CIOs looking to harness the power of talent retention should look no further than Apple’s low-paid workers. Wait, something doesn’t sound right about that plan.
Got a BlackBerry? You’re quickly becoming an endangered user.
Did you think Dell had slowed down with the buyouts? Silly reader. Dell is acquiring Quest Software this week.
CIOs got heart, too. Check out the CIO Scholarship Fund event to help future IT leaders with tuition.
The leap second caused outages on many sites over the weekend, like Reddit, BuzzFeed, Gawker, LinkedIn and Yelp, plus scores of Linux sites. Bad news for lovers of cloud computing services. You’ll be happy to know that the TechTarget network was just fine.
And if the leap second wasn’t bad enough, real storm clouds knocked out Amazon’s data center, wiping swaths of the Internet’s cloud computing services with it, including Netflix, Instagram and Pinterest.
At this month’s SearchCIO360 dinner, we met Seth Kutty, director of IT at Tesla Motors, one of the ten companies that are changing the world. Don’t miss our upcoming dinners in Chicago and Boston!
There’s no grappling over IT dollars for DR for a given project. The cost of DR is simply slipped in as part of the monthly subscription fee, whether the service is for a hosted application or hosted infrastructure.
Not too shabby. And, as far as cloud proponents are concerned, the providers are going to make sure that disaster recovery is a priority because their business models and reputations are built up (or ruined) if they don’t get DR right.
That’s a pretty big incentive for cloud providers, so it makes sense that their DR capabilities are going to be better than their customer’s.
That’s one way of looking at it. Another is that disaster recovery in the cloud could hurt — rather than help — your DR plans.
Some IT executives are ground-shipping their data to their cloud provider because they don’t want their data on a public network. They’re also afraid that, once the cloud provider has their data, they may lose access to it for a variety of reasons: the cloud provider’s employees could mishandle the data, or the company could experience a service disruption or even go out of business. Cloud provider outages do not exactly make a case for disaster recovery in the cloud.
According to a blog post on ReadWrite Cloud on the Top 5 cloud outages, “Mark Williams, a cloud computing consultant based out of the United Kingdom, found 23 reports of cloud computing failure in 2010. Google had 12 outages. Amazon had five. He reported that Microsoft had four outages. Salesforce.com had two.”
Laura Smith, features writer for SearchCIO.com, recently wrote about the top 10 public cloud risks, and while security was number one, availability made the list, too.
“It’s all about quality, not about low-cost services anymore,” Lalitendu Panda, global CIO of D&M Holdings Inc., pointed out in Smith’s story. “Interruption of service is an issue; we have had a couple of situations. It’s not like having your own [infrastructure] that you can modify. You have no control over what else is running on the cloud that could degrade performance.”
If the company can afford to lose access to an application, it would seem that disaster recovery in the cloud is a moot point. As cloud providers push to accommodate more mission-critical applications, enterprises will inevitably lose access at some point.
But that happens anyway in some organizations. So maybe we should just call it a draw.]]>
As more midmarket firms explore virtualization to achieve cost savings and other benefits, they’re discovering that adequate bandwidth and compliance with regulations as part of their disaster recovery solutions are drivers for success.
“You never know when a disaster is going to hit,” says Ray Lucchesi, president of Silverton Consulting Inc. in Broomfield, Colo., and a 30-year industry veteran with early patents on tape and disk storage devices. Lucchesi now counsels clients on issues such as bandwidth and regulatory requirements when setting up disaster recovery solutions in virtualized environments.
When transferring data from a virtualized production environment to a hot site, having enough bandwidth is critical — and that means planning and budgeting for it ahead of time. “Bandwidth issues can be expensive,” Lucchesi says. “If you’re going to support disaster recovery, you need to support the bandwidth requirements. In the end, you have to transfer the data. Bandwidth is the key governor to that.”
Sun National Bank learned that lesson the hard way — but then, that happens when you’re ahead of the curve. Three years ago — a generation in virtualization terms –the Vineland, N.J., bank embarked on a project that was slightly delayed due to bandwidth issues during failover from the production environment to the hot site. On the positive side, Sun National made good use of its displaced servers, setting up its own business continuity and disaster recovery site and pocketing the monthly fee it had been spending on third-party services. More importantly, the bank, now in charge of its own disaster recovery architecture, can comfortably guarantee it will deliver on user requests within 24 hours.
