Notwithstanding the obligatory announcements tilted to developers, this year’s VMworld – as always - was mostly about the infrastructure.
For the first two days of the event and beyond, a main conversation – in keynotes, in sessions and among attendees – was the opportunity presented by the newly available VMware Cloud (VMC) run out of Amazon’s AWS. There were questions about who is going to fill in all the missing functions for the service: Where was the data protection going to come from? Better yet, how is all the data going to get to the cloud in the first place? And how is this all going to work as more enterprises adopt multi-cloud strategies? What is notable is that, for the first time, there was consensus that all this needs to happen.
Enterprises are struggling to move to the cloud because they literally must make copies of data to do it – but doing so takes forever. Primary storage, backups, disaster recovery, historical data, geographic distribution – all of these come with demands for extra copies, which slows the march to the hybrid cloud.
Organizations simply don’t have the bandwidth to make all those copies, move them around, and manage them. While they figure out how to tackle this challenge, companies are paying for their data to just…sit.
And yet, a true hybrid cloud is the ideal option for putting costly secondary data centers firmly in the past. It is the only way to expand compute, storage capacity, and IT power for efficient, cost-effective DR – among other uses. To make this work, the conversation needs to start with getting rid of legacy data replication infrastructure.
There are multiple questions that have to be part of that discussion. These include:
- How can we improve our RPO and RTO ?
- Are we going to be exploring the cloud for other uses besides DR?
- Which clouds are we going to use? More than one?
We went to VMworld to talk about a radically different approach to enterprise storage: a consumption-based model, where enterprises pay only for the capacity they actually use. Get rid of the separate infrastructure for primary and secondary storage, for backups and DR. Get rid of the software licenses, the pain of managing disparate systems and the endless data copying.
As the industry moves decidedly toward the hybrid cloud, our network approach to elastic data storage is becoming a larger part of the conversation. A question I heard again and again at VMworld: How are you going to get your data to the AWS/VMware cloud? The answer is that enterprises must be able to talk about accessing their data from anywhere without having to create yet another copy of it.
Few other vendors can honestly claim the same, and the group that can even try is shrinking. The storage industry is going through consolidation – again. With less noise in the market, enterprises can more easily tune into the right messages and realize the promise of a hybrid, multi-cloud strategy. In Vegas last week, that goal was widespread.