And so there emerges another cloud buzzword. You too can have fun with business buzzwords by playing with the Wall St. Journal’s Business Buzzword Generator here.
Hi Google, welcome to the Enterprise Cloud party… we’ve been expecting you 🙂
It was an exciting week for our friends in Mt. View, I’m sure, as the new head of Google Cloud, Diane Greene (former VMware Founder and CEO) and her team took the wraps off of Google’s enterprise ambitions and strategy. As you might expect, the IT pundits and editors returned to their corners to remind everyone that it’s still very early days for cloud computing, and that Google has as good of a chance as anyone at winning the prize. The problem is, I feel we’re having the wrong discussion.
Today I submit to you that the storyline around #cloud #winning is a bit flawed and the discussion should really be around what federation of clouds will win.
Multi-Cloud. Looks and sounds a lot like ‘hybrid cloud’, right? Yes and no. Hybrid cloud is a concept that is largely focused on bridging legacy data center models from on-prem to an IaaS service and ideally enabling applications and data to easily and securely flow from one data center to any other data center. Once an organization’s hybrid cloud journey ends up with all of their applications in the cloud (all in), what do we call this steady state mode of operation?
Search Google for the term “AWS all-in” and you’ll find some amazing stories of pioneering companies who are aggressively embracing the cloud. GE and CapitalOne made waves at the last major AWS conference. Netflix, Intuit, Notre Dame and (CTERA customer) Live Nation have made similar commitments. What’s more… the all-in principle doesn’t exist exclusively for AWS, but as the leader in the space they are certainly a lightning rod in this discussion.
With all the talk of “all-in”, why does multi-cloud matter? Well, the answer to this question is a combination of preserving optionality and business continuity. When making the choice to go “all cloud”, it’s important for organizations to understand their alternatives as the market evolves, and position themselves to extract the greatest business value from cloud platforms as their business requirements evolve. Today, a multi-cloud strategy that has a mindfulness of multiple cloud providers is an essential component of jumpstarting any modern enterprise cloud initiative. Here are the key considerations.
#1 Price. As always, cash is king.
With an increase in market entrants comes price competition, all of this serves the purpose of keeping your vendors honest. In this case, it’s important to understand a) competitive pricing for services you consume, as well as b) differences in technology approaches and offerings that may result in fundamentally different pricing discussions.
For example, a customer may move to Microsoft Azure because they are bundling cloud credits with their Office 365 sale. At face value this may seem attractive, but if you are storage centric, it’s important to understand that Amazon offers additional storage tiers, such as S3 Standard -Infrequent Access (which CTERA now supports) which can be 60% less than the cost of Azure object services.
In essence, I can see tomorrow’s CIO strategy looking much like today’s, when you can go into any enterprise data center and find it nearly all full of Vendor A‘s equipment, only to find a little bit of Vendor B‘s products there too, purchased by the customer to keep Vendor A honest. In the future, organizations will ensure multi-cloud price optionality by:
- deploying applications that are portable across multiple cloud platforms – like CTERA
- looking beyond initial incentive pricing to develop a long-range TCO model
- fully understand the cost of migration to know when it makes sense to move, and when it doesn’t
- and finally, keeping relationships and some amount of applications hosted with all relevant cloud providers, only to ensure that if you decide to migrate applications and data that your team is familiar with their platform and can easily migrate
#2 Business Continuity. When you’ve moved your applications to the same cloud data center you used to use for offsite protection of your private data center, where is your new “offsite”?
As organizations think about moving all in on the cloud, data protection becomes a complicated and critical component of a risk management strategy. Unlike the days of yore where you could do a VM-level image backup to a deduplication appliance, the game has totally changed as organizations look to protect servers running in the cloud. And while the cloud providers offer some level of backup services to protect cloud servers and applications, these services will typically lock vendors into a specific cloud platform – limiting portability, price optionality and the ability to recover to another cloud in the case of a truly catastrophic event.
In the future, we expect enterprises will ensure business continuity for applications running on cloud IaaS by:
- choosing a backup platform that supports a variety of cloud storage APIs (S3, Azure, SWIFT, etc.) so that it’s easy to backup to any IaaS or on-prem object storage of choice
- optimizing backups at the source using granular backup tools, deduplication and compression to ensure it’s possible to backup and recover across wide area networks
- implement cross-cloud replication of backups to enable near-instant recovery after any major cloud outage via a secondary cloud provider (without requiring expensive application-level live replication and CDP tools)
I hope this helps get the juices flowing around the “all-in” era. With only 10% of the market in public cloud IaaS today, there’s certainly a lot to consider as the market matures, ebbs and flows. The future of cloud is bright, but let’s stop talking about “winner take all” and let’s start talking about how enterprises can intelligently plan to be the real winners in this race by using the right clouds for the right jobs.
What will your cloud fabric of the future look like?