But this isn’t the brand of home-grown disaster recovery solutions we’re talking about. CIOs are dealing with disaster recovery requirements that are sometimes specified by financial or health regulations that require them to be operational within a certain time frame. “In some environments, such as a bank, being down for an hour can mean the loss of millions of dollars that you can’t retrieve, to a large extent,” Lucchesi says. And when one bank fails, another may follow, given the complex web we live in.
Surprisingly, European companies have been more actively involved in establishing failproof disaster recovery solutions than Americans — perhaps due to terrorist activity on the continent or more governmental regulations requiring it, Lucchesi suggests. After 9/11, the U.S. government initiated similar regulations, recently urging financial organizations to create plans for surviving a regional disaster. Some of these regulations require industries in the critical infrastructure — health care and power — to audit their disaster recovery solutions.]]>
Although cloud computing is not exactly being embraced by all for disaster recovery, due to incomplete SLAs — and the possibility of the cloud provider actually causing a disruption — there is a flip side.
Many companies buy business continuity operations insurance and have trouble proving the cost of an outage when it comes time to collect from their insurers.
With an on-demand DR plan, that problem can be resolved, said Chris Pyle, CEO of Champion Solutions Group in Boca Raton, Fla.
The company introduced a pay-as-you-use “cloud continuity” DR service. A hurricane hits, a switch is flipped to turn on your backup site, the hurricane subsides, and the switch is turned off.
“Because you’re only paying for what you use, you can send a bill to the insurer that shows the normal state of your operations and compare it to the capacity you needed during the disaster,” he said.
Hey, if it can decrease wrangling with an insurance provider, midmarket companies may be all for it. Still, such a benefit will remain unrealized until midmarket companies start to adopt cloud services for more than online storage and email.
Adoption of the cloud for disaster recovery is on the fast track with SMBs, according to Mike Martin, director of cloud computing at Farmington, Hills, Mich.-based Logicalis Group.
He estimates that 25% of the inquiries about its cloud services are for DR. “They want an infrastructure partner that can keep data synchronous [with their production site], and leverage virtualization to spin up virtual systems for data replication,” he said.
Time will tell if such inquiries turn into implementations. For now, a lot of education is under way, with CIOs still navigating their way through what is really capable in the cloud, and at what cost.
Email me about your DR plans at email@example.com.]]>
“Since [the cloud providers] have the data and system images hosted for you, they will recover you at their site on virtual sites now, for some time,” said Forrester analyst Stephanie Balaouras.
SMBs are backing up desktops, servers and storage in the cloud. This is the first time many of the companies have had any type of disaster recovery plan, for desktops in particular, she said.
Based on a report titled “How the Cloud Will Transform Disaster Recovery Services,” Balaouras outlines some of the benefits and costs of disaster recovery plans in the cloud:
• Pricing is transparent and subscription-based. Pricing includes all the software, infrastructure and services to deliver the solution. You are typically charged per gigabyte of data, per server, or for a combination of the two. The only cost not included is the cost of network connectivity. You pay for only the servers you want to protect.
• Deployment is fast and easy. Most of the recovery configuration can be done online, because there’s no need to reserve identical hardware, set up proprietary links or negotiate your specific SLA. Since the backup is to virtual volumes and servers, you simply have to ensure that the right virtualization layers are in use. However, these services are typically limited to your x86 server environments.
• Oversubscription risk is minimized. In traditional DR services, the provider subscribes several clients to the same IT resources, closely manages the oversubscription ratio and avoids subscribing clients from the same region to the same equipment. But there’s a chance that multiple, simultaneous disasters will be declared, in which case you won’t get access to the IT equipment you’ve spent thousands of dollars a month holding in reserve. While the same can still be true with cloud DR services, the risk is minimized because far more customers can be packed onto the same physical IT infrastructure.
• The penalty for rehearsing is reduced. What good is a DR plan if you don’t rehearse it to make sure it works? Traditional DR service providers recognize this but have to schedule rehearsals, reserve equipment and often be on call for you during the rehearsal. All this prep costs money and is typically above and beyond the DR services contract. With cloud DR services, there’s usually minimal or no prep required, allowing you to rehearse more easily and at much lower cost.
SMBs aren’t giving up all their data. Many are still backing up data locally and using cloud providers as a storage vault. If anything, the cloud is another tool that SMBs can use to cut costs and form the basis of a disaster recovery plan. For more advice on DR strategies, check out our guide on developing a DR plan.]]>