The promise of e-commerce won`t be fulfilled until the Web can consistently deliver higher and more reliable performance, faster downloads, and fewer outages.
So says Jannie van Zyl, executive: managed solutions, Dimension Data Network Services. "An assortment of remedies has been proposed - from load balancers to accelerators, from faster switches to optical networking gear.
"Each of these has its place, but none gives e-commerce managers a holistic view of networks, devices and applications," he says. "As a result, practitioners have access to limited information about a limited set of network elements."
Van Zyl says Dimension Data Network Services is able to provide companies with a holistic view of their networks, systems and applications, and so maximise the availability and performance of their e-business infrastructure.
Dimension Data, which represents Concord Communications locally, uses the US-based company`s Live Health solution to collect data and generate alarms when networks and systems are under-performing. "Today`s solutions are good at letting people know when something`s gone wrong after the fact, but with the Live Health solution we`re taking a more proactive approach," says van Zyl. "We are ensuring faults are detected and resolved before potential outages and delays impact critical services."
The Live Health system collects performance and response-time data across the network, attached systems and applications, polling as often as the customer chooses. The data is then ported into the Live Health `engine`, which says van Zyl, "is where the analysis and crunching takes place, which can generate alarms."
Once the system has generated an alarm and sent it to a display, users can access a variety of analyses, such as historical and trend reports or real-time views to see how the affected element is performing. The data is analysed in real time.
The enablers for the real-time capabilities are two new algorithms, which use historical data to verify that an operational condition is truly a problem. This process is intended to ensure that managers only deal with `real` problems rather then a flood of data from SNMP traps.
A `time-over-threshold` algorithm informs users if any transaction or process is exceeding predefined thresholds for a consistent period of time. Van Zyl says by using this approach instead of generating a trap every time a threshold is crossed, "we can reduce the noise by a factor of 25".
While users can set thresholds as desired, Live Health also comes with preset profiles.
Live Health`s `deviation from normal` algorithm automatically learns the personality of a specific e-business delivery system to identify unusual behaviour. It then sends alerts to operations to notify them of the problem.
With Live Health users have fast, easy access to historical data with a simple `point and click` - providing greater intelligence for understanding the context and business impact of the problems they are experiencing.
"These issues can then be quickly resolved before they have a detrimental impact on the organisation`s critical systems," concludes van Zyl.
Share
Dimension Data
Dimension Data is a leading global network services and i-Commerce Solutions provider focused on enabling enterprises, telecommunication service providers and new economy companies to communicate and transact effectively in an e-enable environment.
Dimension Data listed on the London Stock Exchange in July 2000 and is a constituent of the FTSE 100 index. The Group has over 11 000 employees and operates in over 30 countries on six continents. It has achieved a compound annual growth rate in US dollars of 57 per cent in revenue and 36 per cent in earnings per share over the past two years.
Editorial contacts