Do you have an IT job involving enterprise data? More importantly, do you want to keep it?
If you answered “Yes” to both questions, you’ll want to make sure you’re fully prepared for three big shifts in technology that are reshaping today’s enterprise data center. In the next few blog posts, I'll talk about these "mega-shifts," such as the software-defined data center and flash SSD storage.
But in this first blog post, I’ll discuss cloud computing.
Mega-shift #1: The cloud is here to stay. Embrace it or risk being outsourced...
It’s hard to have any IT discussion today without mentioning the cloud.
- There are private clouds deployed behind a company’s firewall—typically existing data center equipment that has been reconfigured into a shared, elastic pool of on-demand servers and storage.
- Then there are public cloud offerings, which allow you to rent computing power and seemingly infinite storage capacity for pennies per GB.
- Lately we hear more of an emerging hybrid-cloud model, which holds the promise of blending private and public clouds together into a single infrastructure, with data flowing back and forth from on-premise to off-premise locations based on available resources and the demands of users during peak periods.
Whatever the variation, one point has become undeniable: Cloud computing will ultimately replace a major portion of today’s IT services.
Why? Two reasons: time and money. As this KPMG survey concluded:
With our survey showing that more than half of organizations are already working in the cloud, it is encouraging to see that cost reduction and speed to adoption remain high on the list of objectives for executives.
Here's just one example to illustrate
A major online bank used its private cloud to provision ‘bank-in-a-box’ copies, so its developers could quickly test out new applications.
Instead of having to wait months for servers and storage to be racked up, the bank was able to develop a self-service cloud platform that let its developers self-provision. They could bring up simulated banking environments in just a few minutes by cloning them from a template—the process could be sped up even further by employing an intelligent storage that copies-on-write (also known as thin cloning).
For the bank, faster time to deployment transformed IT from an obstacle to an enabler. For the IT staff, however, project success required letting go of the old way of doing things. It also meant embracing process changes needed to make the new, cloud-enabled platform work.
Incidentally the bank won a CIO100 award this year thanks to this project.
Fight cloud all you like, but it will hurt your career
You may choose to fight the cloud. You may explain all the technical reasons why cloud isn’t ready for primetime.
- You might say things like, “The public cloud isn’t secure enough for our company’s data.”
- Or, “Cloud performance SLAs are non-existent.”
- Then there’s my favorite: “Did you hear about the outage that $PROVIDER had last week?”
But taking an anti-cloud position is just avoiding the inevitable.
Here’s the reality:
- Public cloud security breaches are exceedingly rare. This is thanks in large part to the massively scaled, object-anonymous architectures of Google, Amazon, Microsoft Azure, and others like them. When your data is randomly mixed amongst billions of other data objects, it’s darn near impossible for even the best hacker to reconstruct it. Even the CIA is moving to an Amazon cloud (albeit using ringfenced hardware).
- IT network speeds have been steadily improving for the past 25 years (remember 10BASE5 thick Ethernet?). Cloud performance and reliability levels are state of the art, as evidenced by the reports from many cloud traffic watchdogs and cloud benchmarks (but be careful to size for a worst-case scenario). Many resources are available (such as from the Cloud Security Alliance) to help define and enforce strict cloud SLAs.
- Yes, there will be the occasional outage from cloud service providers. Which is why the best cloud strategy will likely involve a combination of private cloud for mission-critical apps and public cloud for important but non-essential data.
By leading your company’s transition to cloud, you give yourself the chance to become an indispensable asset to the organization. While no one yet knows if all applications will ultimately go to the cloud, your expertise in this area can help guide your company on its own journey.
Want to know how to get started? I recommend you give yourself a cloud audit:
- Assess the vendors that are currently providing you with any sort of IT cloud service. Ask yourself: Are you making best use of all their capabilities?
- Assess your applications and data looking for those that would be better suited for cloud storage. Start with the easy ones such as the data copies you make for backup, archival, and regulatory compliance.
- Take a long-term view of the cloud. Is there a way that your cloud-based applications and data could be folded together and made even more productive? For example, by combining your email archival and legal retention files into a single cloud repository with a single discovery engine? (This area is a bit more far-reaching but as my life-insurance salesman says, there's no time like the present to invest in your future.)
It may also help save your job: IT is changing. You need to change with it.
Next up: Mega-shift #2: The software-defined data center.