HPE Reference Configuration for Red Hat OpenShift on HPE Synergy and Red Hat Hyperconverged Infrastructure

Reference Architecture

Reference Architecture

Contents Executive summary ................................................................................................................................................................................................................................................................................................................................ 3 Introduction ................................................................................................................................................................................................................................................................................................................................................... 3 Solution overview ..................................................................................................................................................................................................................................................................................................................................... 4 Solution components ............................................................................................................................................................................................................................................................................................................................ 6

Hardware ................................................................................................................................................................................................................................................................................................................................................... 6 Software ..................................................................................................................................................................................................................................................................................................................................................... 7

Best practices and configuration guidance for the solution ............................................................................................................................................................................................................................. 9 Capacity and sizing ............................................................................................................................................................................................................................................................................................................................ 15 Summary ...................................................................................................................................................................................................................................................................................................................................................... 16 Implementing a proof-of-concept ......................................................................................................................................................................................................................................................................................... 16 Appendix A: Bill of materials ...................................................................................................................................................................................................................................................................................................... 17 Resources and additional links ................................................................................................................................................................................................................................................................................................ 19

Reference Architecture Page 3

Executive summary In today’s digital world, organizations are under increasing pressure to deliver applications faster while reducing costs. As these applications grow more complex, this puts stress on IT infrastructure, IT teams, and processes. To remain competitive, organizations must adapt quickly, and developers need to be more effective, efficient, and agile. Container technology provides the right application platform to help organizations become more responsive and iterate across multiple IT environments as well as develop, deploy and manage applications faster. But implementing a containerized environment across existing infrastructure is a complex undertaking that can require weeks or months to mobilize, particularly for enterprises. To help accelerate container application delivery, HPE and Red Hat® are collaborating to optimize Red Hat OpenShift Container Platform on HPE platforms, including HPE Synergy, the industry’s first composable infrastructure.

Red Hat OpenShift Container Platform on HPE Synergy provides an end-to-end fully integrated container solution that, once assembled, can be configured within hours. This eliminates the complexities associated with implementing a container platform across an enterprise data center and provides the automation of hardware and software configuration to quickly provision and deploy a containerized environment at scale. Red Hat OpenShift Container Platform provides organizations with a reliable platform for deploying and scaling container-based applications and HPE Synergy provides the flexible infrastructure you need to run that container platform to dynamically provision and scale applications, whether they run in VMs or containers, or are hosted on-premises, in the cloud, or somewhere in between.

This Reference Configuration provides architectural guidance for deploying an entry level PaaS configuration based on Red Hat OpenShift environment on Hewlett Packard Enterprise Synergy Composable Infrastructure along with Red Hat Hyperconverged Infrastructure (RHHI) and Red Hat OpenShift Container Storage (OCS).

This Reference Configuration describes:

• How to leverage HPE Synergy strengths in rapid provisioning along with the automation capabilities of Red Hat Ansible to automate many of the manual tasks involved in the deployment of a Red Hat OpenShift Container Platform on a single frame HPE Synergy.

• How to efficiently layout an OpenShift configuration in a virtualized environment using Red Hat Hyperconverged Infrastructure.

• How to configure container persistent storage using Red Hat OpenShift Container Storage and HPE Synergy D3940 Storage Module.

This Reference Configuration demonstrates the following benefits of utilizing HPE Synergy for Red Hat OpenShift Container Platform:

• Automated initial installation of a highly available Red Hat OpenShift Container Platform on HPE Synergy reduces the initial solution deployment from several days as provided by a services organization to less than a day.

• Deployed the OpenShift master, infrastructure, and application nodes on virtual machines running on Red Hat Hyperconverged Infrastructure to optimize resource usage in a single Synergy Frame.

Target audience: Chief Information Officers (CIOs), Chief Technology Officers (CTOs), data center managers, enterprise architects and implementation personnel wishing to learn more about Red Hat OpenShift Container Platform on HPE Synergy Composable Infrastructure. Familiarity with HPE Synergy, Red Hat OpenShift Container Platform, container-based solutions, Ansible Engine, and core networking knowledge is assumed.

Document purpose: The purpose of this document is to describe a Reference Configuration that describes the benefits and technical details of deploying an entry level PaaS configuration consisting of Red Hat OpenShift on HPE Synergy Composable Infrastructure with Red Hat Hyperconverged Infrastructure and Red Hat OpenShift Container Storage.

