The most complete realtime AIOps platform for critical networks — Part Three
OptOSS AI — The most complete AIOps Platform for Telecoms
This is the third instalment of a series of blog posts that helps navigate the rapidly changing world of critical telecom networks and provide insights on the future of AIOps technology. The series aims to contextualise the requirements of technological solutions for highly complex data-driven operational environments and to highlight OPT/NET’s very own flagship AIOps platform: OptOSS AI — the most complete real-time AIOps monitoring & analysis platform for critical infrastructure networks.
The health of a telecom network is illustrated by the billions of log lines and performance indicators it outputs every day. Teams of System Operators are tasked with sifting through this veritable ocean of data in order to manage these critical networks and ensure zero down-time. Human operators simply cannot manage the colossal amounts of data in real-time to uncover the “needle in a haystack” that may ultimately lead to a service disruption. That’s why they rely on the alarms and alerts from complex Operations Support Systems (OSS) in an attempt to stay on top of the deluge of data. As we discussed in the previous article, it’s not just the volume of data that is increasing, but the variety and velocity as well. Gone are the days of networks dominated by voice-traffic, replaced by an ever-growing demand for interactive data usage.
We look to some painful statistics to illuminate how truly overwhelmed many networks are becoming under these new & evolving circumstances:
For telecom operators, ensuring reliable network availability is imperative in order to consistently gain new customers and maintain the trust of existing customers. But the reality today is that most network operators are typically working in a ‘reactive mode’ fighting proverbial fires, often only responding to major disruptions after many customers start complaining. Not only does this poor quality of service negatively impact the customer’s perception of the brand, but in some cases, it may even lead to costly breaches of SLA’s.
With our extensive experience in the telecom industry, we are all too familiar with this problem and the daunting task of working in these increasingly complex and mission-critical environments.
From Telecom consultants to Telecom startup
The OPT/NET BV team comprises telecom network experts that have over 50 years of combined experience in the industry. Having served and protected the networks of clients all over the globe, the OPT/NET team was especially suited to develop advanced AI products capable of dealing with an ever-increasing volume of data and complexity. Initially serving as a tool for our own telecom consulting practice, OptOSS AI has grown into a stand-alone platform that is truly best-in-class.
From being ‘just’ a highly rated open-source network discovery/management tool — NG-NetMS — the journey towards evolving into a stand-alone commercial AIOps platform began during a business incubation granted by the European Space Agency (ESA), where the tool was augmented with spaceworthy “rocket science” know-how and assistance from ESA engineers and data scientists in the area of advanced operations concepts.
The next phase of evolution came through a collaboration with the University of Amsterdam (Top 5 European AI Research Institute), which led to a highly innovative AI module being devised and infused into the platform. Finally, with over 10 years of R&D behind it (and millions € of cash investments), OptOSS AI was launched as a commercial product for the Telecom industry in 2018. Due to the novel nature of the innovations involved, OPT/NET BV has secured Dutch, European, US patents and has pending International patents covering the relevant know-how and technology used in OptOSS AI.
What is OptOSS AI
OptOSS AI is an Artificial Intelligence enabled platform for monitoring, analysing, and managing complex networks in real-time. Both structured and unstructured time-series data can be streamed into OptOSS AI, where irregular patterns are detected within 2s of the data being received. A patented process helps to detect, cluster, label, and recognise millions of known and unknown irregularities that crop up on critical infrastructures and to distill the gist of meaning helping human operators in their daily jobs. Crucially, the majority of new and repeating issues are automatically spotted by OptOSS AI, without the need for clean labeled training data. Spotting the unexpected, noticing the unknown patterns is where OptOSS AI excels, often surprising even its creators.
OptOSS AI — The virtuous cycle of ‘education’
Being able to detect known and unknown anomalous patterns without any training data allows Operators to extract meaningful insights with OptOSS AI from Day 1. As soon as data starts streaming in, OptOSS AI begins to autonomously detect and cluster anomalies into groups based on similarity. These are then brought to the Operator’s attention, who then works to educate the system on the significance of the presented clusters. As the Operator spends more time working with the system they are not only granted the ability to get a full overview of the network, but they are empowered to intelligently automate away the mundane and repetitive tasks and re-orient themselves towards diagnosing new anomalous incidents. This virtuous cycle of education fits seamlessly into an Operator’s day-to-day activities and allows them to work much more effectively.
OptOSS AI — What to expect from a deployment
In any deployment, OptOSS AI first connects to all routers, retrieves the most current network topology, and creates a list of all networks to be audited. Then, it collects all hardware and software configurations and parameters from every discovered and supported element and automatically monitors all additions and deletions of devices, components, and their respective network links. It creates precise audit reports and maps, rapidly collecting the most complete information about the managed network’s inventory, topology, map of IP address space, and most important, analysis of Syslog events and SNMP alarms.
