-Add in a description of the family, what makes the species linked together
- Needs to focus on Industrial IoT use cases at the Smart Device Edge (eg IoT edge compute deployed outside of the data center), not mobile devices/client
- There can be other families that cover other smart device edge use cases-
- refer a lot to the white paper
- 256 MB single node to a small server cluster
- gateway, router, hub, server
- not just about gateways,
- This would be covered by the smart device edge under the User edge- Taxonomy Whitepaper page 3
is there a min hardware? maybe as small as a RPi, but really focused on true industrial machinesneed some type compute functionality at the edge
- agnostic to silicon (eg ARM and x86 cpu, tpu, gpu, etc)
- comprehend the unique needs of the Industrial IoT edge compute use cases
- latency and safety critical applications
- priority for uptime and safety
- prevention of sending control messages back to critical infrastructure from the cloud
- comprehension of OT protocols
- a mix of on prem and cloud back end systems...
- deployment of legacy VM and apps plus modern containers
- add in a couple more UCs for this
- add in alignment with Akraino blueprints
- describe a baseline infrastructure
- horizontal infrastructure- market-tecture need to scope this out; draw the picture
- support to workload themes
- computer vision, eg stream video
- high bandwidth is needed
- the killer app is to process this at the edge
- machine intelligence, eg structured telemetry
- many different protocols,
- varying amounts of data volume (eg vibration data vs temperature data)
- computer vision, eg stream video
- enable workload consolidation
Need to update the text in the table to reflect the changes from the taxonomy
Use Case Details:
<add in the taxonomy image from the white paper>
add in text explaining the smart device edge
Family- IoT Device Edge-
Use Case Attributes | Description | Informational |
Type | New-approved March 2020 | |
Blueprint Family - Proposed Name | Industrial IoT at the Smart Device Edge | There are many possible UCs that could be done at the Industrial Smart Device Edge, so this might |
Use Case | Predictive Maintenance | This family is focused on how to bring data into the smart device edge and then do some type of compute there on the device. |
Blueprint proposed | Predictive Maintenance- Using a thermal imaging camera | This is the first proposed blueprint |
Initial POD Cost (capex) | Varies widely depending on the Blueprint | |
Scale of Servers | Smart Device Edge size | Largest would be an IoT Gateway, with the smallest being the compute power of a Raspberry Pi. These are very limited devices as compared to machines that are currently found at the On-Prem Data Center Edge. |
Applications (Edge Virtual Network Functions) | EVE, Fledge | |
Power Restrictions | None/Varies |
|
Preferred Infrastructure orchestration | Docker/K8 - Container Orchestration OS - Linux | |
Additional Details |
This is for the Family, Blueprint creators are welcome to join by adding their name.
Committer | Committer Company | Committer Contact Info | Committer Bio | Committer Picture | Self Nominate for PTL (Y/N) |
@bill hunt | Dianomic | ||||
Shiv Ramamurthi | Arm | Shiv.Ramamurthi@arm.com | |||
Cplus Shen | Advantech | ||||
| Ashwin Gopalakrishnan | Dianomic | ashwin@dianomic.com | |||
| Erik Nordmark | Zededa | erik@zededa.com | |||
| Daniel Lazaro | OSIsoft | dlazaro@osisoft.com | |||
| Aaron Williams | LF Edge | aaron@lfedge.org | Y | ||
| Vladimir Suvorov | AI Solutions | hello.fleandr@gmail.com |
Contributors: