Service Level Management System

IT PERFORMANCE MATTERS

Your IT infrastructure supports critical business processes that require complexity and flexibility. By monitoring these critical processes in real-time you prevent

• Slow performance and outages
• Poor customer service
• Non-compliance with Service Level Agreements
• Lost revenue

THE SLMS SOLUTION | MONITORING IN REAL-TIME

By measuring actual end-user experience as transactions flow across diverse IT systems, InsightETE records true end-user IT performance to verify SLA compliance. Our open architecture maximizes flexibility by integrating into your existing frameworks.
The insightETE Service Level Management System (SLMS) gathers all the data needed to

• measure performance
• analyze problems
• track application availability
• All from the customer perspective

PROBLEM ASSESSMENT THAT MAKES SENSE

Now that IT problems can be identified quickly and accurately, you can prioritize them in a way that makes sense. With insightETE, IT issues can be prioritized based upon their effect on your business rather than the inefficient “first come, first served,” method used today.

THE BOTTOM LINE: HOW WE SAVE YOU TIME AND MONEY

• Automated problem detection
• Elimination of manual SLA data collection
• Reduction in staff-hours for analyzing and correcting problems
• Reduced tool maintenance fees
• Rapid installation and configuration
• Reduction in equipment purchases by efficiently deploying new system assets
• More efficient business operations from reduced downtime and consistent responsiveness
• Rapid problem resolution
• Improved customer satisfaction
• Increased revenues from customer-facing applications (e.g.: online banking, credit card sales, call center operations, reservation systems, etc.)
• Intelligent Availability Tracking

What makes it so smart?

Normally, synthetic transactions are designed to run at a pre-defined schedule which never varies. While there is certainly benefit to having synthetic transactions in place, it just isn’t necessary to run them at all times, even when real users are obviously still accessing the system being monitored. Our Synthetic Transactions can be configured to only execute when there is no active real user traffic on the servers it is monitoring.
Our tool passively observes real user traffic going back and forth between your end users and your application servers. If there is a lull in the traffic, then our tool can then send out a synthetic user to test the system. This helps us to answer a critical question: Is it down or is it idle? This provides an unprecedented ability in even the most modern of IT Performance Monitoring and Availability Tracking tools. Namely, the ability to truly and completely minimize the impact of synthetic transactions on a production infrastructure.

What Impact could Classic Synthetics have on my Production Environment?

First, let’s level-set: the idea that well-crafted synthetic transactions can bring down the infrastructure of a business-critical production application is highly unlikely. Our goal is not to claim to fix a problem that doesn’t exist. However, synthetic transactions do add additional load to a production environment. If that environment is already taxed by over-use, then one runs the risk of having their synthetic transaction (the tool that is supposed to help ensure the stability of the infrastructure it is monitoring) be the straw that broke the camel’s back and finally take down a system. More realistically though, it could take up a valuable thread which would make a real user, with real business, wait in line behind a robot.

There’s more . . .

If you’re in an industry which relies heavily on external vendors or mainframe transactions, then each time a user (or robot user in this case) accesses those resources, you may get charged real money for a fake transaction. And while those charges are normally fairly small, when they are added together over the course of a year, the cost of monitoring suddenly is much higher than just license fees and resources.

A REAL WORLD EXAMPLE
A huge financial institution had 80 robot agents playing scripts which accessed 5 applications that had direct calls to a mainframe which charged around $0.40 per transaction. Each of those 80 robot agents played each of those 5 transactions once every 10 minutes 24 hours a day, 7 days a week.
So let’s do the math:

× 80 robots x 5 scripts x 6 runs/hour x 24 hours/day x 365 days/year x $0.40 cost/transaction =
$8,409,600 spent annually servicing “users” which will NEVER spend a dollar!

Why use Synthetics at all?

A purely passive tool has a serious problem when presented with an application that has no traffic. It has to decide if the reason is because the application is down and unable to service customers or if the application is just not busy at the moment. A purely passive tool could keep track of the typical volume for a time of day and track against that (which, we do!) but even that is just letting the purely passive tool make an educated guess as to the health of the infrastructure when nobody is using the system. A synthetic transaction allows a monitoring tool to test the application at a specific interval to ensure the application is able to process user transactions when they do happen. The cost associated with a synthetic transaction is justified when there is no other way to verify the availability of a system. It used to be that cost was justified all the time. Now that InsightETE has broken new ground with combining the benefits of passive monitoring and active monitoring into a single engine, the cost of 24×7 synthetics is no longer justifiable.

What's the bottom line?

In the same types of systems described above the formula changes quite a bit. First: the reason those systems required so many robot agents is because it is a business critical application with world-wide users. As a result, the application is rarely idle. In fact, it was typically only at low volumes between 4am and 6am EST. Additionally, the number of agents was only so many because they had no passive monitoring solution to dive into the Real End User Experience. Thus, if the goal for synthetics now is strictly to track availbility those could be limited to regional robots rather than so densely spread out. As a result, the new formula could look a lot like this:

× 10 robots x 5 scripts x 6 runs/hour x 2 hours/day x 365 days/year x $0.40 cost/transaction =$87,600 spent annually ensuring real end users can access the application when it is needed.

The bottom line impact?

× $8,409,600 original cost – $87,600 cost with intelligent synthetics = A savings of $8,322,000 annually! (For those keeping track, that is a 98.958% savings!)

While we obviously cannot promise that every situation will be as stunning as this one, we can say with the utmost confidence that any company will see a significant savings and substantially more impactful monitoring using the combination of methods we are currently pioneering in the industry.