After initial discovery and integration, OptOSS AI becomes an intelligent autonomous and dynamic system that helps operators with their daily operational routines. The base dashboard interface (see below) provides operators with a graphical 360° overview of Network Activity & Performance.
The Streaming Alarm-view can also be engaged if needed. When incidents are detected and brought to the Operator’s attention, this point-and-click interface provides Operators with the ability to zoom in on the full context of any anomalous activity and allows for rapid root-cause analysis & remediation when necessary. OptOSS AI learns from this process, and allows for the correct remediation or alerting procedures to be enacted automatically next time. This frees up the Operators to focus on new incidents if and when they appear.
The visualisations of activity across the network are highly customizable and can provide Ops teams with a variety of useful and pragmatic insights. For example, the “CT-Scan” visualisation intuitively illustrates the extent of different types of detected anomalous activity across a network over time — clearly visualising and communicating any hidden patterns to the operators.
By deploying OptOSS AI, telecom providers can rest assured that they have the full and up-to-date picture of everything occurring in their network. Human operators can finally shift from a reactive ‘fire-fighting’ (e.g. reacting to customer complaints) to a proactive mode of operations with unmatched actionable intelligence. Within days of deployment, it’s possible to improve the monitoring capacity & capabilities by several orders of magnitude by empowering Ops teams to work better, faster, and more cost-effectively.
OptOSS AI also brings great value to the Data Science teams, by offering unprecedented tools for the rapid creation of new analytical products, covering new areas in a fraction of the time and effort.
OptOSS AI — Benefits for customers
Save massive costs and deliver a better service by avoiding outages
Any anomalous patterns that emerge, which may have never been seen before, are detected instantly before they can cascade and bring down the entire network. Spot ‘the needle in the haystack’ before it causes a disruption, be it during routine maintenance, 3rd party updates, or a service rollout procedure.
Empower Operators with insights
Provide Ops teams with context-rich and actionable real-time insights into the issues currently brewing in the network, before your customers notice them and start calling to complain! OptOSS AI can detect performance issues across the network, identify affected and unhappy customers/problematic sites, and kickstart the remediation process before things spiral out of control.
Intelligently automate sophisticated rectification procedures for frequently occurring and/or mundane events. Empower the Ops teams to focus on monitoring the live condition of the network. Leave it to OptOSS AI to trigger troubleshooting when needed, lock out intruders, reroute traffic, or restart relevant network assets or even initiate the service tickets for engineer deployment when things go awry.
OptOSS AI — Built for speed at scale
Performance: In a standard configuration, the platform can handily ingest and process >50,000 separate data points per second. Typical service provider deployments demonstrated performance of hundreds of thousands of data points per second on general-purpose COTS server hardware and it possible to arrange the product for mega-scale deployments, running modules concurrently or in certain hierarchical distributed system arrangements can even unlock processing performances of many millions of events per second.
Scalability: A typical OptOSS AI deployment can ingest data from over 10,000 connected devices, thereby covering all of your critical systems. The deployment architecture relies on customisable and modular design, which is highly flexible and provides room for future expansion.
Speed: OptOSS AI completes full network discovery in a matter of minutes if required, and takes less than 3 seconds from data ingestion to detection of anomalous behavior.
Value: Our ‘rule-of-thumb’ goes as follows: 1 Million events can be processed daily per €100 investment in hardware! Furthermore, negating the need for labeled training datasets (not to mention expensive data scientists!) also saves a significant amount of time and expense compared to other AI approaches.
OptOSS AI features all of the hallmarks of a suitable cognitive AI approach
If we think back to our previous article, where we discussed the hallmarks of a suitable AI approach for future-proof telecom network monitoring, we see that OptOSS AI fulfils each and every requirement!
OptOSS AI — live and ready to be deployed on your network today!
Schedule a demo to find out more about how OptOSS AI was able to help a leading Dutch Telecom generate real-time actionable insights, or how a leading CDN company reduced security exposure and saved 12% in base infrastructure costs within days of deployment. Our team is ready and excited to share how OptOSS AI can help you cope with ever-increasing complexity on your networks today!
At OPT/NET B.V., we focus on enabling our partners & users to rapidly deploy real-time AI solutions in mission-critical environments. Mission-critical can be defined as “crucial for the continued functioning of an organisation”. From telecom networks to disaster zones, from farmlands to the open ocean. In a world of increasing technological connectivity, maintaining the resilience and robustness of these environments becomes increasingly relevant (not to mention complex).
Follow us on LinkedIn to stay up to date & learn more about how exactly OptOSS AI helps to facilitate the transition to a more agile, digital-first type of organisation, and navigate an increasingly disruptive environment.