Introduction This Reference Configuration describes a Red Hat OpenShift Container Platform deployment on HPE Synergy Composable Infrastructure and includes details on how the environment was configured. When combined with the accompanying Deployment Guide (https://github.com/HewlettPackard/hpe-solutions-openshift/tree/master/synergy/entry-level-config/rhhi), it provides a comprehensive example demonstrating how Red Hat OpenShift Container Platform can be set up to take advantage of the HPE Synergy composable architecture and Red Hat Hyperconverged Infrastructure in a highly available environment. The configuration consists of three (3) OpenShift Container Platform master instances, three (3) OpenShift Container Platform infrastructure instances, and three (3) OpenShift Container Platform application node instances all running as virtual machines on Red Hat Hyperconverged Infrastructure. This configuration uses a single (1) frame HPE Synergy with three (3) compute modules installed with Red Hat Virtualization configured in a Red Hat Hyperconverged Infrastructure cluster to support

Reference Architecture Page 4

this deployment. Red Hat OpenShift Container Storage is used to provide persistent storage for the OpenShift Container Registry and the container applications that require persistent storage.

The HPE Synergy platform is designed to bridge traditional and cloud-native applications with the implementation of composable infrastructure. Composable infrastructure combines the use of fluid resource pools, made up of compute, storage, and fabric with software-defined intelligence. Composable pools of compute, storage, and fabric can be intelligently and automatically combined to support any workload. The resource pools can be flexed to meet the needs of any business applications including containerized workloads.

Solution overview This Reference Configuration provides an overview of an entry level / PoC deployment of Red Hat OpenShift Container Platform on HPE Synergy and Red Hat Hyperconverged Infrastructure with Red Hat OpenShift Container Storage (OCS) solution as described in greater detail at https://github.com/HewlettPackard/hpe-solutions-openshift/tree/master/synergy/entry-level-config/rhhi. Figure 1 provides an overview of the solution components.

Figure 1. Solution layout

The solution deploys Red Hat OpenShift Container Platform 3.10 on virtual machines provided by Red Hat Virtualization deployed in a Red Hat Hyperconverged Infrastructure three (3) node cluster. The OpenShift Master, Infrastructure, application nodes, and load balancer are deployed as virtual machines running on three (3) HPE Synergy 480 Gen10 Compute Modules running Red Hat Virtualization Host 4.2 (RHVH) configured in a Red Hat Hyperconverged Infrastructure cluster and managed by Red Hat Virtualization Manager (RHV-M). The hypervisor, Red Hat Virtualization Host (RHVH) operating system is installed via Kickstart files over PXE and post-installation configuration steps are performed, in part, using Ansible. Red Hat Open Container Storage and HPE D3940 Storage Module provides support for persistent container volumes.

Reference Architecture Page 5

The HPE Converged Architecture 750 (CA750) was used as the reference platform for the Red Hat OpenShift deployment. Customers can also customize their OpenShift configuration based on their workload needs, without using or just leveraging parts of the CA750 design. The CA750 approach provides pre-integrated, modular, scalable converged systems that reduce deployment risk. To implement this solution as an HPE CA750, please work with your HPE Authorized Channel Partner.

Figure 2 shows the layout of the hardware within the rack.

Figure 2. Front view of the solution with individual components highlighted

Reference Architecture Page 6

The solution environment assumes the presence of certain products and services to insure proper functionality.

• Ansible Engine

• DHCP

• DNS

• NTP

• TFTP/PXE

Solution components Hardware The following hardware components were utilized in this Reference Configuration as listed in Table 1.

Table 1. Components utilized in the creation of this solution

Component Qty Description

HPE Synergy 12000 Frame

– HPE Synergy Composer

– HPE Virtual Connect SE 40Gb F8 Module

– HPE Synergy 12Gb SAS Connection Module

1

1

2

2

One (1) HPE Synergy 12000 Frame houses the infrastructure used for the solution

Core configuration and lifecycle management for the Synergy components

Ttwo (2) HPE Virtual Connect SE 40Gb F8 Modules provide network connectivity into and out of the frame

Two (2) SAS Connection Modules providing connectivity for the D3940 Storage Module

HPE Synergy 480 Gen10 Compute Module 3 Three (3) Red Hat Virtualization Hosts (RHVH) for Red Hat Hyperconverged Infrastructure

