Managing the complexity to harness the power of multi-cloud disaster recovery
By Johan Scheepers, Country Head for South Africa at Commvault
As organisations send more and more data into the cloud and adopt multi-cloud strategies with solutions and services from a wide range of providers, the challenge of managing these environments effectively increases. Disaster recovery (DR) in particular needs to be carefully considered. While multi-cloud offers unprecedented levels of flexibility for DR, it also has the potential to vastly increase complexity. Organisations need to plan their strategy carefully and partner with a provider who can help to automate workflows and deliver the speed, agility and granularity necessary for DR in multi-cloud environments.
According to poll results from a Commvault webinar on Cloud Disaster Recovery, many organisations (33.6%) have multiple tiers of Recovery Time Objectives (RTOs). The majority of respondents (36.4%) would prefer recovery time of four hours or less. However, in reality, recovery processes seldom go smoothly following an actual disaster. In order to achieve such ambitious RTOs, a DR solution must provide numerous automated features.
Many organisations (37.5%) also have multiple tiers of Recovery Point Objectives (RPO), reflecting that it is important to backup and replicate frequently as businesses simply cannot afford data loss. Flexibility is essential to meet these increasingly strict RTOs and RPOs, ensuring you can recover what you need, where you need to recover it and when you need to recover it.
In a multi-cloud environment, organisations can take advantage of the ability to back up and replicate data across any number of different cloud vendors. Using multiple clouds for data protection also means that you can avoid ‘putting all of your eggs in one basket’ so to speak. It gives the organisation the ability to back up one cloud to another, between clouds, across clouds or within clouds. Multi-cloud DR is therefore incredibly flexible. In today’s business environment where data is a mission critical asset, this flexibility has become essential to meet increasingly strict data compliance rules, combat ransomware and maintain business continuity.
However, it is the very nature of this flexibility that can make multi-cloud DR difficult to manage. Without effective data management, this scenario could become unmanageably complex. Many businesses become overwhelmed when it comes to the multi-cloud and do not have a clear picture of how to recover data. Unfortunately, this only becomes apparent when something goes wrong.
It is also important to consider that recovering data is not just about a ‘disaster’. From lost or accidentally deleted files to ransomware, natural disasters and even internal threat factors, there are many things that can happen to your data. Organisations need granular recovery to the point that a single file can be recovered quickly and in isolation without the need to recover an entire system or virtual machine.
Certain elements such as policy-based automation, Artificial Intelligence (AI) and other tools that watch for anomalies in the data infrastructure become critical to a successful multi-cloud DR plan. This is true for all levels of data recovery, from lost files to system recovery to full on disasters – the underlying technology and processes required are consistent.
Multi-cloud DR needs to start with the business goal in mind. It centres on strategy around data, including where it is kept, who has access to it, what dependencies exist around it and how it needs to be protected. It is not just about a solution or a piece of software, although this is a critical component in getting it right. Companies need a technology partner that understands the nature of their business’s data and what can go wrong. Moreover, organisations need a technology partner that can deliver and implement a solution that enables the business to leverage omnidirectional data recovery, from the cloud, to the cloud, in the cloud, across the cloud and between cloud regions.
Recent Computing News
CIOs need to save… and lead28 Sep