10 reasons why a hybrid cloud is better
Editor’s Note: We’ve updated this post, originally published in 2013, to reflect the realities of cloud in 2020.
Technology trends eb and flow over time. One consistency is that with each passing year, new trends and recommendations replace the old ones. Well, seven years later, I’m pleased to report that everything in this original article remains as true today as it was when it was written. That being said, there have been some changes that enhance everything we stated seven years ago.
The SaaS Element
In 2013, SaaS was relegated to your CRM or email. So much has changed in the past seven years — SaaS is now a very real option for several of your enterprise workloads. Salesforce is now so much more than a sales tracking application and is extremely extensible, even allowing for complex workloads — not to mention what’s available in FinancialForce.
In 2013, the private cloud was often associated with physical infrastructure running in the data center of your choosing. This could mean a physical server, a virtualized environment or an OpenStack deployment. During the intervening years, hyperscale cloud providers have developed services that enhance capabilities within your data center. For example, Microsoft offers Azure Stack to extend Azure services across data centers, edge locations and remote offices. Microsoft has also recently launched Azure Arc to help create management consistency across Azure, on-premise data centers and other clouds.
The Public, Private Cloud
Another enhancement is the ability to run VMware-based virtual workloads natively on AWS or Azure. While this might feel like it creates confusion, you can think of this still as a private cloud – it’s just that the underlying infrastructure is running on a hyperscale cloud provider’s infrastructure.
One last amendment
While the original article identified security as a driver for which cloud infrastructure to choose, security today can be delivered consistently across all platforms. More than that, security should be delivered consistently. Today, you have the ability to monitor and manage the entire attack surface uniformly and effectively with a range of security services on the market from both managed service providers and specialized boutique firms.
What this means to you:
Infrastructure decisions used to be made on a frequency of every one to three years. But business and technology are changing so quickly now that a company’s entire technology stack should be on a roadmap for reconsideration every quarter. This doesn’t mean you change direction every quarter, it just means you have to stay aware of what has changed in your market, your business and in the larger technology market. Specifically, you need to reconsider in which platform each of your applications should be running. Here are the broad buckets:
- Private cloud (VMware, HP, DellEMC, Microsoft, OpenStack, RedHat)
- Public cloud (AWS, Azure, Google Cloud Platform, OpenStack, Microsoft 365)
- Software-as-a-Service (Salesforce, Workday, ServiceNow, Marketo, Okta, HubSpot)
My challenge to you: Either establish a team that can focus on this as a full-time job, or partner with a company who can bring an outside-in view to your technology stack.
Editor’s Note: this article was originally published in 2013.
The hybrid cloud is the future for many companies. Businesses large and small are turning to blended infrastructure because it combines the best of all worlds: public cloud, private cloud and dedicated servers working together in any combination.
Individually, the public cloud, private cloud and dedicated servers can stand on their own legs as valuable architecture. But what our customers have found is that when it comes to the cloud, one size does not fit all. Instead, you can gain a synergistic value from combining different infrastructures into a single product portfolio. The ability to create a common computing architecture can deliver better reliability, optimized performance and greater cost value. But that’s not all.
Here are 10 reasons why a hybrid cloud might be better for your business and your customers:
1. Greater architectural flexibility
A complete hybrid cloud portfolio enables you to place workloads where they make the most sense. You can align your architecture to take advantage of performance requirements that only dedicated servers can offer. At the same time, you can meet financial objectives with contract term billing to meet regulatory or investor expectations, while also taking advantage of utility billing for unknown or short-term needs. And of course, the cloud enables rapid deployment times that come in handy when deadlines get shortened or requirements change.
2. More technical control
Root access can be granted to appropriate personnel, and custom network segmentation can be defined to logically and physically segment your architecture where needed. Unlike cloud-only solutions, a hybrid cloud portfolio enables you to define if and where multi-tenancy is acceptable. You can also request customized hardware as needed to meet your specific application performance requirements.
3. Better security
A common concern about cloud technology among enterprises is security and multi-tenancy. A hybrid portfolio eases these concerns by allowing you to choose dedicated servers and network devices that can isolate or restrict access. Furthermore, your devices can be configured so your dedicated servers and cloud servers can communicate on a private network, turning what were two separate solutions into a single integrated architecture. Rackspace has been providing this option to customers for several years.
4. Compliance options
With better security comes the ability to address compliance requirements. Many auditors frown upon multi-tenancy and require dedicated solutions for some (and sometimes all) aspects of your hosted infrastructure. With proper network security, a hybrid solution blending cloud and dedicated hardware can satisfy audit requirements.
5. Peace of mind
Managing vendor relationships can be daunting. Different billing capabilities, pricing methodologies, differences in technological priorities and varying levels of support can make it difficult to keep your outsourced solutions running smoothly. Using a single provider that offers both a robust portfolio and customer-centric account team can greatly reduce your stress levels thanks to the concepts of “a single throat to choke” or “a single person to hug”.
6. Utility billing
We have often said that the cloud is for everyone, but not everything. There are simply some workloads that demand the performance and security that only dedicated hardware can offer. However, you’ll still need the flexibility and speed to market of cloud services when it comes to completing batch computing jobs, bracing yourself for traffic surges (whether expected or not) and preparing for peak business periods, among other reasons. A hybrid cloud lets you own the base configuration and rent the spike, so you can pay only for what you use.
7. Technical predictability
Developers work hard to create stable applications. Development and testing are common uses for cloud servers (think “rent the spike”), but when it comes time for production, you want to know exactly how your platform will perform. In many cases, public cloud will work fine, but some businesses may prefer to stick with dedicated servers for at least some of their mission-critical launches. For example, dedicated servers can be configured to meet performance needs and then be supplemented with multi-tenant cloud servers for overflow traffic.
8. Operating system choice
Most public cloud vendors offer a wide variety of operating system choices. This can be valuable if you need to test products or engage different technical audiences. If you resell hosted services, your customers will have the flexibility to choose their preferred environment, and you will know that you have the ability to rely on dedicated options for customers that need them.
9. Fostering innovation
The ability to spin up and spin down cloud servers is very appealing to developers, especially if they can do so on their own without the process or time restrictions often required by internal IT departments. This building block mentality can promote research and development flexibility by liberating technical creators quick set up/tear down of cloud servers — think proof-of-concept projects, pilots, software trials, etc. Once goals are achieved and tests are complete, cloud or dedicated resources can be deployed to meet production requirements going forward.
10. A technology safe harbour
Application sprawl is the concept of business units purchasing IT resources from external vendors for a wide variety of reasons. For example, maybe your developers are going to a provider to spin up and tear down servers to meet a short-term project timeline. A company could create a single sandbox environment on dedicated servers with the performance and security features you need, while also being able to connect to flexible cloud resources. This could then provide a controlled and finite reusable technology pool where internal departments, external vendors, partners and maybe even customers can deploy and test solutions. Internally such a configuration might even become a profit center.
One consideration we’ve deliberately excluded from the list is cost. That’s because there are many variables that determine the value and cost of any given hosting solution, at least from our perspective. Is a cloud solution less expensive than a dedicated solution? Sometimes. But in reality, that answer depends on a number of factors. Every hosting solution is unique, and every customer has specific requirements related to technology, support, financial and business objectives. Working with a hybrid cloud provider can give you the luxury of choice with the confidence of consistency.