HPE D3940 Storage Module 1 One (1) D3940 Storage Module to provide storage for Gluster based RHHI and OCS persistent storage

HPE FlexFabric 5940 2-Slot Switch

– HPE 5930 24p SFP+ and 2p QSFP+ Module

– HPE 5930 8p QSFP+ Module

2

2

2

Each switch contains one (1) each of the HPE 5940 modules listed below

One module per HPE FlexFabric 2-Slot Switch

One module per HPE FlexFabric 2-Slot Switch

HPE Synergy HPE Synergy, the first platform built from the ground up for composable infrastructure, empowers IT to create and deliver new value instantly and continuously. This single infrastructure reduces operational complexity for traditional workloads and increases operational velocity for the new breed of applications and services. Through a single interface, HPE Synergy composes compute, storage, and fabric pools into any configuration for any application. It also enables a broad range of applications from bare metal to virtual machines to containers, and operational models like hybrid cloud and DevOps. HPE Synergy enables IT to rapidly react to new business demands.

HPE Synergy Frames contain a management appliance called the HPE Synergy Composer which hosts HPE OneView. HPE Synergy Composer manages the composable infrastructure and delivers:

• Fluid pools of resources, where a single infrastructure of compute, storage, and fabric boots up ready for workloads and demonstrates self-assimilating capacity.

• Software-defined intelligence, with a single interface that precisely composes logical infrastructures at near-instant speeds; and demonstrates template-driven, frictionless operations.

• Unified API access, which enables simple line-of-code programming of every infrastructure element; easily automates IT operational processes; and effortlessly automates applications through infrastructure deployment.

HPE Synergy Composer provides the enterprise-level management to compose and deploy system resources to your application needs. This management appliance uses software-defined intelligence to aggregate compute, storage, and fabric resources in a manner that scales to your application needs, instead of being restricted to the fixed ratios of traditional resource offerings. HPE Synergy template-based provisioning enables fast time to service with a single point for defining compute module state, pooled storage, network connectivity, and boot image.

HPE OneView is a comprehensive unifying platform designed from the ground up for converged infrastructure management. A unifying platform increases the productivity of every member of the internal IT team across servers, storage, and networking. By streamlining processes,

Reference Architecture Page 7

incorporating best practices, and creating a new holistic way to work, HPE OneView provides organizations with a more efficient way to work. It is designed for open integration with existing tools and processes to extend these efficiencies.

HPE OneView is instrumental for the deployment and management of HPE servers and enclosure networking. It collapses infrastructure management tools into a single resource-oriented architecture that provides direct access to all logical and physical resources of the solution.

Logical resources include server profiles and server profile templates, enclosures and enclosure groups, and logical interconnects and logical interconnect groups. Physical resources include compute modules, interconnects, and storage modules.

The HPE OneView converged infrastructure platform offers a uniform way for administrators to interact with resources by providing a RESTful API foundation. The RESTful APIs enable administrators to utilize a growing ecosystem of integrations to further expand the advantages of the integrated resource model that removes the need for the administrator to enter and maintain the same configuration data more than once and keep all versions up to date. It encapsulates and abstracts many underlying tools behind the integrated resource model, so the administrator can operate with new levels of simplicity, speed, and agility to provision, monitor, and maintain the solution.

Within the context of the solution, HPE OneView for Synergy is utilized to:

• Configure the profiles of the HPE Synergy Compute Modules.

• Apply and maintain compliance for firmware across the HPE Synergy infrastructure.

• Configure networking from the HPE Synergy Compute Modules to internal and outbound destinations.

HPE Synergy 480 Gen10 The HPE Synergy 480 Gen10 Compute Module delivers an efficient and flexible two-socket workhorse to support the most demanding workloads. Powered by Intel® Xeon® Scalable Family of processors, up to 3TB DDR4, and large storage capacity within a composable architecture. HPE Synergy 480 Gen10 Compute Module:

• Is the most secure server with exclusive HPE Silicon Root of Trust. Protect your applications and assets against downtime associated with hacks and viruses.

• Offers customer choice for greater performance and flexibility with Intel Xeon Scalable Family of processors on the Synergy 480 Gen10 architecture.

• Offers Intelligent System Tuning with processor smoothing and workloads matching to improve processor throughput/overall performance up to 8% over previous generation.

• Features a maximum memory footprint of 3TB for large in-memory database and analytic applications.

