Managing Cloud Hardware - Inspur Systems

Transcription

Managing Cloud HardwareAlfie Lew, InspurJohn Leung, IntelSai Dasari, Facebook

Agenda Hardware Management Overview Leverage Existing Standards Open Source Projects OpenBMC firmware Open Source Firmware OpenRMC Summary Q&AOpen SoftwareOpen HardwareOpen Firmware2

Hardware Management (Legacy) In-band Uses main CPU, primarymemory, etc. Off-platform Manageability of thesystem to a remote client IPMI over RMCP Issues: BMC firmware is not open Security concerns with RMCP RemoteClientOut-of-band Uses controller (BMC) May operate at low powerstatesOn-platform Manageability of deviceswithin the platform IPMI over variousprotocols and UNICNC-SIBMC FWBMCIPMIdeviceOn-platformotherdevicePlatform

DMTF Manageability StandardsRemoteClient Off-platform (Redfish) A RESTful interface for manageability (HTTP, JSON)In-band Includes resource models for systems, chassis,managers. Other SDO's have extended Redfish tomanage other type of platforms and services On-platform (PLDM over MCTP) MCTP - A route-able transport protocol for manageabilitymessages (connection independent)Out-of-band(Redfish)BIOSCPUNICNC-SIBMC FW PLDM - A bit-based request/response command model RDE - A PLDM model designed for devices to supportRedfish form

Redfish Overview Uses cloud/web protocols, structures, security models HTTP, JSON (leverage existing tool chains) Resource schemas are accessible, to enable interface programmatically json-schema, OpenAPI/swagger schemaHTTP/SHTTP GET https:// ip addr /redfish/v1/Systems/CS 1PythoncoderawData urllib.urlopen(‘https:// ip addr /redfish/v1/Systems/CS 1’jsonData json.loads(rawData)print( jsonData[‘SerialNumber’] )Output1A87CA442K

Redfish Computer System ModelService Root/redfish/v1RootTasksCollection Resource Single Resource/redfish/v1/SystemsCollection of Systems"Logical view"Sessions./Systems/ id ProcessorsComputerSystemMemoryComputer SystemsAccounts/redfish/v1/ChassisEventsCollection of Chassis"Physical view"RegistriesHTTP GET /redfish/v1/Systems/CS 1/Processors/2DisksNICs./Chassis/ id ChassisPowerThermalSchemasPlatformHW MgmtManaged By/redfish/v1/ManagersCollection ofManagers"Mgmt hierarchy"6Compute./Managers/ id BMCLogServiceNW protocol

Managing Beyond Computer SystemsThe Redfish Forum works with other SDOs tocreate models for new management domains Storage services and networked storageRedfish ClientRedfish Off-platform interfaceAug 2015SystemAug 2016Storage Ethernet Switch - map YANG to RedfishInternet Draft 2017NetworkYANG Industrial IoT Telco - Customer Premise EquipmentIIoT DCIM - DC Infrastructure Mgmt (e.g. CRACs,PDUs, chillers)(Starfish)DCIM (facilities)With the number of models increasing, how would OCPspecify the required manageability of OCP platforms?Telco

OCP Platform Manageability Direction OCP will use Redfish profiles to specify platform manageability The OCP Hardware Management projectInterface Will specify the manageability common across OCPplatforms Has approved the "OCP Baseline HardwareManagement rofile(s)TelcoProfile(s)Rack &PowerProfile(s) Other OCP projects Will create platform specific profilesHardware Mgmt Baseline Profile Profiles will extend the "Baseline HardwareManagement Profile" "OCP Server Profile" is available1opencompute.org/wiki/Hardware Management/SpecsAndDesigns

OCP San Jose Compute Node Successfully ran the Redfish Interop Validator testing against the "OCP BaselineHardware Management Profile" Demonstrated at OCP Summit (March 2018)OCP Server Profile

Enabling Interoperable Manageability OCP will prescribe platform manageability in machine readable JSON profile Implementers can use the profile to test conformance of an implementation Linux Foundation governs the open-source implementation of BMC firmware (OpenBMC)10

OCP San Jose Compute Node Incorporated OpenBMC The implementation passed the OCP Baseline HW Manageability Profile Demo’ed at OCP Regional Summit (Oct 2018, Amsterdam) Changes will be upstreamed to the OpenBMC repository1 Target Date: March 2019 available on GitHub1The Linux Foundation project

A Brief History of OpenBMC ProjectTwo independent implementationsFacebookIBMLinux Foundation Project(IBM, Google, Intel, Microsoft, Facebook)12

OpenBMC Development ModelTraditionalOpenBMC3rd PartyHW Vendor3rd Party BMCFW Vendor3rd PartyHW Vendor3rd Party BMCHW speed/Nuvoton)3rd Party BMCFW VendorBMC FWSource Code3rd Party BMCFW VendorOpenBMCProjectODMServer VendorBMC FWSource Code13

OpenBMC WAREUPDATERMCP DGPIO/ADCBootloader& IPMIIPMBEWSENDPOINTLinux Kernel/DriversBoot Loader (e.g. U-Boot)BMC SoC / SPI Flash / TPM14

Additional Resources Source Code: https://github.com/openbmc Continuous Integration Testing https://openpower.xyz Code Reviews http://gerrit.openbmc-project.xyz Contact Mail: openbmc@lists.ozlabs.org IRC: #openbmc on freenode.net Riot: #openbmc:matrix.org Web Page: www.openbmc.org15

Mission: Develop an open source philosophy based ‘system firmware’(BIOS) modules, to support different Operating Systems and differentCPU silicon r Companies Contributing to OSF development: Microsoft, Intel,Google, Facebook, Lenovo, IBM, Two Sigma, ITRenew, 9 Elements,Cavium, AMD and many more.Major Work Streams: Open EDKII DXE core, Linux Boot, Core Boot,Silicon Interface Firmware Module, Intel FSP, AMD AGESA, ARM bootcode, HW platform module, Build tools, Automated test support, HWrequirements, etc.GitHub Repositories Collateral eekly OSF discussions : Architectural reviews, workstream progress,design reviews, agenda setting and other collaborative discussions onOSF development. Contact the following to subscribe, participate e Community development model

Proposed OpenRMC sub-project Rack The rack is the fundamental building block for the Data Center Rack Manager Controller (RMC) The RMC is fundamental to Data Center Management hierarchy The RMC can be located standalone, within power shelf, or within the switch Opportunity Common language/model between RMC and nodes (aggregation, etc.) Provide an open source reference RMCnodes

Open Source Firmware for OCP EcosystemOCP DesignCollateralPresentationsBinary BlobsProduct/Company SpecificSource RepoSpecificationsFirmware(Build Meta Data)SchematicsDesignPackagesData Sheets3D/CAD FilesOpen Source Community Repo(e.g. OpenBMC, OpenRMC, LinuxBoot,Coreboot, OpenEDKII)OCP Specific Source Repo(e.g. LED Config, RedfishProfiles)18

Hardware Management (Legacy) In-band Uses main CPU, primary memory, etc. Off-platform Manageability of the system to a remote client IPMI over RMCP Platform Out-of-band (IPMI) BMC OS NIC NC-SI BIOS CPU BMC FW SW Agent In-band device device IPMI Remote Client Issues: BMC firmware is not open Security concerns with RMCP .