Organizations need to meet five core requirements of the data they store; encryption, mobility, performance, retention and disaster recovery. To deliver these elements, IT is often forced to use five or more products just for on-premises data, let alone the cloud. Once cloud providers are added to the IT strategy the problem gets more complicated – The same 5 functions are needed, but they are implemented and managed differently. An organization with data on-premises and in just two cloud providers may have to manage 15 or more products to meet the core requirements of data. Datrium’s Automatrix Platform promises to simplify the multi-cloud problem by reducing it to a single solution.
The use of separate products to meet each of the core requirements of data storage not only complicates data management it also adds to the costs. Primary, retention (backup and archive) and disaster recovery often each require their own storage silo. As a result, not only do software costs increase, but so do hardware costs.
Each cloud provider delivers these services in different ways and, again with different products. Each provider also has different tiers of storage. Amazon AWS will use Elastic Block Storage (EBS) for high performance storage and Simple Storage Services (S3) for secondary use cases.
Another challenge is that each of these products and their resulting storage silos lock data into that product. Mobility between products or between on-premises storage and the cloud requires time consuming conversion, if it is possible at all.
Storage Switzerland has covered Datrium in the past. As a refresher, the platform first came to market as an on-premises Hyperconverged Infrastructure (HCI) solution, Datrium DVX. The product is unique in the HCI market because it solves the scalability and efficiency challenges of typical HCI solutions. DVX uses host-local flash-based storage for high performance but then also stores data on a centralized storage repository, which can be either flash or hard disk based. Last year Datrium delivered Cloud DVX which enabled customers to use cloud storage for backup and long term data retention.
Datrium recently announced the next stage in its evolution and is moving from a company with a product to a company with a platform. Datrium Automatrix is an autonomous data services platform. A new key component of Automatrix is ControlShift which provides workflow automation for disaster recovery (DR).
A critical difference between disaster recovery and standard recovery is the number of objects being recovered and the complexity of the recovery goals. In a typical recovery from backup, IT is usually recovering a single item (file, application, virtual machine). During a disaster there are multiple objects to be recovered, often an entire data center, and those virtual machines often need to be recovered following a runbook. They need to be started in a specific order, sometimes with pre- or post-scripting, often mapping across different local IP schemas, re-registering with destination vCenters, and so on. Recoveries may be from primary storage or, following a cyber-crime incident, from months-old backups. The interdependency is what makes DR so difficult and is the reason that most DR tests fail in some way.
As part of the Automatrix platform, Datrium delivers ControlShift, which essentially enables organizations to create a runbook, for their workloads, that automatically executes in the event of a disaster. ControlShift will automatically restart applications in the right order. Also, ControlShift is cloud based so that the loss of network connectivity doesn’t trigger an automatic fail over. Because ControlShift is deeply integrated with a converged primary and backup data platform, compliance testing can be continuous. The system tests full resource compliance every 30 minutes. In many comparable products, a recovery compliance objective (RCO) is only as frequent as a full DR test, every quarter or two.
Replication between sites or to the cloud is not disaster recovery. It is the creation of a second copy of data. Putting that second copy to use in the event of a disaster requires, in most cases, a lot of manual effort, hopefully following a runbook. Automatrix takes Datrium’s replication and Cloud DVX features to the next level by automating application restart in the right order, in a way that’s fully converged with their DVX data plane on prem or cloud.
In the first release, Automatrix will automate recovery from a second customer site, running DVX. Later this year, Datrium plans to leverage VMware Cloud on AWS to enable, with Cloud DVX, Disaster Recovery as a Service (DRaaS).
There are plenty of HCI solutions on the market, some from the hypervisor vendors themselves but most only deliver one key requirement of data storage, performance for production workloads. The customer needs to add other products to meet all the multi-cloud requirements and the concept of an automated DR is out of reach for most. Datrium’s Automatrix platform checks all the multi-cloud data requirement checkboxes and with ControlShift provides automation to enable organizations to have complete confidence in their ability to recover from a disaster.