Verizon Cloud Fallout: What IS an Enterprise-Grade Solution, Anyway?
Verizon recently made waves in the enterprise cloud world when it announced a 48-hour weekend shutdown of its Verizon Cloud service for maintenance. This set off a storm of criticism from analysts and customers, who were left without access to anything stored in or running on the Verizon Cloud platform – email, apps, data, customer portals. Needless to say, that’s never a good situation to be in, especially for consumer-facing businesses that don’t close shop over the weekend.
The actual outage lasted around 40 hours, and Verizon says that it enabled them to make any further updates without disruption. That’s good news for their customers going forward. But the whole affair had observers questioning if Verizon could truly call themselves an “enterprise-grade” cloud solution while scheduling service outages – and raising larger questions about what having an enterprise-grade solution actually means.
“Enterprise-Grade”: Not Just A Marketing Bullet
We can’t speak for other companies, but here’s what we mean when we call ourselves an “enterprise-grade” solution at Zetta.
We’re there when our customers need us. Business never stops, and neither should we. No service can be up and running flawlessly for every second of each and every day – the nature of technology is that yes, things will occasionally go wrong. An enterprise-grade solution should be architected with that in mind, and designed so that a hiccup in one part of the service can be compensated for without affecting the user.
This means investing in quality infrastructure and engineering. We use Tier-4 datacenters, collocated with industry leader Equinix, to ensure the solution we offer our customers can actually deliver the level of service they need. In our case that means customers being able to complete their backups and, more importantly, restore their backups when they need them. Every time. We have a 100% recovery rate – we’ve never had a customer unable to get their data back when it counted.
We’re fast enough to get the job done. A key element for a enterprise-grade cloud backup and restore solution is performance. It’s easy to think of the difference between consumer and enterprise cloud backup needs as primarily a matter of scale. But besides just having an order of magnitude more data than the average consumer, businesses have much stricter timeframes around getting it back.
For most consumers, there’s no firm time requirement for backup…and it often doesn’t get done all that frequently. Likewise, recovery is generally not tied to a specific timeframe. If it takes a few days to get those cat pictures back, the world won’t end.
For business, on the other hand, downtime and data loss can be brutal, with incidents costing thousands of dollars in lost orders, productivity and goodwill. When a data loss incident occurs (and they happen to everyone at some point), it’s vital to get things recovered and running again as soon as possible to prevent further damage. For that reason, businesses need frequent and regular backups, ideally at least once a day, to ensure their data is protected in a disaster. Thus an enterprise-grade cloud backup solution needs to be able to transfer enough data quickly enough to make daily backups of a potentially large dataset, and to restore those backups in a hurry when disaster strikes.
For us, that meant designing our service around the realities of transferring data over the Internet. Thanks to a series of WAN optimization techniques, parallelization and Internet-first architecture, we can handle enterprise-grade data volumes without leaning on an appliance.
We’re serious about customer security. Obviously, data is valuable. More importantly, everyone KNOWS it’s valuable – which is why you couldn’t go more than a few hours in 2014 without hearing about another high-profile data breach. Securing customer data is a big responsibility, and an enterprise-grade solution needs to treat it as such.
For cloud backup, this means high standards of encryption and physical security. Zetta is fully compliant with HIPAA and ITAR, which means we follow strict regulations on how we handle and secure our data. We use SSL encryption for data in flight and AES encryption at rest, and each customer has a unique encryption key. We’re also audited according to the SSAE-16 standard, which is a fancy way of saying we get an external party to double-check that we actually do what we say we do (because “we’re, like, totally secure, take our word for it!” doesn’t exactly inspire confidence).
So what does “enterprise-grade” actually mean? If you ask ten people, the odds are you’ll get ten answers, but we think it’s actually pretty straightforward. Being enterprise-grade means being ready and able to be the partner that businesses need, full stop. And THAT means not leaving them to scramble for the fallout shelter.