Verizon explains its seamless cloud upgrade except not really

In a bid to be more transparent about last weekend’s 40-hour closure of the Verizon Cloud, the company just posted more about that massive upgrade. But, to be honest, it was more a history lesson about Verizon’s cloud ambitions than any new detail about what exactly it did to its cloud.

Mostly the post recapped the message from October, 2013, when Verizon first talked up its planned enterprise-class cloud, and explained how it was built from scratch to be completely virtualized and network addressable at all levels — all the endpoints, all the orchestration, firewalls, load balancers, identity management etc. Thursday’s post was written by Verizon CTO Kevin Clarke who, by the way, stepped in for his predecessor John Considine, who left Verizon last fall in a move that has not been reported but which the company confirmed on Thursday.

What all that virtualized goodness means, according to Clarke is:

Updates and maintenance, even unscheduled maintenance like those required with the Xen hypervisor security flaw last Fall, can be performed while client virtual machines continue to run, without requiring a two-zone setup or a machine reboot. It’s a little like performing brain surgery on yourself while you’re awake. Tricky? Yes, but necessary to keep our commitment to enterprise customers

Soooo, what exactly was done over the 40 hours last weekend to make sure this is the last massive maintenance closure? No clue. We’re just told that “Seamless Upgrade was a key addition” to achieve that aforementioned vision.

To be fair, the post was called: “Why seamless upgrades matter” but in reality everyone already knows that.

[company]Verizon[/company], [company]AT&T[/company], [company]CenturyLink[/company] and other telco-rooted companies moving into cloud have some heavy lifting to do to sell to many customers who think of cloud as [company]Amazon[/company] Web Services, full stop. There is a bias among many of these customers that telephone companies, despite their networking expertise — and, face it, networking is key — are not really in this game.

For that reason alone, Verizon had better hope that its seamless upgrade, whatever it actually was, works as promised.


Verizon Cloud warns customers of 2 day (!) maintenance closure

Verizon’s enterprise cloud may go off line for 48 hours starting next Saturday at 1 a.m. EST for scheduled maintenance, notice of which got the Twitterverse in a bit of a tizzy this weekend.

A [company]Verizon[/company] spokesman confirmed that a major platform upgrade would take place starting January 10, but said the customer service rep probably overstated the amount of downtime they will experience. In a subsequent email, he said the work being done next weekend will enable future service updates to happen without impacting customers.

“We do these upgrades periodically, the last one being right before Thanksgiving and there was zero impact,” he said. In that case, all customer VMs were back up in 24 hours and most within 12 hours, he added.

Verizon had better hope that is the case because two days of downtime, even if the bulk of it happens over the weekend, is not anything any corporate account wants to experience.

Verizon, which initially got into “cloud” via its $1.4 billion acquisition of Terremark in 2011 and then last year announced a brand new written-from-scratch “enterprise cloud,” is trying to compete with Amazon Web Services and a raft of other competitors for corporate accounts. The shutdown impacts that new Verizon cloud, not the legacy Enterprise Cloud versions, another spokesman said in a follow-up email.

Planned shutdowns for cloud maintenance are not unheard of. And sometimes cloud providers have to react faster to more unforseen issues. For example:  Amazon Web Services, Rackspace, IBM and others all had to regroup in late September/early October to address a Xen hypervisor security flaw.

Note: This report was updated at 7:51 p.m. PST with additional comment from Verizon about the nature of next week’s service update and again at 12:01 p.m. PST January 5 to reflect that Verizon said its legacy cloud offerings will not be affected by next week’s action.

Google wants your Windows apps on its cloud

Despite what you may hear from the Linux-and-Mac crowds, a good chunk of today’s enterprise workloads run on Windows Server, which is why Google really wants them to also run on the Google Cloud Platform.

And now they can. Because [company]Google[/company] is now party to the Microsoft License Mobility program, existing SQL Server databases, SharePoint document repositories and Exchange Server mail can run on the Google Cloud Platform without having to cough up additional licensing fees to do so. At least that’s the case if they now run on Windows Server 2008 R2, support for which Google announced in March. But Google is also working on analogous support for the newer Windows Server 2012 and 2012 R2 releases, about which it will talk “soon,” according to this Google Cloud Platform blog post.

As for why a Windows shop would opt for Google’s cloud as opposed to, say, [company]Microsoft[/company] Azure, Google director of product management Greg DeMichillie didn’t hesitate to play the anti-lock-in card.

“Almost every enterprise intentionally wants to be multi-cloud,” DeMichillie noted. “Let’s face it, some of them got locked into on-prem licenses fees from vendors that were much bigger than they expected. Most companies will qualify two or three different cloud vendors.”

Plus, he noted, when customers look at Google Compute Engine’s local SSD storage and data center peering, they’ll see it “as a great place for enterprises to bring their apps” along with other perks like Google Firebase for mobile development and Big Query analytics.

Greg DeMichillie, director of product management, Google.

Greg DeMichillie, director of product management, Google.

Google has indeed been working overtime to portray its cloud as a good home for enterprise data and workloads. It sort of has to since [company]Amazon[/company] Web Services has an 8-year head start and Azure can parlay Microsoft’s branding and existing relationships with enterprise accounts. Google has made some inroads, especially among younger companies with Google Apps et al, but it’s still playing catch-up in big companies.

But I would agree that most large companies do not want to lock into one cloud vendor, and that may play to Google’s advantage.

For better or worse, HP has finally found its cloud religion

HP is all about the enterprise cloud and all about OpenStack, although its approach might seem very different for devotees of open source software or Amazon Web Services. Here’s how HP’s Margaret Dawson explains the company’s strategy.

Cloud app services company rPath acquired by SAS

rPath, a cloud company, is rumored to be for sale and is close to being acquired, according to our sources. The buyer being mentioned: business analytics software maker SAS Institute.

Already awash in cloud cash, Virtustream raises $15M more

Cloud computing provider Virtustream, a hot player in the enterprise cloud space, with a cloud platform designed for mission-critical and heavy-duty enterprise applications such as SAP, has raised another $15 million in investment capital. The money brings Virtustream’s total funding to $75 million.