weeks. And of demand decreases, the organization is stuck with a bunch of infrastructure which it may or may not be able to repurpose. Really, the only advantage in this area is that costs tend to be a little more predictable because organizations aren’t running into “shadow IT” problems.

Upgrades: When a vendor releases an upgrade for on-premise software, the customers usually have the ability to decide when (and if) to roll it out to their users. They can test everything to make sure that it works with their infrastructure and other systems, which can help eliminate hassles.

On the other hand, vendors often release new feature upgrades sporadically, perhaps once a year or less. And there may be a delay between the time that security upgrades are released and when enterprises can deploy them, potentially putting them at risk. And when organizations do install upgrades, they may have to temporarily shut down systems, which can disrupt work.

Making the Decision: Cloud vs. On-Premise Software

In the end, there is no one right answer to the question of whether to deploy SaaS applications or on-premise software. Organizations will need to evaluate their needs, calculate their TCO and determine the best fit on a case-by-case basis.

In recent years, the clear trend has been away from on-premise software and toward cloud-based applications, and most experts believe that trend will continue. However, those same experts are also generally quick to point out that it is unlikely that most organizations will ever move completely to cloud software. The cost and compliance issues, in particular, will make the in-house option very attractive for some applications for many years to come.

Cloud vs. On-Premise Software

 

Cloud Software

On-Premise Software

 

Advantages

Disadvantages

Advantages

Disadvantages

Costs

·         Converts Capex to Opex

·         Low upfront costs

·         Reduced overhead costs related to hardware, real estate, utilities, etc.

·         Maintenance and support are generally included

·         May require effort to optimize costs

·         May be more expensive in the long run

·         May be a one-time fee

·         May be less expensive over time

·         Lower Opex

·         Higher Capex

·         Overhead costs

·         Maintenance and support fees can add more expense

Flexibility

·         Add-ons and 3rd-party tools may be available

·         Little customization may be available

·         More customization opportunities

·         Customer can control the hardware the software runs on

·         Customization takes time and may add expense

·         Customization may lead to problems when upgrades occur

Deployment

·         Usually requires only minutes

·         Usually requires very little effort

·         Easy deployment may lead to “shadow IT”

·         Customer controls the deployment process

·         Vendor may offer services to assist with deployment

·         Deployment may take weeks or months

·         Customer needs to provision hardware and ensure that it will support the application

Security

·         Vendors have large teams of security experts

·         Vendors deploy state-of-the-art security

·         Vendors are an attractive target for cyberattackers

·         Customers don’t have complete control over security

·         May not integrate easily with other security tools

·         Customer has complete control over security

·         May integrate more easily with other security tools

·         Customer may not have as much security expertise as the vendor

Compliance

·         Some vendors have services that meet industry-specific compliance requirements

·         May not customer’s compliance needs

·         Customer must trust vendor for some compliance issues

·         Customer has complete control over compliance

·         Customer has complete control over compliance

Management

·         Vendor handles most hardware and software management tasks

·         Customer has little control over application management

·         Customer may have little visibility of application performance and data

·         Customer has complete control over application management

·         Customer has visibility of application and data

·         Requires customer to have staff to handle hardware and software management

Accessibility (include mobility, latency, DR)

·         Usually accessible from anywhere, anytime, any device

·         Vendor handles disaster recovery

·         Vendors generally have high availability

·         May have higher latency

·         Vendors sometimes experience high-profile outages

·         Cloud applications require Internet service

·         Usually lower latency

 

·         May not be accessible from mobile devices

·         Customer is responsible for availability and DR

 

Scalability

·         Highly scalable

·         Costs can rise unexpectedly

·         Costs are more predictable

·         Less scalable

Upgrades

·         Vendor handles upgrades

·         New features are available immediately

·         Generally not disruptive

·         Customer cannot control when upgrades roll out

·         May result in more support calls

·         Upgrades are predictable.

·         Customer has control over when upgrade takes place

·         Customer has to wait for access to new features

·         Upgrades are often disruptive and may require downtime

·         Potential security vulnerabilities

Subscribe to Data Insider

Learn the latest news and best practices about data science, big data analytics, artificial intelligence, data security, and more.

Similar articles

Get the Free Newsletter!

Subscribe to Data Insider for top news, trends & analysis

Latest Articles