How to Gain an Edge for Cellular IoT Operations with Super SIM Connection Event Streams

The key to running a fleet of reliably connected IoT devices is having the right amount of insight, such as where SIMs are connecting, what networks are rejecting connections, and which SIMs are going rogue in terms of data usage. Without that, you are essentially building your business on hope, not knowledge—a risky proposition.

Performance insight is key

Super SIM is Twilio’s new cellular IoT connectivity platform, offering almost 400 worldwide cellular networks on one SIM. With Super SIM, one of the guiding principles of the product from day one was providing visibility into what’s happening “under the hood” when SIMs try to connect to networks. To help businesses grow and expand their IoT deployments, we created a reliable IoT SIM solution offering global reach, which allows companies to select exactly which networks their device should connect to in each country. So we built our own mobile network core—the key component of any wireless network that decides who gets to connect, and how data flows to external networks such as the Internet. Owning the core gives us the deep, to-the-edge visibility that WE need to let YOU run a successful IoT deployment.

Now, we are giving you new ways to get this level of insight too.

Twilio now streams Super SIM Connection Events

In April 2021, Twilio Event Streams entered public beta. Event Streams is Twilio’s new centralized framework for exposing (near) real-time information on everything that is going on with Twilio products—including Super SIM.

Examples for events in the world of cellular connectivity are:

• a SIM tried to connect to a network (via our Distributed IoT Mobile Core),
• a SIM disconnected from a network,
• a SIM was used with a new and unknown device (IMEI),
• a certain amount of data got consumed by a SIM.

These events are meaningless if you don’t look at them in the context of your operation and if you don’t even have a reliable and scalable way of accessing them. When setting up an event stream on the Twilio Event Streams platform, you can currently choose between Amazon Kinesis or generic webhooks as destinations (with more to come). From there, you can choose how to use the data:

• To automate your operations, such as alerting staff to review an unexpected data spike, deactivating a SIM immediately upon detecting it in an unexpected country, or billing your customers for their data with realtime usage information. React to discrete SIM events that you can enrich with your own internal data to trigger workflows using your own internal or 3rd party APIs.
• For advanced troubleshooting, you might use tools like Elasticsearch with Kibana on top for visualization to follow a SIM’s connection history or create powerful dashboards.
• For business analytics and forecasting, e.g. to see the trend of SIMs connecting for the next few weeks, based on your past traffic, you could put Amazon Quicksight on top of your elasticsearch DB.

If you are reselling our connectivity as part of an IoT solution you are taking to market, subscribing to our event stream also lets you implement much more accurate end-to-end, near real-time usage metering and billing.

At Twilio, we like to “drink our own champagne”. So how are WE using Super SIM Connection Events to improve how we run our own operation?

A glimpse behind the scenes: The path from data to insight

Twilio, as a whole, produces a lot of data as we’re handling billions of messages, calls, and emails every day. Super SIM is no different. With our Distributed IoT Mobile Core capable of connecting millions of IoT devices to the Internet and your clouds, we produce billions of events each month. These events end up on data streams to which consumer services can subscribe to transform and store data or take action.

The various components of our Distributed IoT Mobile Core each publish an array of events with different information in different schemas (and different time zones -- oof). By transforming and combining those different data streams into a single stream, we can now quickly trace the lifecycle of a SIM’s connection from when it first tried to attach, to when it created a data session and continued to use data, to when it disconnected from the network. And by enriching the stream with other information about the SIM’s fleet or the account to which it belongs, we can instantly pivot and filter the events for a holistic view. This single stream became our Super SIM Connection Events.

To enable all the members of our Support, Operations, Product (and yes, even Engineering) teams to benefit from these Connection Events, we funnel the events into Elasticsearch and then use Kibana to inspect them and build visualizations.

Let’s say you as a customer identify a handful of devices that went offline unexpectedly in Paris, France, and you open a support ticket. Here are the different steps we may take:

• Our support engineers may drill down into the individual events for those devices over the last 24 hours to see if they are still connected or are trying to reconnect in real time; a simple configuration change may solve this.
• In that same moment, our carrier operations team may look at eventstream-powered dashboards showing trends for all of France for the last week to ensure your offline devices aren’t part of a bigger issue that our systems didn’t automatically detect.
• Later in the week, our product manager may pull together a heat map showing global usage for the last 30 days and identify a growing opportunity in Europe; they filter and chart the data over different dimensions, preparing to pitch a new initiative.

All of these tasks are being completed by leveraging the same Super SIM Connection Events. If that sounds like something you want for your own organization, you can follow this step-by-step guide to get you started with Super SIM Connection Events via AWS Kinesis, ElasticSearch, and Kibana.

Improving operational efficiency through automation

Another way we use Super SIM Connection Events is to automate usage and billing, with multiple services simultaneously reading from the event stream to carry out different operations. One service is aggregating the data usage to make it available to you via our APIs and Console within seconds or minutes, not hours. At the same time, other internal services, such as our billing system - also subscribed to that data stream - are performing other near real-time operations allowing us to efficiently and scalably run our business.

Next Steps

As you should realize by now, event streams are powerful. If you would like to get started with leveraging Super SIM Connection Event, here are a few resources from our documentation that give you a jump start:

• Learn about Different Connection Event Types and Payloads
• Get Started with Super SIM Connection Events
• How to Monitor Super SIM Connection Events using AWS ElasticSearch and Kibana