Fog computing with Eclipse fog05
1 like233 views
The document discusses fogOS, an open-source software platform for managing resources across cloud, edge, and IoT devices in a unified way. It describes fogOS's architecture which includes a Fog Orchestration Engine (FOrcE) and Fog Infrastructure Manager (FIM). FOrcE manages applications and services, while FIM virtualizes and abstracts hardware resources. FogOS supports various deployment options and aims to provide a decentralized infrastructure for provisioning and managing distributed resources regardless of location.
Fog computing with Eclipse fog05
- 1. Advanced Technology Office Gabriele Baldoni Technologist ADLINK Tech. Inc. gabriele.baldoni@adlinktech.com AT() 05fog
- 2. Hardware Tiers in (I)IoT / MEC A generic IoT/IIoT/MEC system has three different hardware tiers Off-premises data-centre which may be private or public On-premises edge infrastructure Things with computational, communication and storage capacity
- 3. Cloud-Centric Perspective The cloud infrastructure is mature and operationally convenient… Yet cloud centric architectures don’t fit well for a large class of IoT/IIoT applications
- 4. Cloud Centric Perspective The application is deployed, managed and monitored using the Cloud IaaS infrastructure
- 5. Edge-Centric Perspective The main idea of Edge- Centric architecture is that of providing edge-clouds to reduce some of the short- comings of traditional Cloud Centric architectures
- 6. Edge-Centric Perspective The application is deployed, managed and monitored using the Cloud IaaS infrastructure which cooperates with edge Cloud infra.
- 7. Cloud, Edge, and Things Cloud Computing gives operationally convenient abstractions and tools to manage and provision data-centre resources Edge Computing alleviates some of the challenges posed by cloud computing at the cost of introducing some fragmentation in the infrastructure. But how about the Things? In a large class of system we need to manage and provision them too…
- 8. What’s Really Needed? An infrastructure that allows us to federate compute, storage, I/O and communication resources regardless of their location (Fog Infrastructure)
- 9. Vision Storage CANBUS Compute Network GPU VPU FPGA TPU GPIO I2C Distributed Management (no central controller needed) TSN Sensors fogOS aims at providing a decentralised infrastructure for provisioning and managing (1) compute, (2) storage, (3) communication and (3) I/O resources available anywhere across the network. fogOS addresses highly heterogeneous systems even those with extremely resource-constrained nodes. 05fog
- 10. fogOS unifies the virtualisation infrastructure • Support of heterogeneous runtimes, hypervisors and networking • Common API and information model for management XEN LXD docker binaries Unified abstraction of virtualisation infrastructure µControllers ROS2 OCI Virt. Machines Containers Unikernels Monitoring Info. Model APIs Linux Windows STM32 Vision05fog
- 11. Architecture
- 12. Modules fogOS is divided in two modules: • Fog Orchestration Engine (FOrcE) • Fog Infrastructure Manager (FIM) Each layers provides different abstractions and APIs Entity Fog Deployment Unit Node Network Resource Port Router FOrcE FIM 05fog …
- 13. FOrcE Fog Orchestration Engine Provides functionalities for managing applications and services (Entities) in the Fog Infrastructure. Enforces constraints defined in the Entity descriptor Provides a first-fit distributed allocation algorithm for FDUs composing the Entity Entity Fog Deployment Unit Node Network Resource Port Router FOrcE FIM 05fog …
- 14. Entity Fog Deployment Unit Node Network Resource Port Router FIM Fog Infrastructure Manager Virtualises the hardware infrastructure, such as computational, communication, storage and I/O resources, and abstract the key primitives provided by system software, such as the OS Provides primitives for managing these virtualised infrastructure Provides infrastructure level monitoring information. FOrcE FIM 05fog …
- 15. Fog Infrastructure Manager (FIM) The set of abstraction provided by the FIM are: Resource, Node, Network, Router, Port, Fog Deployment Unit (FDU) agent agent FIM05fog
- 16. Resource A fogOS resource is any kind of asset that can be assigned either in a shared or exclusive manner. The different kinds of resources supported by fogOS are (1) computational resources, such as, CPU, FPGA, GPU, (2) storage resources, such as, RAM, disk and NAS, (3) networking resources such as Point-To-Point connections, 802.11 devices, network tunnels and (4) I/O resources, such as, COM, GPIO, and I2C.
- 17. AT() Node A fogOS node is represents a locus of resources. For instance, a server, an embedded board, and IoT device are examples of fogOS nodes. Each node has either a running fogOS agent or a device plugin. Cloud IaaS agent agent agent agent agent agent agent
- 18. AT() Fog Deployment Unit A Fog Deployment Unit (FDU) is an indivisible unit of deployment, such as, a binary executable, a uni-kernel, a container or a virtual machine. An FDU requires a certain kinds of resources as a precondition to its execution. The life-cycle of an FDU is defined by a Finite State Machine (FSM).
- 19. AT() Networks A fogOS network represents a level-2 or level-3 network between a collection of FDU fogOS networks can be isolated, interconnected or bridged/ routed to external networks UK C BE UK VM: Virtual Machine C: Container UK: Uni Kernel BE: Binary Executable uS: micro Service uS BE UKUK VM C
- 20. AT() Information Model fogOS’s information model defines the describes associated with nodes, entities and networks. Additionally it provides an abstract way to describe applications and relations between them.
- 21. Architecture fogOS is composed by: Communication fabric based on zenoh to support data distribution and sharing across the network Location-transparent state access based on YAKS to enable distributed access to system’s state Agent based closed-loop management of the device Plugin architecture for extensible hardware and software management zenoh YAKS AgentPlugins Network Data Link Physical Transport 05fog
- 22. AT() Deployments fogOS offers different deployments options: • ZYAP - armv7l, Linux, 57MB • YAP - armv7l, Linux, 45MB • AP - armv7l, Linux, 30MB • P - NUCLEO-F429ZI, arm-m, mBed, 58KB mbed-os-example- wifi Zenoh µController µController µController Host Host Host P PZenoh Broker YAKS Agent P P YAKS Agent P Agent 56KB 30MB 45MB
- 23. Some numbers
- 24. Container Deployment fogOS Container deployment time is better than Kubernetes and does not add much overhead above Docker and LXD. *
- 25. Service Chain Deployment fogOS service chaining deployment time scales far better than K8s Mediandeploymenttime(sec) 0 22,5 45 67,5 90 Service Chain Length 2 4 8 16 32 fog05 Docker k8s
- 26. use cases • ETSI MEC and ETSI NFV convergence • Fog assisted robotics • Vehicle to Everything • Augmented Reality/Virtual Reality 05fog
- 27. Plan
- 28. AT() Plan First release (0.1.0): • Updated website - https://fog05.io • containerd Plugin • K8s integration • Repository reorganisation • Improve documentation 05fog
- 29. Innovating Together Gabriele Baldoni Technologist ADLINK Tech. Inc. gabriele.baldoni@adlinktech.com
- 31. Demo
- 32. Back-up slides
- 33. How state is distributed fogOS system state can be separated in Public and Private state Public is the state composed by the union of the state exported by each node Private is the state each node does not share Public Private Actual Desired R/W R/W R/W R/W R/W R — — Me Others
- 34. How state is distributed fogOS system state can be separated in Actual and Desired state Actual is the current state of the system. Desired is the set point of the system. Public Private Actual Desired R/W R/W R/W R/W R/W R — — Me Others
- 35. AT() API fogOS’s provides API at different levels • Plugin API • FIM API • FOrcE API (*) And in different languages: • OCaml • Python • REST • Go (*) • Rust (*)