• Features a hybrid HPE Smart Array for both RAID and HBA zoning in a single controller.

• Is more performant (over previous generation) with:

– A 25% performance increase with 1 or 2 processors (up to 205W).

– An 11% increase in memory speeds with a wide range of memory sizes.

– A 2.5x increase in compute and network connectivity speed gets your data where it's needed faster.

– Up to 50% performance improvement in random read IOPs in storage controllers.

The HPE Synergy 480 Gen10 provides the needed compute to power this solution running both Red Hat Virtualization for the core management pieces of Red Hat OpenShift and Red Hat Enterprise Linux to host the worker nodes.

The bill of materials found in Appendix A of this document outlines the configuration of the HPE Synergy Compute Modules used in this solution.

Software Red Hat Enterprise Linux Red Hat® Enterprise Linux® Server powers the applications that run your business with the control, confidence, and freedom that comes from a consistent foundation across hybrid cloud deployments. As the premier platform provider for enterprise workloads, Red Hat works side by side with engineers from major hardware vendors and cloud providers to make sure that the operating system takes full advantage of the latest innovations. This leadership with partners, as well as Red Hat’s influence and contributions to upstream communities, provides a stable, secure,

Reference Architecture Page 8

and performance driven foundation for the applications that run the business of today and tomorrow.1 Red Hat Enterprise Linux is at the core of this solution; each of the Red Hat OpenShift Container Platform control plane nodes running as virtual machines are running Red Hat Enterprise Linux.

Red Hat Hyperconverged Infrastructure for Virtualization Red Hat Virtualization is an open, software-defined, efficient platform for virtualized Linux and Microsoft® Windows® workloads, built on Red Hat Enterprise Linux and Kernel-based Virtual Machine (KVM) technologies. Automated management, scaling, and security features let you build a streamlined, reliable environment for virtualized applications and cloud-native workloads. Virtual-to-virtual (V2V) conversion tooling lets you easily migrate workloads from other hypervisors. A RESTful application programming interface (API) lets you integrate Red Hat Virtualization into your existing infrastructure and easily add innovative new technologies as business requirements change.2 This solution is designed to optimize the use of physical resources by using a combination of virtual machines and physical servers. Red Hat Virtualization provides a robust highly available virtualization platform for the Red Hat OpenShift Container Platform virtual machines. The Red Hat OpenShift Container Platform control plane, including the master, etcd, infrastructure, and load balancer roles, and application nodes are deployed on virtual machines that are distributed across a three node RHV cluster.

Red Hat OpenShift Container Platform Red Hat OpenShift Container Platform unites developers and IT operations on a single platform to build, deploy, and manage applications consistently across hybrid cloud and multicloud infrastructures. Red Hat OpenShift helps businesses achieve greater value by delivering modern and traditional applications with shorter development cycles and lower operating costs. Red Hat OpenShift is built on open source innovation and industry standards, including Kubernetes and Red Hat Enterprise Linux, the world’s leading enterprise Linux distribution.3

Figure 3 highlights how the individual Red Hat OpenShift Container Platform pieces are laid out within the solution.

Figure 3. Red Hat OpenShift solution component layout

1 redhat.com/cms/managed-files/li-enterprise-linux-server-datasheet-f11191jm-201803-en_1.pdf 2 redhat.com/cms/managed-files/vi-virtualization-datasheet-f12281-201805-en.pdf 3 redhat.com/cms/managed-files/cl-openshift-container-platform-datasheet-f9695kc-201711-en.pdf

Reference Architecture Page 9

Red Hat OpenShift Container Storage Red Hat OpenShift Container Storage is a Gluster based software defined storage solution that provides persistent storage for container-based applications. In this solution OpenShift Container Storage is used to provide persistent volumes for the OpenShift Container registry and container-based applications that require persistent storage. There are two (2) OpenShift Container Storage deployments in this solution, one instance running on the three (3) infrastructure nodes for the OpenShift Container Registry, logging, and metrics persistent storage and a second instance of OpenShift Container Storage running on the three (3) application nodes for container-based applications requiring persistent storage volumes.

Best practices and configuration guidance for the solution This section discusses the high-level cabling and configuration of the solution hardware and software. For a detailed explanation of how to build and deploy the entire solution stack, consult the deployment guide and accompanying Ansible deployment scripts at https://github.com/HewlettPackard/hpe-solutions-openshift/tree/master/synergy/entry-level-config/rhhi.

Networking Table 2 describes the configuration of the networks as defined within HPE OneView for Synergy and the bandwidth associated with each network.

Table 2. Networks defined within HPE Synergy Composer for this solution

Network Name Type VLAN Number

Purpose Requested Bandwidth (Gb)

Maximum Bandwidth (Gb)

Management Ethernet 10 Solution management 8 20

Public Ethernet 194 Application, authentication and other user networks 8 20

Storage Ethernet 20 Gluster Storage 8 20

PXE_Boot Ethernet 30 PXE boot for compute 1 20

Table 3 below describes the cabling of the HPE Virtual Connect SE 40Gb F8 Modules for Synergy to the switches and highlights what networks are carried on the connections. All ethernet networks, as described in Table 2 above, are carried on the Uplink Set labeled “Network”.

Table 3. Network Uplink Sets

Uplink Set Synergy Source

Network Enclosure 1, Bay 3, Port Q5

Enclosure 1, Bay 3, Port Q6

Table 4 below describes the bridge aggregation groups and the networks they contain.

Table 4. VLAN definitions by Bridge Aggregation Group

Network Function VLAN Number Bridge Aggregation Group

PXE 10 3

Public 194 3

Management 20 3

Storage 30 3

Reference Architecture Page 10

By utilizing HPE Synergy, the non-storage networks within the solution are able to traverse the Synergy infrastructure in an east-west fashion across high speed, low latency links both within and between HPE Virtual Connect Modules. In particular, communication between the core OpenShift management pieces remains within the HPE Synergy Frames. Figure 4 shows the network configuration in the context of the HPE Synergy Compute Modules utilized for management functions. In this part of the solution, the compute modules are virtualized and as such, ethernet traffic traverses a virtual switch prior to the Synergy LAG groups.

Figure 4. Network configuration for virtualized compute modules

Storage This section describes the configuration of the HPE D3940 Storage Module for:

• OpenShift Container Storage

• Virtual Machines including storage for Red Hat Virtualization Manager

• OpenShift Container Registry

• Persistent Volume Claims

Reference Architecture Page 11

Figure 5 below describes the logical storage layout used in the solution. Local internal storage is used for the Operating System installation on the Red Hat Virtualization Hosts. Storage required for the virtual machine storage domains that host the virtual machine disks is presented to each RHVH host as an eight (8) member RAID6 array from the D3940 Storage Module, 24 disks total 3 RAID6 arrays. This solution requires an additional six (6) disks, two (2) each that are presented to the RHVH hosts. These disks are subsequently passed through to the infrastructure and application virtual machines as raw devices using the Red Hat Virtualization Host Device Passthrough functionality. The six (6) SSD drives will be used for OpenShift Container Storage on the infrastructure and application virtual machines.

Figure 5. Logical storage layout within the solution

Table 5 lists all volumes used within the solution and highlights what storage provides the capacity.

Table 5. Volumes used in this solution

Volume/Disk Function Qty Size Source Hosts Shared/Dedicated

Operating System 3 300GB Internal drive bays 1 RAID1 per RHVH host Dedicated

Virtual Machine Hosting / RHVM – Gluster 3 1.8TB D3940 1 RAID6 (8 disks) per RHVH host Replicated

Persistent Application Data – OCS 3 400GB D3940 1 RAID0 per RHVH host / 1 per OpenShift worker node

Replicated

OpenShift Container Registry – OCS 3 400GB D3940 1 RAID0 per RHVH host / 1 per OpenShift Infrastructure node

Replicated

The array configurations used in the creation of this solution were built to suit configuration testing. Customer requirements around application performance and capacity should be considered when selecting disks and configuring arrays for Red Hat OpenShift Container Platform environments. In some cases, a single disk for OpenShift Container Storage may not be enough to meet performance, capacity, or available requirements of a production environment. Additional disks can be added to meet the demands of a production environment. Each D3940 Storage Module can hold up to 40 small form factor disk drives. When configured as documented in this reference configuration, the solution requires 30 small form factor disk drives to be installed in the D3940 Storage Module. This leaves room for an additional ten (10) disk drives to be added for enhanced scalability. Additional D3940 Storage Modules can be added to the Synergy 12000 Frame which would allow for scaling the solution beyond 40 disk drives provided by a single D3940 Storage Module.

