The web connects, constantly. This means that the internet is always on, apart from when it’s off. To be clear, the internet never actually “goes off” as such because the cloud service provider datacenter servers running the data estates that feed the web with its juice are always powered up… and they have power supply redundancy to cover times of national grid disconnection. Our computers and mobile devices are as good as always-on too, so in theory, there is no powering down for the internet’s eternal flame.
But as we know, the web can appear to be off when an outage occurs due to one of the planet’s global internet services providers experiencing a system malfunction, misconfiguration or miscalculation in load balancing, service capacity provisioning or when a major software code error occurs. When that happens, we users tend to say “the web is down” and the techies running the cloud-centric web resources that we now rely upon say that there’s an “outage” and service disruption inevitably occurs.
Which Bar Has No Beer?
Wouldn’t we be able (as user community) to channel and manage our use of the internet more competently if we more accurately knew when and where internet outages were happening? Given this type of knowledge in any other scenario, we would probably avoid restaurants that we know have run out food, the supermarkets that have no produce left and the bars that have run dry as we channel our custom elsewhere. Why shouldn’t the same logic apply to web consumption? Internet performance monitoring company Catchpoint thinks we shouldn’t.
The company has this month announced the launch of its live Global Internet Outage Map, a free-to-use AI-powered dashboard designed to offer a real-time snapshot of the health of hundreds of global internet services that run our everyday lives. This tool provides users with visibility into outages that may be impacting businesses, retail or consumer websites, core internet services or any web-based entity. The service does not require user registration, but is comprehensive enough to provide real-time, global data from the past 24 hours, sampling hundreds of services monitored by Catchpoint’s own Internet Sonar capability, which provides deeper insights for paying customers.
“With the release of our Live Internet Outages Map, Catchpoint is delighted to be able to offer to the community for free, real-time, global insights into the health of some of the world’s biggest Internet platforms, from Facebook to Amazon and so on,” said Mehdi Daoudi, CEO of Catchpoint. “By providing a subset of the features from [our own] Internet Sonar service, we hope to provide businesses the insights they need to answer the question, ‘Is it me or something else?’ Today we must realize that visibility into the health of internet infrastructure and the layers driving Software-as-a-Service functions for world business are key to staying competitive.”
MORE FOR YOU
How ‘Healthy’ Is My Instgram?
This global connectivity map has been built to provide insights into the health of many internet services and platforms, including social media platforms such as Instagram, LinkedIn and Facebook; ecommerce services from the likes of Amazon, Apple and Walmart; core finance functions from big players including Visa, PayPal, Bank of America; entertainment applications like Netflix, Disney Plus and Spotify; plus also travel services spanning airlines, hotels, booking sites etc.
Daoudi says that this technology allows users to proactively respond to outages that could disrupt business operations. “Unlike other trackers that rely on delayed, crowd-sourced reports, Catchpoint's map utilizes data from millions of real-time tests conducted on the world's largest independent active observability network. This results in an AI-powered real-time, interactive status report,” he notes.
Crowdsourced Data Dilemmas
Catchpoint’s use of real-time tests is interesting. Many of us will have travelled in an Uber (or another rideshare service) and seen the driver avoiding traffic hot spots that are flagged by other drivers’ crowdsourced data inputs. This is (arguably) a useful application of crowdsourced information streams, albeit one that is inevitably subject to the human error that Catchpoint insists we need to avoid if we’re using the web i.e. om a world where the cloud and the web runs mission-critical services for enterprise organizations and public bodies around the world, we need to go beyond community alerts relating to “sheer weight on traffic on interstate-90”, even if those are useful at another level.
“Users receive immediate outage information without waiting for third-party providers to post (often) delayed reports, providing timely insights when they are needed most. Our data is derived from numerous tests run on Catchpoint’s Global Observability Network and curated by our AI correlation engine,” writes the company, in a technical statement.
To reduce the potential of false positives here, Catchpoint says that its branded Internet Sonar AI analytics engine carefully examines outage data to validate results, avoiding inaccuracies that can mislead decision-making. The full commercially-chared version of Internet Sonar monitors thousands of additional Internet services in additional categories, including internet infrastructure (AWS, Azure, Google Cloud etc.), SaaS (AlaaS, UCaaS, SECaS, etc.), marketing technology (analytics, ad serving, video etc.), plus also API and developer tools. The full platform also allows for customizable timeframes, detailed notifications and wider integration into various other software systems.
As part of a busy summer update period, Catchpoint has also now detailed a new set of web performance capabilities to its internet performance monitoring platform. By incorporating WebPageTest’s web performance optimization features into its platform, Catchpoint now offers the a view of web performance, covering everything from front-end code to end-user experience including detailed views into the “internet stack” from thousands of locations around the world, all within a single dashboard.
MTTR Matters
The acronym that enterprises will gravitate their use of web outages around here is course MTTR, standing for mean time to resolution, which clearly really matters. Before that R-factor there’s MTTI for mean time to identification and this duo of digital yardsticks taken together and driven by a technology that serves this space as detailed here is a key step for businesses who need to keep their connectivity quotient as high as possible and, crucially, enable them to see windows where they may have to balance dropouts.
As we said at the start here, the web never really experiences a formal turnoff, but knowing where the "outage offs" are and being able to navigate around them might well be a turn-on.