Solution architects often find themselves specializing in a particular application or platform vendor over time. For my part, I’ve spent the majority of my career focused on Oracle databases – mostly on the infrastructure side. And as far as platforms go, it’s not a bad one to hitch your wagon to. Oracle is by far one of the leading providers of enterprise software products, particularly its own brands of database management systems.
While many IT organizations use Oracle for efficient collection, processing, and storing of data, a common obstacle is the upfront and ongoing cost of deploying Oracle applications and databases. Drawing on my experiences with hundreds of Oracle deployments over the years, here’s my perspective on how hyperconvergence can help you cut down on licensing costs and save on infrastructure needed to support your next Oracle deployment.
#1 – Avoid overpaying for licensing and penalties
Oracle database licensing is highly complex. If you’re new to Oracle or you don’t know the rules of the road, you could find yourself paying millions in licensing and fees that might have been avoided. The good news is that hyperconvergence can help by cutting down on the number of licenses you need, and simplifying the audit process.
First, Oracle licensing is based on the total number of cores running on the physical infrastructure where it is deployed. When Oracle is just one of many workloads running in a large cluster or shared storage environment, you are on the hook for all physical cores in the cluster. This means that unless Oracle is completely isolated, you’re likely paying – or should be paying – for cores that are actually supporting other workloads. Oracle enforces these terms by conducting periodic audits of customers. If they audit your organization and find that you’re not in compliance, you may be subject to a financial penalty.
This complicated licensing scheme is often overlooked and can lead to big penalties. As this helpful TechRepublic article explains:
“A common scenario we uncover when working with customers is that the IT administrator installed Oracle database products on a virtual server, but failed to recognize the licensing impact of this action. They may have assumed that only the virtual server needed to be licensed, when in fact, they should have taken into account the underlying architecture of the physical servers. Or, in the case of a VMware cluster, the whole cluster needs to be licensed for every physical server where the virtual server could potentially be running.
Small ‘oversights’ like the one illustrated above can be very expensive mistakes to make – to the tune of millions of dollars in penalties. It’s only really now that CIOs are uncovering the extent of any non-compliance with Oracle licensing rules, as they either renew their license agreements or respond to requests for an audit.”
For many environments, especially larger scale deployments, hyperconverged infrastructure can help by isolating Oracle workloads and thereby limiting the number of CPUs that could be counted in an Oracle licensing audit. Rather than deploying Oracle in a large shared storage environment, you can size your Oracle deployment for just the number of hyperconverged building blocks you need, and then deploy them as a separate cluster or federation. This not only saves on licensing costs, it also greatly simplifies the audit process. Here’s an excerpt from a recent Forrester study that explains how SimpliVity helped a customer breeze through their audit process:
“One of the interviewed companies was able to avoid an increase in fees after a recent software licensing audit by using SimpliVity’s hyperconverged infrastructure instead of a multiport SAN, which had exposed the company to licensing complications in the past. The director of infrastructure at the healthcare provider told Forrester, ‘We were completely compliant with licensing, and using SimpliVity helped us avoid gray areas that big software vendors try to expose you to when using shared storage.’”
#2 – Lower your data protection costs
When you’re architecting any mission-critical deployment, you have to account for the cost of protecting the applications and data. In a typical Oracle deployment that often means purchasing add-ons like Oracle Data Guard or other third-party solutions.
There are two ways that SimpliVity hyperconverged infrastructure can help in this scenario. First, SimpliVity’s built-in data protection can be used to provide VM-level protection with less overhead and greater ease of use than RMAN. In the latest Oracle reference architecture, we demonstrated a full backup of a VM in just seven seconds, with no performance degradation while the backup was occurring, and a restore of the same VM in four seconds, with no database corruption.
For offsite disaster recovery, we offer bandwidth-efficient replication at the VM level between any SimpliVity systems in a global federation. And you can manage all of it from a single vCenter interface. Our globally-aware deduplication ensures that only unique blocks are transferred between remote sites, which allows you to move very large data sets without the need for special-purpose WAN optimization appliances. You can even automate the recovery of workloads using SimpliVity RapidDR, with recovery point objectives in minutes and recovery time objectives in seconds.
This level of protection is good enough for most organizations. In fact, nearly half of all SimpliVity customers have completely eliminated third-party backup or replication solutions in favor of SimpliVity’s built-in data protection.
And by moving away from Oracle Data Guard, you can further reduce licensing costs because you don’t need to pay for the second Oracle instance in your disaster recovery site, as long as it’s only used for periodic disaster recovery testing.
#3 – Cut down on the number of devices supporting your Oracle deployment
Hyperconvergence can lower your operating costs by reducing the number of devices required to host and protect your Oracle workloads. In fact, SimpliVity reduces number of devices in the data center by 90% on average. That’s a 10-to-1 reduction in power, cooling, and hosting costs.
And by converging all IT infrastructure for virtualized workloads onto simple, scalable x86 building blocks, you eliminate separate compute, storage, storage networking, data protection, and WAN-optimization appliances, which often helps to lower upfront costs and save time by avoiding system refreshes.
SimpliVity’s accelerated data efficiency features further reduce the amount of resources you need, achieving a median data efficiency ratio of 40:1 across primary and backup storage, with 10:1 data efficiency guaranteed. The screenshot below comes from Romac, an industrial products company that is using SimpliVity to host their Oracle deployment. In this screenshot, you can see 108 TB stored on just 7 TB of disk space.
The ability to use your existing servers is another area you can save in terms of cores. For very demanding OLTP workloads that require extra compute power, you can plug any almost x86 server into the SimpliVity deployment and access them as compute-only nodes. This allows you to add core resources using servers that are already in your data center, or to cost-effectively add compute as needed.
Cost reduction certainly isn’t the only reason to adopt hyperconverged infrastructure, but it’s a big benefit. And being mindful of Oracle’s licensing requirements and data protection options will help to make sure you make the right investments, no matter which infrastructure you choose.