compute-nest-best-practice/ehpc-demo/template.yaml

ROSTemplateFormatVersion: '2015-09-01' Description: zh-cn: 该模版用于创建一个弹性高性能计算（EHPCC）集群，包含管理、登录和计算节点，支持Tiny、Simple、Standard三种部署模式，配置实例类型、数量、存储、网络及付费方式，同时自动化设置文件系统与安全组，确保集群运行在指定VPC环境，且提供Web访问入口。 en: 'This template is designed for establishing an Elastic High-Performance Computing Cluster (EHPCC), encompassing management, login, and compute nodes. It accommodates three deployment modes: Tiny, Simple, and Standard, allowing customization of instance types, quantities, storage configurations, networking setups, and billing methods. The process automates the configuration of file systems and security groups, ensuring the cluster operates within a specified Virtual Private Cloud (VPC) environment. Additionally, it provisions a web access portal for convenient cluster management.' Conditions: TinyMode: Fn::Equals: - Ref: EhpcDeployMode - Tiny Parameters: EhpcClusterName: Type: String Label: zh-cn: 集群名称 en: Cluster Name Description: zh-cn: 创建集群名称，长度2-64字符. en: Cluster Name. 2-64 characters in length. InstancePassword: Type: String Label: zh-cn: 集群登录密码 en: Cluster Login Password Description: zh-cn: 长度8-30，必须包含三项（大写字母、小写字母、数字、 ()`~!@#$%^&*-+=|{}[]:;'<>,.?/ 中的特殊符号）。 en: Length 8-30, must contain three(Capital letters, lowercase letters, numbers, ()`~!@#$%^&*-+=|{}[]:;'<>,.?/ Special symbol in). ConstraintDescription: zh-cn: 长度8-30，必须包含三项（大写字母、小写字母、数字、 ()`~!@#$%^&*-+=|{}[]:;'<>,.?/ 中的特殊符号）。 en: Length 8-30, must contain three(Capital letters, lowercase letters, numbers, ()`~!@#$%^&*-+=|{}[]:;'<>,.?/ Special symbol in). AssociationProperty: ALIYUN::ECS::Instance::Password AllowedPattern: '^[a-zA-Z0-9-\`\~\!\@\#\$\%\^\&\*\_\-\+\=\|\{\}\[\]\:\;\<\>\,\.\?\/]*$' MinLength: 8 MaxLength: 30 NoEcho: true EhpcDeployMode: Type: String Label: zh-cn: Ehpc部署模式 en: ehpc deploy mode Default: Tiny AllowedValues: - Tiny - Simple - Standard EhpcComputeInstanceType: Type: String Label: zh-cn: 计算节点实例类型 en: Login Instance Type Description: zh-cn: 填写可用区下可使用的规格； 规格详见：<a href='https://help.aliyun.com/document_detail/25378.html' target='_blank'>实例规格族</a> en: 'Fill in specifications that can be used under the availability zone; see detail: <a href=''https://www.alibabacloud.com/help/en/doc-detail/25378.html'' target=''_blank''>Instance Specification Family</a>' AssociationProperty: ALIYUN::ECS::Instance::InstanceType AssociationPropertyMetadata: InstanceChargeType: ${InstanceChargeType} EphcComputeCount: Type: Number Label: zh-cn: 计算节点数量 en: Computing node number Description: zh-cn: '计算节点数量, 可选值: 1-99.' en: 'Computing node number, which ranges from: 1-99.' Default: 1 MinValue: 1 MaxValue: 99 EhpcLoginInstanceType: Type: String Label: zh-cn: 登录节点实例类型 en: Login Instance Type Description: zh-cn: 填写可用区下可使用的规格； 规格详见：<a href='https://help.aliyun.com/document_detail/25378.html' target='_blank'>实例规格族</a> en: 'Fill in specifications that can be used under the availability zone; see detail: <a href=''https://www.alibabacloud.com/help/en/doc-detail/25378.html'' target=''_blank''>Instance Specification Family</a>' AssociationProperty: ALIYUN::ECS::Instance::InstanceType AssociationPropertyMetadata: Visible: Condition: Fn::Not: Fn::Equals: - ${EhpcDeployMode} - Tiny Default: ecs.c6.large EhpcLoginCount: Type: Number Label: zh-cn: 登录节点数量 en: Login Node Number Description: zh-cn: '登录节点数量, 可选值: 1-8.' en: 'Log node number, which ranges from: 1-8.' AssociationPropertyMetadata: Visible: Condition: Fn::Not: Fn::Equals: - ${EhpcDeployMode} - Tiny Default: 1 MinValue: 1 MaxValue: 8 EhpcManagerInstanceType: Type: String Label: zh-cn: 管控节点实例类型 en: Login Instance Type Description: zh-cn: 填写可用区下可使用的规格； 规格详见：<a href='https://help.aliyun.com/document_detail/25378.html' target='_blank'>实例规格族</a> en: 'Fill in specifications that can be used under the availability zone; see detail: <a href=''https://www.alibabacloud.com/help/en/doc-detail/25378.html'' target=''_blank''>Instance Specification Family</a>' AssociationProperty: ALIYUN::ECS::Instance::InstanceType EhpcManagerCount: Type: Number Label: zh-cn: 管控节点数量 en: Manager Node Number Description: zh-cn: '管控节点数量, 可选值: 1,2,4.' en: 'Manager node number, which ranges from: 1,2,4.' Default: 1 AllowedValues: - 1 - 2 - 4 SystemDiskCategory: Type: String Label: zh-cn: 系统盘类型 en: SystemDisk Category Description: zh-cn: 系统盘类型. en: SystemDisk Category. AssociationProperty: ALIYUN::ECS::Disk::SystemDiskCategory AssociationPropertyMetadata: LocaleKey: DiskCategory InstanceType: EhpcComputeInstanceType SystemDiskSize: Type: Number Label: zh-cn: 系统盘大小 en: SystemDisk Size Description: zh-cn: 系统盘大小. en: SystemDisk Size. Default: 40 SystemDiskPerformanceLevel: Type: String Label: zh-cn: 系统盘性能等级. en: SystemDisk PerformanceLevel. Description: zh-cn: 系统盘性能等级. en: SystemDisk PerformanceLevel. AssociationPropertyMetadata: Visible: Condition: Fn::Equals: - ${SystemDiskCategory} - cloud_essd Default: PL1 AllowedValues: - PL0 - PL1 - PL2 - PL3 SchedulerType: Type: String Label: zh-cn: 调度器类型 en: Scheduler Type Description: zh-cn: 调度器类型，当前支持pbs和slurm en: The scheduler type, currently support pbs and slurm. Default: pbs AllowedValues: - pbs - slurm - opengridscheduler - deadline AccountType: Type: String Label: zh-cn: 域账户服务类型 en: Domain Service Account Type Description: zh-cn: 域账户服务类型，当前支持nis和ldap en: Domain service account types, currently supports nis and ldap. Default: nis AllowedValues: - nis - ldap ZoneId: Type: String Label: zh-cn: 可用区 en: Zone ID Description: zh-cn: 可用区配置决定了新建ECS实例的可用区。 en: ECS VSwitch Zone ID defines the Availability Zone for creating ECS instance. AssociationProperty: ALIYUN::ECS::Instance::ZoneId VpcId: Type: String Label: zh-cn: VPC ID en: VPC ID Description: zh-cn: 请选择已有的VPC 如找不到VPC,请切换地域。 en: Please select the existing VPC. If you can't find the VPC, please switch Regions. AssociationProperty: ALIYUN::ECS::VPC::VPCId VSwitchId: Type: String Label: zh-cn: 交换机ID en: VSwitch ID Description: zh-cn: 现有网络交换机的实例ID en: The instance ID of an existing network switch. AssociationProperty: ALIYUN::ECS::VSwitch::VSwitchId AssociationPropertyMetadata: VpcId: ${VpcId} ZoneId: ${ZoneId} PayType: Type: String Label: zh-cn: 付费类型 en: ECS Instance Charge Type AssociationProperty: ChargeType AssociationPropertyMetadata: LocaleKey: InstanceChargeType Default: PostPaid AllowedValues: - PostPaid - PrePaid PayPeriodUnit: Type: String Label: zh-cn: 购买资源时长周期 en: Pay Period Unit Description: zh-cn: 购买资源时长的周期。 Week为周，Month为月 默认值为月 当ECS实例类型为PrePaid有效 en: The long cycle of purchasing resources. Week is a Week and Month is a Month. The default value is month. When ECS instance types are PrePaid valid AssociationProperty: PayPeriodUnit AssociationPropertyMetadata: Visible: Condition: Fn::Not: Fn::Equals: - ${PayType} - PostPaid Default: Month AllowedValues: - Month - Year PayPeriod: Type: Number Label: zh-cn: 购买资源时长 en: Period Description: zh-cn: 当购买资源时长为Month时，Period取值：1~9 当ECS实例类型为PrePaid有效 en: When the resource purchase duration is Month, the value of Period ranges from 1 to 9, 12, 24, 36, 48, or 60. When ECS instance types are PrePaid valid AssociationProperty: PayPeriod AssociationPropertyMetadata: Visible: Condition: Fn::Not: Fn::Equals: - ${PayType} - PostPaid Default: 1 AllowedValues: - 1 - 2 - 3 - 4 - 5 - 6 - 7 - 8 - 9 Resources: # 创建权限组 AccessGroup: Type: ALIYUN::NAS::AccessGroup Properties: AccessGroupName: Fn::Join: - '-' - - Ref: ALIYUN::StackName - '[1,4]' AccessGroupType: Vpc # 创建权限组规则 AccessRule: DependsOn: AccessGroup Type: ALIYUN::NAS::AccessRule Properties: AccessGroupName: Fn::GetAtt: - AccessGroup - AccessGroupName # 优先级 Priority: 1 # 读写权限类型 RWAccessType: RDWR # 地址 SourceCidrIp: 0.0.0.0/0 # 用户权限类型 UserAccessType: no_squash # 创建文件系统 FileSystem: Type: ALIYUN::NAS::FileSystem Properties: # 协议类型 ProtocolType: NFS # 存储类型 StorageType: Capacity ZoneId: Ref: ZoneId VSwitchId: Ref: VSwitchId VpcId: Ref: VpcId # 创建挂载点 MountTarget: DependsOn: - FileSystem - AccessRule Type: ALIYUN::NAS::MountTarget Properties: # 权限组名称 AccessGroupName: Fn::GetAtt: - AccessGroup - AccessGroupName # 文件系统ID FileSystemId: Fn::GetAtt: - FileSystem - FileSystemId # 网络类型 NetworkType: Vpc VSwitchId: Ref: VSwitchId VpcId: Ref: VpcId # 安全组 EcsSecurityGroup: Type: ALIYUN::ECS::SecurityGroup Properties: # 使用默认名称 SecurityGroupName: Ref: ALIYUN::StackName VpcId: Ref: VpcId # 出方向全开 SecurityGroupEgress: - Priority: 1 PortRange: -1/-1 DestCidrIp: 0.0.0.0/0 NicType: intranet IpProtocol: all # 类型用于创建一个弹性高性能计算集群 Cluster: Type: ALIYUN::EHPC::Cluster Properties: ZoneId: Ref: ZoneId VSwitchId: Ref: VSwitchId SecurityGroupId: Ref: EcsSecurityGroup ImageId: centos_7_6_x64_20G_alibase_20211130.vhd OsTag: CentOS_7.6_64 EcsOrderComputeCount: Ref: EphcComputeCount VolumeType: 'nas' VolumeId: Fn::GetAtt: - FileSystem - FileSystemId EcsOrderManagerInstanceType: Ref: EhpcLoginInstanceType EcsOrderComputeInstanceType: Ref: EhpcComputeInstanceType PeriodUnit: Ref: PayPeriodUnit AutoRenew: 'true' Password: Ref: InstancePassword Name: Ref: EhpcClusterName SchedulerType: Ref: SchedulerType EcsChargeType: Ref: PayType EcsOrderLoginCount: Ref: EhpcLoginCount DeployMode: Ref: EhpcDeployMode Period: Ref: PayPeriod AccountType: Ref: AccountType VolumeMountpoint: Fn::GetAtt: - MountTarget - MountTargetDomain EcsOrderManagerCount: Ref: EhpcManagerCount EcsOrderLoginInstanceType: Ref: EhpcLoginInstanceType # 类型用于获取弹性高性能计算集群节点的列表 ClusterNodes: Type: DATASOURCE::EHPC::Nodes Properties: # 节点类型 Role: Fn::If: - TinyMode - Manager - Login # 待查询的集群ID。 ClusterId: Fn::GetAtt: - Cluster - ClusterId Outputs: VisitUrl: Value: Fn::Sub: - https://${Domain}:8081 - Domain: Fn::Jq: - First - .[].PublicIpAddress - Fn::GetAtt: - ClusterNodes - Nodes Software: Value: 1.0.0 Metadata: ALIYUN::ROS::Interface: ParameterGroups: - Parameters: - EhpcClusterName - InstancePassword Label: default: zh-cn: EHPC集群配置 en: Environment Configuration - Parameters: - EhpcDeployMode - EhpcComputeInstanceType - EphcComputeCount - EhpcLoginInstanceType - EhpcLoginCount - EhpcManagerInstanceType - EhpcManagerCount - SystemDiskCategory - SystemDiskSize - SystemDiskPerformanceLevel - SchedulerType - AccountType Label: default: zh-cn: EHPC资源配置 en: Resource Configuration - Parameters: - ZoneId - VpcId - VSwitchId Label: default: zh-cn: 网络配置 en: Network Configuration - Parameters: - PayType - PayPeriodUnit - PayPeriod Label: default: zh-cn: 付费类型配置 en: PayType Configuration

compute-nest-best-practice/ehpc-demo/template.yaml (410 lines of code) (raw):