Reference Architecture Page 12

Server profiles HPE Synergy Composable Infrastructure using HPE Virtual Connect provides the construct of a Server Profile. A Server Profile allows a suite of configuration parameters, including network and SAN connectivity, BIOS tuning, boot order configuration, local storage configuration and more to be templatized and applied programmatically to compute resources. These templates are the key to delivering the “infrastructure as code” capabilities of the HPE Synergy platform. For the purpose of this solution, a single template was created that was applied to all compute modules.

The critical items configured as part of the template were the connections and storage. Figure 6 describes the configuration of the network interfaces as part of the Server Profile Template. There are three (3) redundant networks, Management, Public, and Storage, and one (1) non-redundant network (PXE) that are defined.

Figure 6. Network interfaces as defined by the Server Profile Template

Reference Architecture Page 13

The storage arrays are defined and initialized in the server profiles as shown in Figure 7. The storage connections in the server profile define the volumes that will be created on the host from physical disks presented from the internal storage drive bays and the D3940 Storage Module.

Each of the RHV / RHHI hosts have the following volumes:

• Bootvol: RAID1 volume provide by the two internal drives in the SY480 Compute Module

• Gluster: RAID6 volume comprised of eight physical disks presented from the D3940 Storage Module

• OCS-1: RAID0 volume single SSD disk presented from the D3940 Storage Module used for Gluster based OpenShift Container Storage

• OCS-2: RAID0 volume single SSD disk presented from the D3940 Storage Module used for Gluster based OpenShift Container Storage

Figure 7. Storage configuration as defined by the Server Profile Template

Reference Architecture Page 14

Red Hat OpenShift virtual machines Red Hat OpenShift Container Platform requires a number of redundant functions. These functions may be hosted on either physical compute modules or on virtual machines, both of which run Red Hat Enterprise Linux 7.5. For this solution, HPE chose to implement these functions as virtual machines. This approach reduces the amount of infrastructure required while introducing enhanced options for management and high availability. Three (3) HPE Synergy 480 Gen10 Compute Modules host the virtual machines shown in Figure 8 below. This figure also shows that the worker nodes run on bare metal.

Figure 8. Red Hat OpenShift implementation with virtual machines for core management functions

The virtual environment is managed by Red Hat Virtualization Manager which runs as a virtual machine within the environment.

Reference Architecture Page 15

Software This section describes the software versions utilized in the solution as well as noting any special installation or configuration requirements. For detailed descriptions of how to install and configure the software, consult the deployment guide and accompanying scripts found at https://github.com/HewlettPackard/hpe-solutions-openshift/tree/master/synergy/entry-level-config/rhhi. Table 6 lists specific versions of required software.

Table 6. Versions used in the solution

Component Version*

Red Hat Virtualization Host 4.2

Red Hat Virtualization, Red Hat Virtualization Manager 4.2

Red Hat OpenShift Container Platform 3.10

Red Hat OpenShift Container Storage 3.10

* Latest sub-version should be installed

Capacity and sizing Sizing for a Red Hat OpenShift Container Platform environment varies depending upon the requirements of the specific organization and type of deployment. In this section we will discuss sizing considerations for Red Hat OpenShift Container Platform virtual machines, host requirements and cluster sizing.

Red Hat OpenShift Container Platform role sizing • Master – The minimum size for a physical or virtual machine running the master node is 4 vCPU and 16 GB RAM with a 40 GB disk space for

/var, 1 GB disk space for /usr/local/bin, and 1 GB disk space for the system’s temporary directory. Master nodes should be configured with an additional 1 CPU core and 1.5 GB RAM for each additional 1,000 pods Three (3) master node virtual machines were deployed for this solution.

• Infrastructure nodes – Application and Infrastructure nodes require a minimum of 1 vCPU and 8 GB RAM with a disk with at least 15 GB of space for /var/, 1 GB disk space for /usr/local/bin, and 1 GB disk space for the system's temporary directory. A total of three (3) infrastructure nodes were deployed for this solution.

• etcd – etcd nodes should be configured with a minimum of 4 vCPU and 15 GB RAM and 20 GB for etcd data. A total of three (3) etcd nodes were deployed for this solution.

• HAProxy – A single HAProxy load balancer VM was deployed for this solution. Default values for CPU and memory were utilized.

