Up, Uptime & Away: Hybrid Cloud Strategies To Improve Reliability
A hybrid cloud deployment can help you improve your IT infrastructure’s reliability and increase uptime.
No matter the focus of your infrastructure, whether it’s supporting documentation and records, storing and supplying media, a public facing web application, or hosting enterprise software like collaboration and enterprise resource planning platforms, cloud hosting remains a solid way to increase service levels.
Enterprises are increasingly turning to hybrid cloud models in order to take advantage of the on-demand, scalable computing power and bolt-on disaster recovery and backup features enabled by cloud providers. One recent EMC survey found that 88% of respondents believe hybrid is “important” or “critical” to their digital business transformation. Digital business transformation was defined as increased agility, data-based decision making, and more responsive services for customers and employees.
Here are a few ways that a hybrid cloud deployment can help you improve your IT infrastructure’s reliability and increase uptime.
Mix and Match On-Premise and Cloud
Combining your existing data center assets with cloud resources allows you to supplement in-house infrastructure as needed. If you have a new application rolling out or anticipate an increase in demand, you can add some virtual machines to accommodate. You might run out of storage on-site and increase it in the cloud; or instead of taking a risk on an emerging technology or one that requires a large upfront investment, you can test it with a cloud provider first.
One example is cloud storage. An Aberdeen Group study from 2013 found that downtime costs were reduced an average of 15% when cloud storage was implemented. A health care organization might have an on-premise data center that is running out of storage space. With new equipment arriving soon that generates very large images and data files, they need to expand. So they turn to cloud object storage, linking it to their onsite systems so doctors can pull up records and media from any medical facility. One day, the power is cut at the hospital data center site, but employees are still able to access this new storage system.
Cloud Disaster Recovery and Backup
It’s common for disaster recovery and backup solutions to recommend geographically diverse sites for storing backup data and configurations. Critical applications should be spread across availability zones in case of disaster, and multiple vendors can also help avoid an outage from a single site.
Disaster recovery can be configured to automatically failover in the case of a major facility outage, so your employees are back up and working in no time. Keeping cold sites, in which the virtual machines are only powered on when necessary, can minimize disaster recovery costs or you can keep downtime to practically nothing by operating a hot-site failover, which is always ready.
Contract and SLA Assurances
Whether you’re moving a single application or your entire infrastructure into the cloud, your provider should offer a 100% service-level agreement (SLA), which basically guarantees constant uptime. In the event of an outage, you are compensated. Make sure to read any contract carefully so you know the SLA details well.
Many provider data centers are constructed to higher standards than the average enterprise or mid-market facility and can endure longer power outages, the loss of more vital equipment, and more extreme conditions. Do your research and find a provider with a reliable track record. You might be compensated under an SLA, but compensation only does so much good if your cloud backups fail when they were supposed to failover.
Supplement Your Staff, Simplify Your Operations
Cloud resources can also help your IT staff be more productive, which itself leads to greater uptime across the entire infrastructure spectrum. A 2013 case study from Cisco and NetApp reported a 166% ROI, and IT capital and operational savings of $1.27 million over three years; cloud resources saved money on hardware procurement, space, energy, and administration. One employee reported that an update that would have previously taken five or six employees four days to complete now took one person only six hours.
Simplified upgrades, testing, and moves to production make it easier to fully test and roll out essential updates to operating systems, applications, anti-virus/anti-malware, load balancing, and network monitoring tools that are necessary to maintain fully functional systems. When you can clone a virtual machine or apply a new configuration to an entire batch of servers, the time to complete updates is greatly reduced.
Many cloud providers also keep all or some of the cloud platform up-to-date without any work required by your staff. This may or may not require an additional service contract, but at the very least the virtualization platform and everything beneath it (hardware, networking, etc.) is guaranteed under your SLA.
It’s clear from the EMC survey and other headlines coming out every day that hybrid infrastructure is a big part of the future of enterprise computing. There’s no need to go all-in on cloud immediately unless you are in dire need of an infrastructure upgrade or have a large project coming up that requires you to reevaluate your data center. But adding some cloud resources is the perfect way to increase uptime while getting your feet wet.
This article was originally posted “Up, Uptime & Away: Hybrid Cloud Strategies To Improve Reliability” from Cloud Strategy Magazine.