An open API service indexing awesome lists of open source software.

https://github.com/dsietz/daas

Overview of the Data as a Service (DaaS) architecture
https://github.com/dsietz/daas

archconf architecture daas data message-broker microservice nfjs patterns rust-lang uberconf

Last synced: about 1 month ago
JSON representation

Overview of the Data as a Service (DaaS) architecture

Awesome Lists containing this project

README

          

![DaaS](./img/cover-small.jpg?raw=true)

The Data as a Service (DaaS) pattern allows for the delivery of the Minimal Viable Product (MVP) for real-time data management, while avoiding many of the anti-patterns that traditional data provisioning and BI systems portray. Unlike traditional BI tooling, building out a DaaS system doesn't require high up-front costs and the welding of multiple products.

---

## High Level Architecture
The DaaS Pattern is the cobination of three logical components _Data Model_, _Processing_, and _Eventing_.

#### Data Model
> The _data model_ mimics an envelope, (DaaS Document) with metadata as the wrapper and the data object as the content. Because of this data model, the orchestration of the eventing can be driven by the data and does not have to be preconfigured.

Metadata Wrapper Data Model

#### Processing and Eventing
> The _processing_ can be supported by microservices or serverless functions, while the role of _eventing_ is fulfilled by a broker. Since brokering supports the publish/subscribe pattern, a plugin model of data provisioning steps can be easily added/removed and it even allows for forked parallel processing on a singluar data message.

Brokering wiht Microservices

---

### DaaS SDK
+ [Rust-Lang](https://crates.io/crates/daas)

---

### Conferences
+ [Uberconf](https://uberconf.com)
+ [ArchConf](https://archconf.com/)