Red Hat OpenShift Container Platform cluster sizing The number of application nodes in an OpenShift cluster depends on the number of pods that an organization is planning on deploying. Red Hat OpenShift Container Platform can support the following maximums:

• Maximum of 2,000 nodes per cluster

• Maximum of 120,000 pods per cluster

• Maximum of 250 pods per node

• Maximum of 10 pods per CPU core

To determine the number of nodes required in a cluster, estimate the number of pods the organization is planning on deploying and divide by the maximum number of pods per node. For example, if the organization expects to deploy 5,000 pods, then the organization should expect to deploy 20 application nodes with 250 pods per node (5000 / 250 = 20). In this environment with a default configuration of three physical application nodes, the Red Hat OpenShift cluster should be expected to support 750 pods (250 pods x 3 nodes = 750 pods).

Reference Architecture Page 16

For more information about Red Hat OpenShift Container Platform sizing, refer to the Red Hat OpenShift Container Platform product documentation at:

https://access.redhat.com/documentation/en-us/openshift_container_platform/3.10/html/installing_clusters/install-planning#sizing

https://access.redhat.com/documentation/en-us/openshift_container_platform/3.10/html-single/scaling_and_performance_guide/#scaling-performance-cluster-limits

Summary Red Hat OpenShift Container Platform on HPE Synergy and Red Hat Hyperconverged Infrastructure provides an entry level end-to-end fully integrated container solution that, once assembled, can be configured within hours. This solution is ideal for proof of concept, departmental, and small business Red Hat OpenShift Container Platform deployments. This eliminates the complexities associated with implementing a container platform across an enterprise data center and provides the automation of hardware and software configuration to quickly provision and deploy a containerized environment at scale. Red Hat OpenShift Container Platform provides organizations with a reliable platform for deploying and scaling container-based applications and HPE Synergy provides the flexible infrastructure you need to run that container platform to dynamically provision and scale applications, whether they run in VMs or containers, or are hosted on-premises, in the cloud, or somewhere in between.

This Reference Configuration demonstrated the following benefits of utilizing HPE Synergy for Red Hat OpenShift Container Platform:

• Automated initial installation of a highly available Red Hat OpenShift Container Platform on HPE Synergy reduced the initial solution deployment from several days as provided by a services organization to less than a day.

• Deployed the OpenShift master, infrastructure, and application nodes on virtual machines running on Red Hat Hyperconverged Infrastructure to optimize resource usage in a single Synergy Frame.

• The ability to provide local scalable storage from the Synergy D3940 Storage Module to the SY480 Gen10 Compute Modules used with Red Hat Hyperconverged Infrastructure and Red Hat OpenShift Container Storage.

Implementing a proof-of-concept As a matter of best practice for all deployments, HPE recommends implementing a proof-of-concept using a test environment that matches as closely as possible the planned production environment. In this way, appropriate performance and scalability characterizations can be obtained. For help with a proof-of-concept, contact an HPE Services representative (hpe.com/us/en/services/consulting.html) or your HPE partner.

Reference Architecture Page 17

Appendix A: Bill of materials The following bill of materials contains the core components utilized in the creation of this solution. Services, support and software are not included in the BOM and should be customized based on customer needs.

Note Part numbers are accurate at time of testing and are subject to change. The bill of materials does not include complete support options or other rack and power requirements. If you have questions regarding ordering, please consult with your HPE Reseller or HPE Sales Representative for more details. hpe.com/us/en/services/consulting.html

Table A1. Bill of Materials

Rack components Qty Part Number

HPE 42U 600 mm x 1200 mm G2 kitted Advanced shock rack 1 P9K10A

HPE G2 Basic Modular 3-phase PDU, high voltage (208V), 24 A 4 P9Q52A

Network components

HPE FlexFabric 5940 Switch Components Qty Part Number

HPE FF 5940 2-Slot Switch 2 JH397A

HPE 5930 24p SFP+ and 2p QSFP+ Module 2 JH180A

HPE 5930 8-port QSFP+ Module 2 JH183A

HPE X712 Back (Power Side) to Front (Port Side) Airflow High Volume Fan Tray 4 JG553A

