OpenStack
Huawei Elected as First Asian Platinum Member of the OpenStack Foundation

2017 OpenStack
Huawei Tops List of Cloud Vendors for OpenStack Users

2017 IDC
Huawei Ranked First in China’s OpenStack Software Market

Solution Overview

FusionSphere OpenStack is Huawei’s commercial OpenStack release with a built-in Huawei KVM virtualization engine based on open-source OpenStack. It incorporates various enterprise-level enhancements to its computing, storage, network management, installation and maintenance, security, and reliability resources. This solution is the optimal commercial OpenStack choice for enterprise private cloud, carrier NFV, and public cloud service providers.

FusionSphere OpenStack

The latest version (6.1), based on the community’s Mitaka release, includes the Nova, Glance, Ironic, Swift, Cinder, Neutron, Keystone, Ceilometer, and Heat core services.

Cloud Provisioning Service

A boot and provisioning module responsible for installing OSs on hosts, deploying service modules, monitoring service status, and arbitrating between active and standby services.

OpenStack OM

This unifies the management of resources that are visible to users in the cloud system and provides integrated resource management experience and Virtual Data Centers (VDCs). Automatic resource provisioning is available to implement automated VM deployment, including infrastructure deployment, automatic service deployment, and service monitoring. The O&M system also provides reports, rights, user management, maintenance, and fault management functions.

Extended VM HA

This High Availability (HA) enhancement was added to meet the elevated reliability requirements of NFV VMs.

VM Faults

VM HA is based on the results of VM monitoring. Faults include abnormalities, such as a system panic in the VM GuestOS or an infinite kernel loop.

Server Faults

When local VMs cannot be restarted, or when physical servers become faulty, VMs are rebuilt on other physical hosts that meet their requirements.

 

 

 

 

Shared Volumes

Logical volumes are attached to at least two VMs as shared volumes. Data can be read from or written to one or multiple volumes at the same time.

Scenarios

In consistent data access and hot standby settings, failover time is reduced and system availability is improved.

 

 

VM Bandwidth Configuration

Extended flavor APIs define VM bandwidth resources to meet the requirements of Virtualized Network Functions (VNFs). Bandwidth is a core resource competed for by products with high requirements, such as media servers and gateways. The community natives provide unified management and scheduling for general resources, such as memory and CPUs, but it cannot preschedule bandwidth resources, leading to bandwidth preemption. Huawei’s release uses nova-compute to update the bandwidth information on each node, extend flavor APIs, define VM bandwidth requirements, and add filters to screen nodes with efficient bandwidth resources in the scheduling phase, so that the high-bandwidth requirements of VNFs can be met.

 

 

 

  •  

  •  

 

 

 

 

 

  •  

 

 

 

  •  

 

 

 

 

 

 

 

Solution Overview

Solution Overview

FusionSphere OpenStack is Huawei’s commercial OpenStack release with a built-in Huawei KVM virtualization engine based on open-source OpenStack. It incorporates various enterprise-level enhancements to its computing, storage, network management, installation and maintenance, security, and reliability resources. This solution is the optimal commercial OpenStack choice for enterprise private cloud, carrier NFV, and public cloud service providers.

Solution Architecture

FusionSphere OpenStack

The latest version (6.1), based on the community’s Mitaka release, includes the Nova, Glance, Ironic, Swift, Cinder, Neutron, Keystone, Ceilometer, and Heat core services.

Cloud Provisioning Service

A boot and provisioning module responsible for installing OSs on hosts, deploying service modules, monitoring service status, and arbitrating between active and standby services.

OpenStack OM

This unifies the management of resources that are visible to users in the cloud system and provides integrated resource management experience and Virtual Data Centers (VDCs). Automatic resource provisioning is available to implement automated VM deployment, including infrastructure deployment, automatic service deployment, and service monitoring. The O&M system also provides reports, rights, user management, maintenance, and fault management functions.

Enterprise-Level Enhancements

Extended VM HA

This High Availability (HA) enhancement was added to meet the elevated reliability requirements of NFV VMs.

VM Faults

VM HA is based on the results of VM monitoring. Faults include abnormalities, such as a system panic in the VM GuestOS or an infinite kernel loop.

Server Faults

When local VMs cannot be restarted, or when physical servers become faulty, VMs are rebuilt on other physical hosts that meet their requirements.

 

 

 

 

Shared Volumes

Logical volumes are attached to at least two VMs as shared volumes. Data can be read from or written to one or multiple volumes at the same time.

Scenarios

In consistent data access and hot standby settings, failover time is reduced and system availability is improved.

 

 

VM Bandwidth Configuration

Extended flavor APIs define VM bandwidth resources to meet the requirements of Virtualized Network Functions (VNFs). Bandwidth is a core resource competed for by products with high requirements, such as media servers and gateways. The community natives provide unified management and scheduling for general resources, such as memory and CPUs, but it cannot preschedule bandwidth resources, leading to bandwidth preemption. Huawei’s release uses nova-compute to update the bandwidth information on each node, extend flavor APIs, define VM bandwidth requirements, and add filters to screen nodes with efficient bandwidth resources in the scheduling phase, so that the high-bandwidth requirements of VNFs can be met.

 

 

 

  •  

  •  

 

 

 

 

 

  •  

 

 

 

  •