10 Way to Improve Data Backup
Every aspect of the data center environment can stand a little improvement. But if your backup capabilities are like most, they are in dire need of an upgrade.
Enterprises continue to place larger portions of their backup and recovery infrastructure on the cloud. After all, it's cheaper and is said to be more resilient than in-house resources because it can be easily mirrored across ever more clouds. Many cloud-based B&R services, in fact, back up their own infrastructure on other clouds.
It seems, though, that the question is not whether the cloud can house all your critical data, but how quickly you regain access in the event of a serious failure. Of course, instant hand-over of enterprise operating environments to and from the cloud is a pipe dream. Re-establishing network pathways, resource connectivity and access to stored data takes time, particularly if the home office is out of commission.
The obvious solution is to keep track of everything that goes on in the cloud, but that is proving to be more difficult than previously thought. First off, it's important to recognize that the cloud is not a single entity. Even if you contract with a single cloud provider, there's still a good chance that your data will be spread across multiple providers. As HP Chief Architect Tara Schreider describes it, data created at one source is often transferred to another for processing and analysis, then to another for storage, another for backup and a final one for recovery. Management and visibility in this distributed network is difficult because it will most likely stretch across multiple jurisdictions and face differing legal requirements for things like retention, discovery and retrieval.
The situation is so convoluted that large numbers of organizations don't think they can quickly recover from a significant outage. Data protection firm Acronis says nearly half of industrial enterprises, namely construction and manufacturing organizations, give a no confidence vote to their recovery infrastructures, many of which have endured years of inadequate budgets. Even cloud solutions are failing to satisfy their concerns, as more than two-thirds say they are not convinced they will be able to effectively migrate data and applications across physical, virtual and cloud environments.
Ideally, all enterprise infrastructure would be managed as a single entity, so that no matter where data ends up, managers can find and retrieve it from a single user interface. We're not there yet, but the trends are pointing in that direction. Companies like TransLattice provide application and VM distribution across multiple entities while maintaining a single security and management footprint. This helps enterprises maintain continuity even if one of those domains, say a cloud provider or enterprise data center, is lost. The company recently teamed up with LynuxWorks and Fritz Technologies to develop the Secure Enterprise Cross-Domain Unified Resilient Environment (SECURE) to enable broadly distributed enterprise data environments.
Recovery is only as fast as the network infrastructure allows it to be, which is why many backup firms are working closely with WAN optimization specialists. WAN optimization-as-a-service provider Aryaka, for instance, utilizes a range of compression, deduplication and TCP optimization technologies to speed the transfer of data between on-premise and cloud infrastructures. The company claims its approach is so effective that distances between resource sets is no longer a barrier to high-speed transfer.