Physical and virtual servers should live by the same business continuity rules
Virtualization provides tremendous benefits for business continuity. The hardware-agnostic nature of virtualization makes it much easier to quickly recover virtual machines, and capabilities like virtual machine migration and high availability have significantly reduced infrastructure complexity. At the same time, virtualization puts pressure on business continuity in two ways:
- The dynamic nature of virtualization makes it much harder to keep Disaster Recovery plans up-to-date, putting critical operations at risk.
- Many key application components still run on physical servers, but business continuity for these needs to be managed separately.
Virtualization can create gaps in business continuity planning and operations, particularly for enterprise applications. Fortunately, Neverfail can extend the scope of business continuity functionality offered by VMware’s continuity features and products. Neverfail has deep and ongoing experience of collaboration with VMware and Neverfail’s technology is at the core of VMware vCenter Server Heartbeat, VMware’s continuous availability solution for VMware vCenter Server. This experience has afforded Neverfail unparalleled insight into opportunities to deliver solutions that enhance business continuity for customers running VMware vSphere and other products.
Extending VMware Site Recovery Manager to Physical Servers
VMware Site Recovery Manager (SRM) has become one of VMware’s most popular products - and with good reason. When it comes to protecting virtualized applications, SRM has dramatically reduced the complexity and cost of implementing a reliable disaster recovery plan. Indeed, many customers now wish to standardize on SRM for all business-critical workloads. However, SRM can only protect applications running on vSphere, and while the vSphere footprint continues to grow, some categories of applications (or components of them) remain running on physical platforms for a variety of reasons. These could include vendor support, cost, performance or internal resistance from application owners. Consequently, the cost and complexity of business continuity can increase as you maintain legacy Disaster Recovery strategies for your physical servers in parallel with your virtual infrastructure. And for most customers, the server landscape will remain heterogeneous across physical and virtual and, increasingly, across multiple hypervisors for the foreseeable future.
Neverfail solves the problems posed by a heterogeneous server landscape by seamlessly integrating physical servers within SRM Recovery Plans. By leveraging Neverfail’s technology, you can now standardize on SRM to protect your entire server inventory, not just applications deployed on vSphere. Business continuity plans that once required several separate solutions and consoles can now be managed directly through vSphere client leveraging SRM as the single point of management and control. Neverfail’s vSphere management plug-in allows your administrators to manage the failover of physical servers directly from within the vSphere Client. With Neverfail’s SRM integration, administrators can then insert failover steps to protect any non-vSphere server directly within the SRM run-list.
When SRM Recovery Plans are triggered, Neverfail’s runtime infrastructure seamlessly fails over non-vSphere servers alongside the vSphere infrastructure. Neverfail also offers industry-leading recovery time and recovery point objectives, allowing customers to meet the stringent availability targets associated with business-critical applications. Tier-1 applications that span physical and virtual servers are quickly returned to service without further intervention. Neverfail also supports SRM’s run-list test mode with zero disruption to the active primary server.
Leveraging VMware vSphere HA to Protect Applications
VMware vSphere HA has become the most widely deployed solution for protecting applications from downtime caused by local host server hardware or operating system failures. It automatically restarts affected VMs on a different host server within a virtual cluster. Like SRM, its strength lies in its ease-of-use; with a single mouse click, any server can be made highly available and protected within a virtual cluster. While VMware HA provides complete protection from host server failures, it can’t detect problems with applications running on those VMs. This is the key limiting factor for customers seeking to protect critical workloads since many threats that cause downtime are due to application defects or operator error. If VMware HA were able to extend its fault detection to identify sick or failed applications and then restart them on another server within the virtual cluster, you’d be able to provide more complete protection in the datacenter. This is where Neverfail comes in.
Neverfail deploys a broad range of instruments within the VM to monitor the health of your critical applications. In real time it determines whether they are running correctly and measures their consumption of system-level resources and application-level behaviors such as response times, service availability and database availability. Monitored parameters are interpreted by a rules-based model that identifies the failure signature of any application component before it actually fails. When application-level failure events are detected, Neverfail calls the VMware HA APIs to initiate failover within the virtual cluster.
The combination of VMware HA and Neverfail delivers protection from the broadest range of threats in order to keep your critical applications continuously available. And like Neverfail’s SRM integration, it is managed from right within your vSphere client for a single point of management and control.
Neverfail believes that true continuous availability can only be delivered through the judicious and appropriate use of monitoring, detection and failover automation so that application issues can be resolved so rapidly that users are unaware that a problem even occurred. Any alternative solution that promises to recover from a disaster has, by definition, already failed to eliminate user downtime. This is the essential difference between reactive recovery-based solutions and proactive continuous availability from Neverfail. It’s also behind the decision by industry technology leaders, such as VMware and Solarwinds to license Neverfail technology to protect their customer’s critical infrastructure.
“We chose Neverfail for continuous application availability because it is the only solution on the market that protects our key business applications with simple one-click failover and failback between sites.”