Build your private cloud with OpenStack and HPE solutions and services

[ServicesExperts]

OpenStack is a proven, enterprise-grade private cloud technology that’s cost-effective, scalable and flexible. HPE can help you get the most from it.

By Kyaw Thiha, Chief Solution Architect, Cloud Native Computing Practice Area, HPE Advisory & Professional Services

Hyperscalers have seen healthy growth rates exceeding 30% for the past few years. Many enterprise leaders thought that all workloads would move to public cloud environments and never come back on-premises. However, there are some workloads which are not able to be migrated to the hyperscalers, and some which have returned to on-premises environments because of factors such as organizational policies, data sovereignty, security, compliance, total cost of ownership and workload requirements. For such cases, a scalable, flexible, secure, affordable, and configurable on-premises private cloud environment is required. (See: What is private cloud?)

What is available now in the market? There are hyperconverged infrastructure (HCI) solutions, software-defined stacks, and on-premises appliances from a variety of vendors.  However, many of them lack the scalability, flexibility, and customizability of OpenStack. 

OpenStack is the most popular and mature open-source private cloud solution. Telecommunication companies, large enterprises, governments, and research institutions are continuously adopting the technology for their private cloud environments and contributing to its technical evolution through community activities.

OpenStack continues to mature. The following are some benefits of its use:

• Cost effectiveness:  OpenStack provides transparent and predictable costs, even with the growing requirements of workloads. That’s especially helpful for workloads which will be running long-term, at scale, and for controlling operational costs.
• Flexibility: A private cloud powered by OpenStack is fully under the control of the customer organization. It is designed and implemented to meet the organization’s workload requirements, security compliance and organizational policies.
• Interoperability: It provides software-defined networking (SDN), hardware, load balancing and integration with external or in-house solutions.
• Shorter time-to-market: With OpenStack’s software-defined infrastructure and self-service portal, deployment of IT resources can be completed within minutes. End users can start using the network and applications services in a shorter time.
• No vendor lock-in: OpenStack can be implemented on commodity hardware, which offers the freedom of choosing hardware vendors, equipment and scale.
• Enterprise support: OpenStack is open-source, but businesses can still get enterprise support from RedHat and Canonical when using their distributions.

These benefits are possible because OpenStack is a set of software components developed in a parallel manner by its open community. Behind the scenes, many open-source projects are integrated and working together as one OpenStack solution. Major components include:

• Nova, an Infrastructure-as-a-Service (IaaS) cloud computing platform which gives control over instances
• Neutron, which provides Networking-as-a-Service
• Cinder, software-defined storage which provides block storage
• Keystone, which offers service discovery, authentication, and authorization services.

The set of software components is developed through open community activities and based on the needs of users in the field. The development is not only for major IT components, but also for specific use cases such as high-performance computing for the telecommunication industry and SDN integration for the service provider industry.

High-performance workloads

OpenStack offers high performance capabilities for workloads with HugePage, CPU pinning and single root I/O virtualization (SR-IOV) implementation.

• HugePage: The page size is typically 4 KB for general purpose computing. With this OpenStack function, you can configure page size to 2 MB and 1 GB. These larger page sizes are known as huge pages. By using huge pages, virtual-to-physical address translation is reduced, increasing performance.
• CPU pinning: Virtual machine instances can run on a dedicated CPU. This enables smarter scheduling and improves guest performance. 
• Single root I/O virtualization: This allows near-bare-metal performance by allowing instances from OpenStack direct access to a shared PCIe (peripheral component interconnect express) resource through virtual resources. For better performance of SR-IOV VMs, deploy guests that have CPU pinning and huge pages.

SDN integration

OpenStack Neutron, the Networking-as-a-Service project, handles the creation and management of a virtual networking infrastructure, including networks, switches, subnets, and routers. Most telco, enterprise and financial enterprises are using SDN solutions and considering integrating with OpenStack.

Cisco’s Application Centric Infrastructure (ACI) is one of the SDNs which we integrate with OpenStack frequently. ACI provides an intelligent, controller-based network switching fabric with policy-based architecture. OpenStack can integrate and manage ACI, which allows dynamic creation of networking constructs to be driven directly from OpenStack. It provides the flexibility of software overlay networking with the performance and operational benefits of hardware-based networking.

How HPE can help you build it

OpenStack remains one of the most mature open-source-based cloud platform software solutions and has been continuously evolved to match field needs and industry use cases.

Hewlett Packard Enterprise has extensive offerings to fulfil customers’ needs around open, flexible, and agile private cloud platforms leveraging OpenStack. HPE ProLiant DL servers, HPE 3PAR StoreServ Storage, HPE Primera, and HPE Alletra 9000 are validated for OpenStack Platform.

Also, HPE has validated its hardware for specific use cases, for example with HPE Telco Blueprints. HPE Telco Blueprints are available for telco implementations with the desired scale. HPE ProLiant DL360 and DL380 are used in HPE Telco Blueprints.

HPE Advisory & Professional Services has strong expertise and experience backed by thousands of code contributions and hundreds of OpenStack project deliveries around the globe. The Global Cloud Native Computing Practice within HPE Advisory & Professional Services guides enterprises on their open-source-software-based cloud transformation journey with a proven methodology and vast experience/expertise.

Learn more about advisory and professional services from HPE Pointnext Services.

Kyaw Thiha has been working with HPE for four years as a Chief Solution Architect. He has over fifteen years’ experience in developing, designing and implementing enterprise IT solutions, including container platforms, hybrid cloud systems, cloud management platforms and private cloud. His main areas of interests are containers, virtualization, hybrid cloud, automation and workload transformation. Kyaw has helped telco and enterprise customers by providing architectural and design consultation to build high-performance clouds. He has extensive experience on workload onboarding to cloud platforms.

Services Experts
Hewlett Packard Enterprise

twitter.com/HPE_Pointnext
linkedin.com/showcase/hpe-pointnext-services/
hpe.com/pointnext