Cloud Scaling: Three Ways To Be Smarter About IT

- Ray Solnik, President, Appnomic Systems, says:

Despite the progress so far, capitalizing on Cloud remains a vast open field ahead and a tough challenge for many. Can it be reliable enough, fast enough, secure enough, affordable enough or scale enough?

Few Cloud providers are providing visibility and solid, measurable, validated proof of their performance. Of course, we hear about the major outages by Amazon, Netflix and others, but there’s more going on behind the curtains than these major events.

To truly achieve the potential of Cloud systems and applications, the IT/Web Operations function needs to get better at focusing on what’s happening in the application layer and then down into the data center stack. Static monitoring of data center devices and then up the stack from lower layers is no longer going to cut it in the new Cloud paradigm.

Monitoring and analyzing performance from the application perspective, end user experience down into the data center, is now the best way to ensure Cloud apps are managed effectively and efficiently – preserving the bottom line while serving users. YES, the Cloud is actually turning out to be pretty expensive if improperly managed. 

To meet customer expectations and achieve the potential of the Cloud, application owners may now leverage advanced analytics to help determine when an app stack, along with underlying data center resources, must scale up to adequately serve customers and then scale down to preserve budgets. Advanced analytics can now also help identify issues before they become real IT “accidents.”

Read on to hear about three key ways to be proactive and smart in the Cloud – leveraging this top down approach and advanced analytics:

Link scaling management to ROI analytics

Scalability is intrinsically linked to ROI. Often this painful realization occurs when Cloud service invoices highlight that Cloud under-utilized resources left “on” are significantly more expensive than effective resource self-management.

Instead, organizations should be moving towards elastically managing resources, automatically scaling up with peak demand and down during low-usage. 

With advanced analytics, forward thinking operators can even crank up performance, for example, when a new online advertising campaign drives traffic to a website and the website wants to provide as good an experience as possible for end users interacting at the application layer. This approach maximizes customer conversions and can lead to better customer loyalty over time as well – both, huge drivers of ROI.

Fire prevention is infinitely better than fire fighting

Preventing fires instead of fighting them requires Monitoring + Analytics + Automation. This combination is now available and ready for commercial application from a variety of vendors.

Advanced analytics are enabling web operations teams to operate with a MTTP (More Time to Prevent) IT “accidents” vs. a reactive MTTR (Mean Time To Repair) approach which focuses on putting out fires once an incident has occurred.

Advanced analytics run on real-time monitoring can implement sophisticated, yet actionable pattern matching to find deviations in production environments and help IT professionals get in front of issues before they turn into “accidents.” In many cases, systems can also be configured to then automate a remediation.

With MTTP operations, downtime is dramatically reduced and cost savings range from 30-50%. It also turns out customers are more loyal and happier as they are calling and complaining less frequently about application slowness, outages, and the like.

Be smart so both IT and the bottom line win the war

These Advanced or “Big Data” Analytics deliver impressive results in saving critical IT time and positively affecting a business’ top and bottom line.  While it’s early for this market, real results are beginning to be achieved in large, enterprise and Cloud data center environments. 

Think about it, if effective operations analytics can prevent a single failure to critical applications, the cost to get there is probably worth many times the original investment. IT resources are freed up and businesses can provide premium user experiences.

Gartner has recently launched coverage for a software category for Big Data Analytics that they call IT Operations Analytics (ITOA) and they are forecasting this category of software will permeate web and enterprise IT operations over the next five years – in the order of $2 billion.

Cloud scalability is not just about providing more data, applications and services, it’s about being smart about how you do it. And locating and preventing potential “fires” is critical in any enterprise IT environment, but especially in the Cloud. While it’s still early, it is certainly time to get flying on smart analytics for your data center operation.

As president of Appnomic Systems, Ray has P & L responsibility with a focus on business growth in North America. He brings to Appnomic twenty years of experience in cloud computing, managed network services, and data communications. Prior to Appnomic, Ray was president and COO of OpSource, an early SaaS/IaaS provider, which was acquired and is now the core Cloud offering of Dimension Data - a $4 billion systems integrator. Ray has helped multiple next generation companies develop and drive strategies resulting in successful fundraising from top venture capital investors, including Gengo, PowerCloud Systems, and CrowdFlower. Earlier in his career, Ray was chief development officer of New Edge Networks (acquired by EarthLink), and president of AT&T’s consumer Internet services business, AT&T WorldNet. He has a bachelor's degree in economics from the University of Michigan and an MBA from Stanford Graduate School of Business. He lives in Silicon Valley.