IEC Type 3: Android cloud native applications on Arm servers in edge for Integrated Edge Cloud (IEC) Blueprint Family

Attributes

Description

Informational

Type

New

Industry Sector

Telco, Cloud

Business driver

• Edge cloud requires initiatives for cloud gaming on Android platform  
• 5G + edge bring low latency and high-throughput for cloud gaming, which improves user experience
• More and more Android applications will migrate into edge compute platform. Building an android platform is necessary, and it's rigid demand. 

Business use cases

1.   Android Cloud Gaming
2.   AR/VR  Android Application

Business Cost - Initial Build Cost Target Objective

Using virtualization technology instead of mobile chip board on ARM server will reduce the cost of android platform.  ARM server board also let android platform more flexible and more agile.

In order to support cloud gaming and AR/VR, ARM server need have GPU inside. 

• 2 ARM Nodes
• Kubernetes v 1.17 for ARM
• Android 10.0
  
  • GPU (AMD, NVIDIA)
• Android virtualization layer: Anbox, docker.

Business Cost – Target Operational Objective

It more like a cloud platform, but it's specific for android application.  

• It needs Helm and Ansible for the automation and management tools to keep operational cost lower
• It needs to monitor the status of android system status and application status. 
• Maintain a mixed edge platform including x86 and ARM. It's more complex
• Kubernetes v1.17 for ARM
• Android 10.0
  • GPU (AMD, NVIDIA)
• Host OS kernel optimization;
• Android virtualization layer: Anbox, docker.
• GPU (AMD, NVIDIA)
• Both ARM and X86 can support it.

Security need

Regulations

N/A

Other restrictions

N/A

Additional details

N/A

Case Attributes

Description

Informational

Type

New

Blueprint Family - Proposed   Name

Use Case

Cloud Gaming and AR/VR

Blueprint proposed Name

IEC Type 3: Android cloud native applications on Arm servers in edge

Initial POD Cost (capex)

Scale & Type

For the smallest deployment, this requires 2 ARM bare metal machines. For large deployments, this could span to large number of bare metal machines.

Applications

Power Restrictions

N/A

Infrastructure orchestration

Host:

• Orchestrator: Kubernetes
• Bare Metal Provisioning：Ansible
• Kubernetes Provisioning：KuD
• OS: Ubuntu
• GPU Driver: AMD,NVIDIA:
• Network: OVS 
• Android virtualization layer: Anbox, docker.
• Host OS kernel optimization

Guest OS Andriod:

• GPU Driver (AMD, NVIDIA)

SDN

Workload Type

• Android applications 

Additional Details

N/A

Committer

Committer

Company

Committer Bio

Committer Picture

Self Nominate for PTL (Y/N)

CPU

2x 32 core 2.4 GHz Hi1616 Processors

2x 32 core 2.6 GHz Hi1620 Processors

RAM

8x 16GB RDIMM-Tx2 

8x 16GB DDR4-DIMM

Storage

2x 1TB SSD

2x 1TB SSD

Networking

2x 1/10Gbe BASE-T (1 connected)
1x IPMI  Lights-out Management

2x 1/10Gbe SFP+ (connected)

1x IPMI / Lights-out Management

GPU

1 AMD GPU ,1 NVidia GPU

Server Name

IPMI Address

Public Network Address

Switch Port(s)

OS Installed

93

10.x

10.x

gigabyte-edgecore2: left 40Bbe Port 1, right 40Gbe Port 2, left 10Gbe Port 7 Breakout 1,  right 10Gbe Port 7 Breakout 2

Ubuntu 18.04

174

10.x

10.x

gigabyte-edgecore2: left 40Bbe Port 3, right 40Gbe Port 4, left 10Gbe Port 7 Breakout 3, right 10Gbe Port 7 Breakout 4

Ubuntu 18.04