Secret Agents. The stuff of James Bond legend.
But there is another kind of Secret Agent that lives amidst our technology-driven lives.
They live on your phone; they live on your television and on your home internet router. And they also live throughout your business’ infrastructure performing a crucial service.
Literally every agent’s job, just as in real life, is to closely monitor and inform.
They often speak different languages too; not Russian or Chinese, but languages like SNMP, SOAP, and JSON. And every agent uses secretly coded messages.
Every agent on every device has his secret code book; those who speak SNMP use a code book called a MIB to carefully translate what they see happening into a number-based syntax that is easily communicated to those who can translate these messages.
Federos’ flagship assurance product, Assure1, knows how to communicate with nearly all secret agents with the sole purpose of ensuring your business continues to operate in an efficient manner. Think of Assure1 as the Government Security Service who receives translates and informs those who can react to world problems, or in this case device problems in your datacenter.
Agents constantly pass messages that let the Security Service know what is happening. They inform of big problems like if the temperature in a device gets too hot, or if the CPU is too busy. And they inform of smaller problems or non-issues, like if just a network interface is disconnected and reconnected. They inform if a configuration is changed or if they are rebooted.
These informer messages are generally called SNMP Traps; and the Security Service Assure1 receives and processes all these messages and then communicates to the Network Engineering Staff exactly what is going wrong and where.
But the agents also collect measurements and statistics that they do not generally send as messages. They collect and store information and need to be asked for this deeper level of detail.
In the assurance field these statistics are called Metrics, and generally fall into the scope of Performance Management. The assurance service, Assure1, collecting these measurements serves a very important purpose: to predict problems before they happen
When Assure1 collects a series of data at a fixed point in time, say every 5 minutes, there are magical predictive analytics that can inform the operations team BEFORE something bad happens.
For example, if the consumed memory or traffic on a port gradually increases, you can use the projected slope of a linear function to anticipate WHEN a threshold will be breached.
For an operator, this means that the operations team can be alerted WHEN a device, or a business service, or a customer, will have an issue BEFORE it happens and have time to remedy the issue – preventing impact to SLAs and the customer experience.
The Assure1 platform has many other metric based algorithms, one measures what is normal for that point in time and can alert when a metric becomes abnormal – for example, if a network interface is usually busy during work hours and today there appears to be little traffic. That model, Triple Logarithmic Smoothing, alerts your operations team when something abnormal is going on.
Assure1 has many different mathematical algorithms it leverages across your agent’s data to predict or inform when something business or service disrupting is about to happen.
Our Assure1 is the only product on the market that works at hyperscale – speaking to the largest number of agents and leveraging both metric and fault data together to ensure the operational success of your business.
Come back to our blog to learn more about the assurance space, and leading-edge technologies – such as artificial intelligence and machine learning – that we have incorporated into Assure1. We will share customer success stories here, and how our multiple different machine learning techniques assure better, at a lower cost, and at a better result. Here at Federos, we know assurance.