IT professionals have become increasingly comfortable with the cloud’s ability to deliver scalable performance. But concerns about the cloud’s ability to provide satisfactory high availability (HA) and disaster recovery (DR) protections have kept many applications on premises, where the staff enjoys full control over the environment. Three technology trends in 2019 will improve the suitability of the cloud for even the most critical of applications, including those like SAP for supply chain management and enterprise resource planning.
Broad Adoption of Mission-critical HA Clustering in the Cloud
IT departments have come to depend on purpose-built, third-party failover clustering technology to protect their mission-critical applications in the enterprise datacenter. These software-only HA/DR solutions include, at a minimum, real-time data replication, continuous monitoring capable of detecting failures at the application level, and configurable policies for failover and failback. In 2019, their platform- and application-agnostic designs will evolve to provide more seamless deployments in hybrid and public cloud environments.
Cloud service providers (CSPs) will also be enhancing their basic uptime provisions to become more effective and affordable for a broad range of enterprise applications, with the primary focus being on those that can tolerate some downtime and data loss. To achieve the desired service levels for the most critical applications that require less than an hour downtime per year with no loss of data, CSPs will be partnering with the failover clustering vendors.
Cost Savings from Dynamic Allocation of Redundant Resources
On-demand provisioning in the cloud is nothing new. These capabilities will be enhanced in 2019 with the emergence of new orchestration policies that enable redundant resources to also be provisioned automatically on-demand, which will make achieving continuity in the cloud more affordable for a broader set of applications.
Both HA and DR require redundancy to ensure rapid recovery from failures. Traditional failover clusters replicate the full operational environment of the primary VM—CPU, memory and storage resources, along with all data—to one or more secondary VM(s). The need to fully provision fully-redundant (and very rarely used) resources can double the cost of HA in the cloud, which is difficult to cost-justify for many applications.
To remedy this situation, failover cluster management techniques are being enhanced to orchestrate recoveries through the dynamic allocation of fully-provisioned resources at the time of failure. With this approach, the standby VM(s), while operating in a standby mode, get configured with only the resources needed to handle the minimalist role of a data replication target for the primary VM. When a failure occurs, the cluster immediately and dynamically reconfigures an available standby VM with the full complement of resources needed to deliver the level of performance required for its fully operational role of the primary VM. This dynamic allocation will enable HA and DR protections to benefit from significant cost savings with negligible impact on uptime or performance.
Automation Accelerates Reliable Deployment with “Quick-start” Templates
Properly configuring—and testing—HA/DR provisions can require time-consuming and error-prone manual processes owing to the inherent complexity of the different systems, services and software involved. And just a single mistake can result in the failover provisions failing when needed.
Beginning in 2019, CSPs and third-party providers will greatly expand the availability of quick-start templates that simplify and accelerate the deployment of properly-configured failover clusters without any need for highly trained specialists. These templates are wizard-based interfaces that employ automated scripts to provision, configure and orchestrate the resources and services needed to protect specific applications.
The use of templates will substantially increase IT staff productivity by minimizing the time and effort—and talent—it takes to deploy and adequately test reliable HA/DR provisions. And greater confidence in being able to get the configurations right the first time will help overcome concerns about using the cloud for mission-critical applications.
With these three enhancements, built atop an already-solid foundation, 2019 may well be the year the cloud achieves that long-anticipated tipping point where it becomes the preferred platform for a majority of applications for a majority of organizations.