Users demand a seamless digital experience. When something goes wrong they expect their data and applications to quickly return, preferably without them ever knowing something went wrong in the first place. For IT this seamless expectation means data has to be more than protected, it has to be available and ready for use. VeeamOn 2017 focused on meeting these new user expectations and solving the data protection problem.
The Data Protection Problem
IT looks at data protection as a scheduled task that occurs once per day. That protected data is stored in a format that conserves disk capacity and organizes it so administrators can find the data later.
The first problem with this classic understanding of data protection is data changes so quickly that only protecting it once per day leaves too much data at risk. Compounding this problem is the fact that there is seldom a “hard copy” from which to re-key data.
Consider order processing. In the 90’s and even early 2000’s orders came in via fax machine or were at least printed after a user keyed it into an order entry system. Now most orders are digital from the start, there is no hard copy to fall back on. If that order, then, is after its creation, but before the once-a-day backup, there’s no record of that file anywhere.
The second problem is the time it takes to re-position data after a failure. The primary backup target for most backup software solutions is disk-based. Typically that disk-based target is adding compression and deduplication, once again to drive down costs. If there is a failure, data needs to be extracted from the backup format, and then sent across the network to the new source location.
The third problem is data is everywhere. While virtualization dominates the enterprise, there are often a group of applications based on MS-SQL or Oracle that are on bare metal servers or in their own application-specific cluster. The data protection process has to protect virtualized and non-virtualized systems.
In addition, data is no longer uniquely in the data center. As organizations continue their cloud embrace, they are creating unique data in the cloud. Services like Office 365, as well as custom applications running in clouds like Amazon Web Services (AWS), all need protection.
The Availability Solution
IT needs to re-think data protection. Capturing a copy of data and storing it somewhere is no longer enough. Data needs to be copied much more frequently, every few hours for some data sets and continuously for others. IT also needs to consider how quickly it can return that data to users and applications if something goes wrong in production. In some cases, this means having a ready-to-go standby environment. In others it means being able to serve the data up directly from the backup storage device, saving network transfer times. Finally, it means protecting data uniquely in the cloud, either to an on-premises copy or to another cloud. The data center needs to be prepared to be a host to cloud-hosted applications.
Greater Target Flexibility
Most data protection solutions are solely focused on disk as their backup target and a few are adding a form of cloud connectivity. The data center needs more than that.
Object storage is becoming the secondary storage system of choice for many data centers. It acts as their archive and as the repository for IoT data storage. As we discuss in our recent article, “Does Object Storage Make Sense for Backups?” these systems also make ideal backup storage solutions. Most object storage systems support the S3 protocol and so modern backup applications need to support S3 connectivity.
Finally, there is tape. While experts repeatedly pronounce that tape as a backup target is dead. It is still here and is experiencing a resurgence. It is cost-effective, has very low power requirements and is disconnected, which, given the threat of ransomware and other malware, is of high value.
At VeeamOn 2017, Veeam Software started to introduce to the public what will eventually be Veeam Availability Suite v10. It will decrease the gap between protection events by introducing continuous data protection leveraging VMware’s IO filtering capability. The software will also extend its snapshot management capabilities to IBM and INFINIDAT storage systems. These capabilities will build on Veeam’s Instant Recovery capability that eliminates network transfers during restoration.
Veeam Availability v10 also extends Veeam’s cloud support to include expanded Office 365 support, AWS application protection and can leverage Microsoft Azure for DRaaS. AWS protection is delivered through a Veeam – N2WS strategic partnership. It offers a cloud-native, agentless data protection solution designed to protect and recover AWS applications and data. The solution mitigates the risk of losing access to the organization’s applications and ensures protection of their AWS data against accidental deletion, malicious activity and outages.
V10 also expands the support backup storage targets and will have native support for S3 protocol, which means most object storage systems will work as a storage target for Veeam backups. Veeam also announced a partnership with StarWind Software which created the “StarWind Cloud VTL for AWS and Veeam.” The solution offers cost-effective and scalable tape replacement by leveraging Amazon S3 and Glacier object storage, helping businesses meet regulatory requirements for data retention with no changes to the established tape-centric data archival processes.
Veeam has evolved from a niche VMware data protection product to an enterprise availability solution, systematically adding features and capabilities required to be considered enterprise. But it has done more than cover its bases, it is a leader in extending the availability status quo. Amazingly it made all this progress while keeping its customers happy.