Serverless Computing is the new design pattern that is emerging when creating business applications. It does not mean that , there are no Servers in executing the functionality. It is more of , the complexities about managing the servers, compute capacities are hidden from the application itself and the application only concentrate about its functionality.
Serverless architectures are more closely aligned with Event driven Computing, where the applications are viewed as a set of small functions which execute based on the triggering events. This model is different from traditional procedural and monolithic application architectures.
There are lot to be read about Event Driven and Serverless architectures in the context of agile applications which follow Microservices design pattern.
Azure Functions : Azure Functions is an event-driven compute experience which allows you to execute your code, written in the programming language of your choice, without worrying about servers. Benefit from scale on demand and never pay for idle capacity.
BigchainDB : In my earlier articles we have seen the introduction of BigchainDB as one of the Scalable Blockchain databases. With Bigchain DB we can build De-centralized applications whose transactions are immutable.With its current schema model, BigchainDB is a good choice for Asset Registrations, Ownership and transfers. Typically BigchainDB will work in tandem with Smart Contract blockchain platforms like Ethereum.
One of the important aspects of BigchainDB is its support for Web Sockets. BigchainDB provides real-time event streams over the WebSocket protocol with the Event Stream API. Connecting to an event stream from your application enables a BigchainDB node to notify you as events occur.
With BigchainDB ability to store assets and considering the fact that transactions can be notified to client applications using Web Socket we could think of many real life applications which are event driven and use the serverless architectures.
The below the architecture of Serverless Event Driven application using BigchainDB web socket interface.
The following are the brief description of components involved in this architecture.
BigchainDB Web Socket Event Stream API : BigchainDB provides a stream for all validated transactions. Each stream is meant as a unidirectional communication channel, where the BigchainDB node is the only party sending messages. The following API will show the end points for connecting the Bigchain DB Event Streams.
Azure Web Jobs : WebJobs is a feature of Azure App Service that enables you to run a program or script in the same context as a web app, API app, or mobile app. Web Jobs are typically used for creating and running back ground jobs. Webjobs can be continuous or triggered jobs. Web Jobs in continuous mode can play the role of Web Socket client so that it received events from BigchainDB for process them. The following is a simple code that listens on a Event Stream and logs the output.
using (var ws = new WebSocket("ws://****.westus.cloudapp.azure.com:9985/api/v1/streams/valid_transactions"))
ws.OnMessage += (sender, e) =>
Console.WriteLine("Event Data: " + e.Data);
Azure Functions : As mentioned earlier, Azure functions is the event driven and serverless component of the architecture. Use Azure Functions to run a script or piece of code in response to a variety of events. Azure Functions has multiple kind of triggers which makes the underlying process to run.
Generic webhook is one of the triggers which Process webhook HTTP requests from any service that supports webhooks
Now if we look into above architecture from the following scenario.
- Bigchain DB stores the Car Leasing data for a major car leasing company
- The initial Car Asset is created with the ownership of car leasing company
- Every time the car is leased out a new TRANSFER transaction is created in BigchainDB
- BigchainDB event stream web socket trigger notified the Web Jobs
- The Web Jobs which is created and owned by an insurer will be notified of car leasing ownership changes
- The Web Jobs triggers a serverless API call using Azure functions
- The data gets stored at its destination.
The insurance company on getting notified of changes in Driver, can adjust the insurance premium according driver's past history.
This is a simplistic example of a architecture that uses Event driven and Serverless architecture from the perspective of BigchainDB and Azure.
Currently I face a few errors with Web Socket implementation but they will get solved. Write to me if you are interested in knowing how Event Driven Architectures will be of use in a Blockchain application.
Also current Azure functions does not support Websocket directly and hence we need to introduce Web Jobs as a intermediate component, however this can be avoided if Azure functions support Web Sockets as one of the trigger.