Yesterday’s problems with Amazon's Web Services facility in Virginia, US, causing S3 cloud storage ‘challenges’, has highlighted the need for resilience in data storage, particularly for core, business impacting data.
Let’s face it, failures such as these are not uncommon across the IT industry – so you need to design to avoid the business impact of IT outages, and respond well if and when the worst happens.
The old adage of not putting all your eggs in one basket remains true. As all of our businesses become data dependent, it’s just common sense to adopt an appropriate IT strategy, to reduce dependency on single points of failure and provide business continuity in the event of problems.
Hybrid IT is a great platform to deliver resilience. I’m not saying we never have issues – no software company can say that. But I am delighted that every single one of our Hybrid IT customers were unaffected by this S3 production outage.
This is because:
- Our product is built hybrid first with resilience in mind – if cloud fails you continue locally, if local fails you can continue in the cloud. Even our AWS hosted management system has a locally usable version.
- S3 buckets used for some of our services in that facility are replicated elsewhere – we didn’t lose services, we just had less redundancy for a while – and our Customer Success team were on that.