HPE 58x0AF 650W AC Power Supply 4 JC680A

INCLUDED: Jumper Cable - NA/JP/TW 4 JC680A B2B

HPE X240 40G QSFP+ QSFP+ 1m DAC Cable 2 JG326A

HPE X240 40G QSFP+ QSFP+ 3m DAC Cable 2 JG327A

Synergy Frame components

HPE Synergy 12000 Frame components Qty Part Number

HPE Synergy12000 CTO Frame 1xFLM 10x Fan 1 797740-B21

HPE Synergy 12000F 6x 2650W AC Ti FIO PS 1 798096-B21

HPE Synergy Composer 2 804353-B21

HPE Synergy Frame Link Module 1 804942-B21

HPE Synergy 12000 Frame Rack Rail Option 1 804938-B21

HPE Synergy 12000 Frame 4x Lift Handle 1 804943-B21

Synergy Compute components

HPE Synergy 480 Gen10 Compute Module components Qty Part Number

HPE SY 480 Gen10 CTO Premium Cmpt Mdl 3 871942-B21

HPE Synergy 480 Gen10 6142 Kit 3 872138-B21

HPE Synergy 480 Gen10 6142 Kit 3 872138-L21

HPE 32GB (1 x 32GB) dual rank x4 DDR4-2666 CAS-19-19-19 registered memory kit 24 815100-B21

HPE Smart Array P416ie-m SR Gen10 (8 Int 8 Ext Lanes/2GB Cache) 12G SAS Mezzanine Controller 3 804428-B21

Reference Architecture Page 18

Rack components Qty Part Number

HPE Smart Array P416ie-m SAS Cable Kit 3 871573-B21

HPE Synergy 3820C 10/20Gb CNA 3 777430-B21

HPE 96W Smart Stor Battery 260mm Cbl Kit 3 875242-B21

HPE 300GB SAS 10K SFF 6 872475-B21

Synergy Fabric componets

HPE Synergy Fabric components Qty Part Number

HPE VC SE 40Gb F8 Module 2 794502-B23

HPE Synergy 12Gb SAS Connection Module 2 755985-B21

Synergy Storage componets

HPE Synergy Composable Storage components Qty Part Number

HPE Synergy D3940 CTO Storage Module 1 835386-B21

HPE Synergy D3940 IO Adapter 1 757323-B21

HPE 300GB SAS 10K SFF 24 872475-B21

HPE 400GB SSD Mixed Use 6 872374-B21

Cable and Transciever components

Cables and Transceivers Qty Part Number

HPE Synergy 40GbE/4x10GbE/4x8Gb FC QSFP+ Transceiver 4 817040-B21

HPE Multi Fiber Push On to 4 x Lucent Connector 5m Cable 4 K2Q46A

HPE CAT6A 4ft Cbl 1 861412-B21

Reference Architecture

Sign up for updates

© Copyright 2018-19 Hewlett Packard Enterprise Development LP. The information contained herein is subject to change without notice. The only warranties for Hewlett Packard Enterprise products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. Hewlett Packard Enterprise shall not be liable for technical or editorial errors or omissions contained herein.

Red Hat is a registered trademark of Red Hat, Inc. in the United States and other countries. Linux is the registered trademark of Linus Torvalds in the U.S. and other countries. Intel and Xeon are trademarks of Intel Corporation in the U.S. and other countries. Microsoft and Windows are either registered trademarks or trademarks of Microsoft Corporation in the United States and/or other countries.

a00060153enw, January 2019

Resources and additional links HPE Synergy, hpe.com/info/synergy

HPE FlexFabric 5940 switching, hpe.com/us/en/product-catalog/networking/networking-switches/pip.hpe-flexfabric-5940-switch-series.1009148840.html

HPE OpenShift Solutions on GitHub, https://github.com/hewlettpackard/hpe-solutions-openshift

HPE Reference Architectures, hpe.com/info/ra

HPE Servers, hpe.com/servers

HPE Networking, hpe.com/networking

HPE Technology Consulting Services, hpe.com/us/en/services/consulting.html

Red Hat, redhat.com

Red Hat OpenShift Container Platform 3.10 Documentation, https://docs.openshift.com/container-platform/3.10/welcome/index.html

Red Hat OpenShift Container Platform, openshift.com

Red Hat OpenShift Container Storage, redhat.com/en/technologies/cloud-computing/openshift-container-storage

Red Hat Virtualization, redhat.com/en/technologies/virtualization/enterprise-virtualization

Red Hat Hyperconverged Infrastructure for Virtualization, redhat.com/en/resources/hyperconverged-infrastructure-for-virtualization-datasheet

To help us improve our documents, please provide feedback at hpe.com/contact/feedback.