Have you ever watched an auction? Whether people are seeking to buy fine art, cars or livestock, they will spend time at an auction house, in order to place their bids and hopefully snag a deal. A well-run auction is a fascinating event to watch, and most interesting of all is the person who calls out the prices and acknowledges the bids. The auctioneer uses a keen eye to identify the most subtle of gestures from the audience: the flick of a single index finger, a raised eyebrow, or a gesture of the head is all that is needed to read the intentions of the bidding floor and move the process ever upwards.
Wouldn’t it be great if that same subtle ability existed for IT managers? To be able to look over a collection of dashboards and readouts and immediately detect a subtle spike or an errant digit that signifies trouble ahead? Granted, there are some people who are able to do just this – usually after many years of constant practice and daily experience – but for companies struggling with continued growth as well as modifications and upgrades to their systems as well as a highly volatile marketplace, the capacity to notice subtle warning signs just can’t keep pace.
An inability to spot trouble or page load delays due to the overwhelming amount of data to monitor is the reason why all IT managers need to ensure they are working hand-in-hand with applications that can do the job for them. As a recent PEER-1 report shows, shoppers insist on a faultless online shopping experience, and will readily abandon an online shopping cart because of slow websites, time-outs and long page load times. The PEER-1 report reveals also that such dissatisfaction then becomes negatively linked in the customer’s mind, to the brand they were seeking to purchase.
Another recent report, published by Load Impact shows that 40% of e-commerce respondents to their study “reported losing money in the last year due to performance issues, compared to only 25 percent of sites in other categories.” Other crucial facts from this study included:
- Sixty percent of companies overestimate their site’s capacity to handle user traffic.
- Ninety-eight percent of e-tailers thought sub 2-second response time was desirable.
- Very few respondents load-test the mobile versions of their sites.
Eric Anderson, CTO and co-founder of Austin, TX-based CopperEgg, points out that the essential missing piece in the monitoring puzzle is custom metrics: in other words, building a system that speaks directly to the operator. “Nobody can cover everything,” he says, “so you need to be able to feed in all kinds of important business metrics and parameters. Whether these data points are granular or high level, they need to be visualized both over time and just in time.” He explains that when there is an issue that affects a critical element such as load time, the problem might be an immediate one, or it may be necessary to look back over the past 24 hours, or 30 days, to identify a pattern, a spike or some other tell-tale indicator of an anomaly – something that displays not only that something did change, but when it changed.
This, he says, is the difference between basic monitoring and the more custom monitoring systems that companies such as CopperEgg provide. The high-speed platform upon which all servers operate, both under the hood, but equally importantly in front of the end client requires the keen eye of a seasoned auctioneer, since, in both worlds, a great deal of money is at stake. When admins are empowered to fine tune the monitoring interface to speak to the key issues and put aside potential distractions, the goal of attaining maximized uptime and reliability gets a little closer.
By Steve Prentice
Sponsored Post By Copperegg
- The Need For A Security Incident Response Team - October 6, 2015
- Facebook’s Lesson In Effortless Marketing - September 30, 2015
- Cloud-Based Collaboration And The Heart Of The Tin Man - September 28, 2015
- Lawsuits and Legacy – Ensuring Legacy Material Will Withstand Scrutiny - August 3, 2015
- Unscrambling An Egg: How CIOs Can Enable Business Through Unstructured Data - June 1, 2015