White Papers

IntegratinghooksandtoolsforbettermanageabilityofHPCCluster



ByMuniraHussainandCalvinJacob



Managinglocalandremoteservernodesinaclusteratscalesoftenstothousandsofnodesisalwaysa

challenge.Inordertoreducethecluster‐managementoverheadandsimplifysetupofclusterofnodes,

adminsseektheconvenienceofaone‐snapshotglimpse.Rapidchangesintechnologymakeseaseof

management,tuning,customization,changingandupdatingsettingsanongoingrealityasinfrastructure

isrefactoredandrefreshed.

Tosimplifysomeofthesechallenges,itisimportantthathardwaremanagementbefullyintegratedinto

thesolutionforclustermanagement..TheintegrationbetweenBrightClusterManager7.1andDell

PowerEdgeHaswellbasedserversdepictaleadingexampleofwhatisachievabletoday.Criticaltothis

integrationisIntegratedDellRemoteAccessController(IDRAC).IDRACisembeddedintothe

motherboardsofDellserversforin‐bandandout‐of‐bandsystemmanagementandcandisplayand

modifyBIOSsettingsaswellasperformfirmwareupdatesthroughtheLifeCycleControllerandremote‐

console.Collectively,eachserver’sin‐depthsystemprofileinformationisgatheredusingsystemtools

andutilitiesandisavailableinasinglegraphicaluserinterfacefortheeaseofadministration,thus

reducingtheneedtophysicallyaccessserversdirectly.



PAGE 1
Integrating hooks and tools for better manageability of HPC Cluster By Munira Hussain and Calvin Jacob Managing local and remote server nodes in a cluster at scales of tens to thousands of nodes is always a challenge. In order to reduce the cluster‐management overhead and simplify setup of cluster of nodes, admins seek the convenience of a one‐snapshot glimpse.
PAGE 2
Figure 1. BIOS‐level integration between Dell PowerEdge servers and Bright Cluster Manager 7.1.  DELL API Integrated  RACADM/IPMI commands capture XML schemaand profile of the server System Management fabric/ IDRAC IP in two modes. Shared or Dedicated Shared (LOM1) Dedicated Figure 1 (above) depicts the configuration setup for a single node in the cluster. The fabric can be accessed via the dedicated iDRAC port or shared with the LAN‐on‐Motherboard capability.
PAGE 3
Figure 2. Snapshot of the Cluster Node Configuration via Bright Cluster Manager 7.1. Figure 2 is a screenshot from Bright Cluster Manager 7.1 that shows BIOS version and system profile information for a number of Dell PowerEdge C6320 servers. This is a particularly useful overview as inappropriate settings and versions can be easily and rapidly identified. An example usage of this would be when new servers are added or replaced in a cluster.
PAGE 4