Time-Critical Edge Compute

Time-Critical Edge Compute

Project Technical Lead: Shane Dewing. Elected 1/17/19.

Project Committers detail:

Initial Committers for a project will be specified at project creation. Committers have the right to commit code to the source code management system for that project.

 

 

A Contributor may be promoted to a Committer by the project’s Committers after demonstrating a history of contributions to that project.

Candidates for the project’s Project Technical Leader will be derived from the Committers of the Project. Candidates must self nominate by marking "Y" in the Self Nominate column below by Jan. 16th. Voting will take place on January 17th.

Only Committers for a project are eligible to vote for a project’s Project Technical Lead.

 

Please see Akraino Technical Community Document section 3.1.3 for more detailed information.

 

Committer

Committer

Company

Committer

Contact Info

 Committer Bio

Committer Picture 

Self Nominate for PTL (Y/N)

Shane Dewing

Intel

shane.dewing@intel.com 

At Intel Corp, Shane works in the Industrial Solutions Division of the Internet of Things Group, where he leads software and open source strategy. Shane is a veteran software developer and technologist, and was on the technical leadership team of the AllSeen Alliance and the AllJoyn open source project (which Shane helped launch in 2013). Shane also works in IOT and Industrial Standards, including the Open Process Automation Forum, the OPC Foundation, and the Open Connectivity Foundation, where Shane chaired the Industrial TG 2015-2017.

Y

 

HPE

 

 

 

 

 

Dell

 

 

 

 

Khemendra Kumar

Huawei

khemendra.kumar@huawei.com

 

 

 

Keith Steele

IOTech Systems

keith@iotechsys.com

Keith is the CEO of IOTech. He is an EdgeX Foundry Board Member and the Chair of the Technical Steering Committee at EdgeX Foundry

 

 

Steve Osselton

IOTech Systems

steve@iotechsys.com

Steve Osselton is CTO at IOTech. Steve is Chair of the Device and Device SDK Working Group at EdgeX Foundry

 

 

Kevin Midkiff

Intel

kevin.midkiff@intel.com

Kevin is a lead architect in the Industrial Solutions Division at Intel

 

 

Jerry Yu

Intel

jerry.yu@intel.com

Jerry works in the System Software Products group at Intel

 

 

Sanrio Alvares

Intel

sanrio.alvares@intel.com

Sanrio works in the System Software Products group at Intel

 

 

 Kailas Maneparambil

 Intel

 Kailasnath.s.maneparambil@intel.com

 Kailas is a system architect in the Industrial Solutions Division

 

 

Taimoor Imtiaz

Intel

taimoor.imtiaz(at)intel.com

Taimoor is a Systems Engineer at Intel's Industrial Solutions Division.

 

 

Tiejun Chen

VMware

tiejunc@vmware.com

Technical leader and staff II engineer

 

 

 

Time-Critical Edge Compute git repo: https://gerrit.akraino.org/r/#/admin/projects/tc

Community Call Notes: https://etherpad.akraino.org/p/tc

 

Akraino New Use Case Template:

Attributes

Description

Informational

Type

Time Critical Edge Compute

Functionally safe capable for Industrial implementations, and enabled for Real-Time

Industry Sector

Industrial Edge with applicability to applications where time-sensitive compute and/or functional safety are required

 

Business driver

Industrial Edge is shifting dramatically with manufacturers, oil and gas, energy and other major industries transitioning to software defined infrastructure just as the telecommunications sector has been undergoing a similar transformation.  Akraino has the opportunity to play a critical role in hosting scalable, tuned software defined infrastructure for Industrial potentially ushering in this transformation and coalescing the ecosystem.

 

Business use cases

Use cases in Manufacturing, Smart Buildings, general IIOT.

  • Virtualized PLC

  • Computer vision inference

  • Machine, sensor data inference

  • Process or discrete manufacturing closed loop control

  • Ethernet TSN

  • Functional Safety capable use cases

  • Discrete manufacturing soft PLC

  • Onramp for 5G-URLLC

 

Business Cost - Initial Build Cost Target Objective

Example: Dell 3000 - ~$800 list

Example: Dell 5000 - ~$1300 list

Example: HPE EL300 - ~$2500 list

 

Business Cost – Target Operational Objective

Leverage cloud-native and other IT infrastructure to enable end-users to deploy time-critical and functionally safe applications at the industrial edge on general purpose compute HW.

 

 

Security need

Functional Safety (FuSa) core OS certification.

Security for Industrial is a key vector, which is why this stack is in essence functionally safe - by using open source leadership software such as Zephyr and ACRN - projects designed to meet both functional safety and real time requirements for security at the core of the stack.  By leveraging Kata Containers, all container workloads can leverage VT for containerized workload isolation.

Trusted Execution Environment compatible.

 

Regulations

IEC 61508 - Safety Integrity Level

 

Other restrictions

For industrial deployments:-

  • IEC-600068-2-27, IEC 600068-2-64

  • IP50 rated

  • Operational humidity: 8 to 90%, noncondensing

  • Tested in accordance with IEC-600068-2-56 Non-operating humidity: 5 to 95%, noncondensing

  • Tested in accordance with IEC-600068-2-56

  • MIL – 810G Tested

  • Storage temperature range: -40˚C to +70˚C

  • MIL-PRF-28800F class 3 limits

  • Operational, -30˚C to 70˚C (-22˚F to 158˚F)

 

Additional details

Differentiation:

Ruggedized implementation.  Supports functional safety workloads.

Potential to attract new members to Akraino project

○ Industrial ODM’s e.g. Advantech, Adlink

○ Industrial OEM’s/ISV’s e.g. TTTech, Nebbiolo, IOTech

○ Industrial end-users e.g. ExxonMobil

 

Akraino Species Blueprint Template:

Case Attributes

Description

Informational

Type

New

 

Blueprint Family - Proposed   Name

TBD

 

Use Case

Use cases in Manufacturing, Smart Buildings, general IIOT.

  • Virtualized PLC

  • Computer vision inference

  • Machine, sensor data inference

  • Process or discrete manufacturing closed loop control

  • Ethernet TSN

  • Functional Safety capable use cases

  • Discrete manufacturing soft PLC

  • Onramp for 5G-URLLC

 

Blueprint proposed Name

Time-Critical Edge Compute

 

Initial POD Cost (capex)

Example: Dell 3000 - ~$800 list

Example: Dell 5000 - ~$1300 list

Example: HPE EL300 - ~$2500 list

 

Scale & Type

Intel Atom/Core; ARM -- Ruggedized hardware, low power footprint, real time, safety critical implementations.

 

Applications

Can span from functionally safe applications, to real time and time-sensitive applications, Android apps and containerized edge workloads orchestrated by Kubernetes.  

Validated applications will include OpenVINO for visual inference workloads, TensorFlow via Kubeflow for Machine learning workloads, Closed loop control, Software Programmable Logic Controllers (PLCs) and EdgeX Foundry

 

Power Restrictions

  • Industrial temp

  • Low power

  • Fanless design

 

Infrastructure orchestration

Kubernetes orchestrated Docker and Kata container formats

 

SDN

As needed by use case, can support OVS-DPDK

 

Workload Type

VM and Containers in a Type1 Hypervisor, ACRN - https://projectacrn.org/

Safety/Host OS - Zephyr - https://www.zephyrproject.org/

For Industrial, Windows support is critical, and therefore Windows will also be validated on ACRN as part of this blueprint.

 

Additional Details

See deck and demo link

 

Presentation:

 

Video Demo of this stack 

Embedded below (Open the URL to be able to view in fullscreen https://youtu.be/1qkRJlulUSY)