Roadmap

2019 Goals

Baetyl will embrace cloud native to edge creative, including:

• Deliver a lightweight, immutable, all-in-memory edge computing operating system based on Linux kernel.
• Build and manage service mesh across cloud, edge and IoT devices to enable free flow of applications and data.
• Improved troubleshooting and tunability of Baetyl under production conditions.
• Supports edge clusters consisting of a large number of nodes, providing intelligent task scheduling capabilities.
• Provides various forms of device and data connection capabilities such as AMQP, OPC-UA, and ZigBee.

Release 1.0 in Autumn 2019

Baetyl will provide infrastructure and application development tools for IoT and edge computing, including:

• Core framework:
  • System security
    • Support for edge device activation and dynamic delivery of unique certificates
    • Provides a mechanism for bulk device activation
  • Application support
    • Logs can be viewed remotely, support for application acquisition, and can be set to cut off according to factors such as capacity and time
    • Provide inter-service dependencies and support the use of ports for health checks
    • Ability to pre-download application images
  • Remote management
    • Provide open source remote management console
• System related capabilities
  • Linux platform
    • Migrate to containerd, eliminate dependence on docker, further lightweight
    • Elimination of transport overhead introduced by virtual machines in container mode
    • Support for native package management on major Linux distributions
    • Support x86/x86-64/armv7/aarch64/mips/mipsle/mips64/mips64le
  • Windows platform
    • Support WSL+Docker
    • Experimental support for Windows Container
    • Experimental support for Windows 10 IoT Core
  • Kubernetes platform
    • Experimental support for K3S
    • Experimental support function distributed on multiple nodes
• Application development support
  • SDK
    • SDK with C interface
    • API for calling function in function
    • API for invoking ML inference
    • API for image capture and processing
    • API for local KV storage
  • South messaging service
    • Further reduce performance overhead on resource-limited devices
  • Function calculation
    • Support for binary functions such as C++
    • Support for shell functions
  • Streaming calculation
    • Support Flink SQL compatible streaming computing
  • North data synchronization
    • Support for connection to Mosquito, EMQ and AWS's IOTHUB
    • Support for connection to Kafka