The cloud makes comprehensive protection possible for every VM
- 20 December, 2013 22:54
Server virtualization complicates disaster recovery (DR) because the complexity and cost makes it impossible to extend rapid recovery to every virtual machine (VM), and yet, when even supposedly non-critical VMs go down, businesses lose money. But today, the cloud is making those hard DR choices history. In their place, enterprises are finding options that are dramatically better than traditional backup, but with comparable price tags.
By leveraging resources in an elastic cloud in conjunction with the latest platform and workload transformation technologies, data centers get rapid operational recovery, near-infinite capacity, seamless management and easy migration back on-premise ¯ at a price that everyone can afford.
Until recently, there were four choices for data protection in VMware vSphere environments, with four widely divergent price tags and recovery capabilities:
* The first is basic backup, the most affordable and ubiquitous approach to data protection. A traditional backup tool will back up VMs to a target store on-premise or in the cloud, but the data store needs to be carefully managed, since storage infrastructure is often expensive and therefore capacity is usually limited, especially on-premise. Most of these tools offer deduplication in order to make the most of available storage, but they don't generally self-test, and VMs might not be restorable if manual testing isn't done well. Two other significant drawbacks to these tools are that administrators need to learn another management console to use them, and fully restoring operations from backup can take a long time many hours or even days depending on the environment.
* Beyond basic backup, replication tools have offered data centers the means to replicate data from one virtualization cluster to another and house copies at an offsite location, at a hot site, on a rack of computers or in the cloud. For this to work, the secondary site has to be more or less identical to the primary one, down to the storage hardware, as well as the compute, memory, network and VMware vCenter server. The network configuration settings also need to be automated. All of these requirements add up to significant cost, as well as preparation and ongoing testing and management. The operational restoration time is faster than backup, though. It can be as fast as 15 minutes or as long as several hours depending on the execution and availability of capacity to replace the out-of-commission hardware.
* Near-continuous data protection, sometimes called replication on steroids, offers greater speed of restoration (for greater cost, of course). However, the major obstacle here is the need for redundant infrastructure and computing capacity at a remote location, which requires an even more significant investment in hardware, software and networking at the mirror site. Companies also need to acquire specialized skills to manage the mirror site and failover in the event of on-premise downtime.
* Continuous data protection (CDP) has delivered the most robust capabilities. Given the extremely high cost, most data centers reserve CDP for mission-critical applications. To make CDP work, users have needed the mirror site, as well as very low-latency networking and hardware. The data protected by CDP is generally of extreme importance to the business, and so these solutions have demanded careful management, monitoring and testing. The benefits include operational restorations that are complete in seconds.
These traditional options for VMware backup, restoration, near-continuous data protection and continuous data protection have several obstacles in common. They require standby capacity, special skills and knowledge, maintenance, management, testing and consistent VMware vSphere environments for recovery. All of that adds up to significant total cost to recover and continue operations after a failure.
Most VMware shops simply can't afford to have extensive ready spare capacity and specialized data protection personnel in their budgets, so for the majority of VMware installations, traditional data protection just isn't cutting it for the mainstream.
In pursuit of protection for every VM
The expense of traditional backup solutions is, of course, overshadowed by the expense of non-functioning servers. When hardware fails, the fix is expensive, but when engineers and other teams can't operate, the productivity losses can be enormous. CDP closes the gap for mission-critical workloads, but that doesn't describe the majority of VMs in operation today.
So where is the better-than-backup solution for non-critical production systems or test and development workloads? Data center leaders have long sought something better than backup only for the same price, which would also be easy to implement, operate, test and manage. The availability of backup combined with rapid recovery in the cloud is finally answering this need, and it's changing the data protection landscape.
Data protection today can span backup, replication, DR and BC for every VMware workload, not just the business-critical ones. This is finally possible because the resources of the elastic cloud, such as Amazon Web Services (AWS), offer organizations consumption-based pricing and near infinite capacity.
Because the costs in the cloud are so much lower than that of on-premise spare capacity, the overall expense of disaster recovery and business continuity is falling, and the capability for data restoration in minutes is now within reach for day-to-day workloads. Seamless interoperability between VMware on-premises and AWS off-premises is enabled through new transformation technologies that enable workload portability and extend VMware vCenter management to AWS ¯ a solution to the thorny problems that had the industry settling for basic backup and slow recovery for so long. Enterprises can take an investment that previously only covered backup and replication and apply it to also deliver disaster recovery and business continuity, creating a holistic, end-to-end solution.
In the past, the high price tag of DR and BC mirror sites meant comprehensive data protection was only available for an elite class of applications and enterprises. The future of data protection is one in which DR and BC tools are easy to implement without professional services, simple to test without time-consuming processes, intuitive to use without specialized training, and easily manageable, all for the same price as the traditional backup and replication tools. By leveraging the latest platform and workload transformation technologies combined with pay-as-you-go public cloud services, we are finally seeing the democratization of DR and BC ¯ ensuring business continuity for all workloads, not just the most critical ones.
LeBlanc, CEO and founder of HotLink Corporation, has more than 25 years of enterprise software and technology experience at both Fortune 500 companies and Silicon Valley start-ups. Prior to founding HotLink, LeBlanc was founder and CEO of FastScale Technology, an enterprise software company acquired by VMware, Inc.
Read more about data center in Network World's Data Center section.
Join the Computerworld Australia group on Linkedin. The group is open to IT Directors, IT Managers, Infrastructure Managers, Network Managers, Security Managers, Communications Managers.
Optus goes over the top with VoIP service
Turnbull asks how the NBN got that way
U.S. retailers insist on PIN requirement in smartcard rules
Yelp speeds database access with flash storage
Thanks a million, Drupal