50290 Understanding and Administering Windows HPC Server 2008

Code: 50290
Course duration: 3 days
Price:

50290: Understanding and Administering Windows HPC Server 2008 (3 Days)

About this Course

This three-day instructor-led course provides students with the knowledge and skills to manage and deploy Microsoft HPC Server 2008 clusters.

Audience Profile

This three-day instructor-led course provides students with the knowledge and skills to setup and administer high-performance clusters running Microsoft Windows HPC Server 2008.

At Course Completion

After completing this course, students will be able to:

  • Plan external services for the cluster.
  • Create and modify node templates.
  • Configure the head node.
  • Configure networks.
  • Prepare nodes for deployment.
  • Deploy nodes.
  • Diagnose provisioning issues.
  • Validate cluster state by using cluster diagnostic tools.
  • Manage nodes.
  • Manage users and groups.
  • Manage node state.
  • Monitor nodes.
  • Submit a job.
  • Monitor a job.
  • Manage resources.
  • Create a job template.

Course Outline

Module 1: Introduction to High-Performance Computing and HPC Server

This module introduces the field of high-performance computing, and the Microsoft product Windows HPC Server 2008.

Lessons

  • Motivation for HPC.
  • Brief product history of Compute Cluster Server and Windows HPC Server 2008.
  • Brief overview of Windows HPC Server 2008 — components, job submission, scheduler.
  • HPC resources: cores vs. sockets vs. nodes.
  • Product differentiators.
  • The goal of your developers — linear speedup.

Lab : Introduction to HPC And Windows HPC Server 2008

  • Submitting jobs.
  • Monitoring job execution.
  • Measuring performance.

After completing this module, students will be able to:

  • Submit jobs to the cluster.
  • Monitor job execution on the cluster.
  • See performance improvements in HPC applications.

Module 2: The Windows HPC Server Job Scheduler in depth

This module discusses the heart of Windows HPC Server 2008 — the Job Scheduler.

Lessons

  • Throughput vs. performance.
  • Jobs vs. Tasks.
  • Job and task states.
  • The definition of a failed job / task, auto re-starting upon failure.
  • Default scheduling policies.
  • Job-level vs. task-level policies.
  • The impact of job priorities and job preemption.
  • Job resources and dynamic growing / shrinking.

Lab : The Job Scheduler

  • Investigating job parameters.
  • Investigating task parameters.
  • Observing the job scheduler in action.

After completing this module, students will be able to:

  • Troubleshoot the job scheduler.
  • Understand the differences between a task and a job.
  • Change default scheduling policies.
  • Understand job scheduling policies and how to change them.
  • Change the default allocation to jobs and understand best scenarios for each one.

Module 3: The main components of a Windows HPC Server cluster

This module overviews the main hardware and software components of an HPCS-based cluster.

Lessons

  • Headnode vs. compute nodes vs. broker nodes.
  • Windows Server 2008 + HPC Pack.
  • SQL Server.
  • Active directory and credentials management.
  • Network topologies.
  • Data storage options: SMB, NAS, PFS.
  • Job Scheduler and the job queue.

Lab : Exploring Windows HPC Server 2008

  • Hardware components.
  • Software components.

After completing this module, students will be able to:

  • Understand the difference between the different node types.
  • How to change the roles of the nodes from a cluster.
  • When to use SQL Server Express vs. SQL Server professional.
  • Understand the scenarios for each network topology.
  • Understand how the job scheduler works and how only having one queue simplifies management.

Module 4: Interfacing with HPC Server — from command prompts to MMC plugins

This module demonstrates the various ways you can interface with Windows HPC Server 2008.

Lessons

  • Cluster Manager.
  • Job Manager.
  • Job Description Files.
  • Clusrun.
  • Console window.
  • PowerShell.
  • Scripts.
  • Programmatically via HPCS API.

Lab : Interfacing with HPC Server

  • Using the MMC plugins.
  • Clusrun is your friend.
  • Using the command line.
  • Using PowerShell.
  • Scripting.

After completing this module, students will be able to:

  • How to use clusrun to execute tasks on all nodes.
  • How to use the command line to control the cluster.
  • How to use PowerShell to carry out management tasks.

Module 5: Basic cluster setup, from hardware to software

This module overviews the standard setup needs and procedures for a Windows HPC Server based cluster.

Lessons

  • Hardware, physical and virtual.
  • Software: Windows Server 2008 editions, Windows HPC Server Pack, Windows HPC Server Pack SDK.
  • Active Directory integration.
  • Common groups, local directories, and network shares.
  • Network topologies, DNS, and DHCP.
  • Runtimes, software, and tools commonly needed by developers.
  • Supporting remote debugging and tracing.

Lab : Installing Windows HPC Server 2008

  • Basic install of Windows HPC Server 2008 on a small, virtual cluster.

After completing this module, students will be able to:

  • Install an HPC Compute Cluster from scratch.
  • Understand the implications of network topologies when running SOA applications.
  • Use tools to simply the deployment process.

Module 6: Advanced setup and integration options

This module dives deeper into the setup of more advanced Windows HPC Server based clusters, along with use of the more advanced setup features available in Windows HPC Server 2008.

Lessons

  • Windows Deployment Services.
  • High-speed networking hardware and drivers, e.g. InfiniBand.
  • Configuring headnode failover with Windows Server Failover Clustering.
  • Enabling support for Open Grid Forum’s basic web profile.
  • Additional monitoring via System Center Operations Manager.
  • Cluster integration (with other Windows HPC Server clusters, and Linux clusters).

After completing this module, students will be able to:

  • Understand how WDS works with Windows HPC Server.
  • Configure a failover cluster with Windows HPC Server.
  • Use SCOM to monitor cluster health.

Module 7: Understanding how developers will use the cluster

This module discusses the goals of HPC developers, and how you can support them in their efforts.

Lessons

  • What applications can run on the cluster.
  • What software technologies will developers typically use.
  • Sequential apps.
  • Parametric sweep.
  • SOA applications.
  • Multi-threaded apps.
  • MPI apps.
  • UNIX apps.
  • The cluster heatmap and how to customize.

Lab : Understanding and Helping Your Developers

  • A parametric sweep application.
  • An SOA application.
  • An MPI application.
  • Customizing the heatmap display in Windows HPC Server 2008.

After completing this module, students will be able to:

  • Understand the types of jobs developers will usually submit to the cluster.
  • Use node resources appropriately for different types of tasks.

Module 8: Configuring Windows HPC Server 2008

This module discusses the most commonly-used configuration options in Windows HPC Server based clusters, and why.

Lessons

  • Node groups.
  • Job templates.
  • Job preemption.
  • Dynamic resource allocation.
  • Submission and activation filters.
  • Job history, job restarting.
  • Applying configurations to HPC users and groups.

Lab : Configuring Windows HPC Server 2008

  • Configuring job preemption.
  • Writing and installing submission and activation filters.

After completing this module, students will be able to:

  • Create node groups for efficient cluster utilization.
  • Understand how to create job templates and grant access to your cluster users.
  • Resubmit failed jobs to the cluster without having to create new ones from scratch.

Module 9: Maintenance and Performance Tuning of your Cluster

This module presents the tools and techniques available for tracking down both cluster and application failures.

Lessons

  • Cluster Manager diagnostics.
  • Defining your own, custom diagnostic tests.
  • Common job and task failures — from the command-line to licensing.
  • App-specific failures with MPI and SOA.
  • Security issues.
  • Remote desktop as a debugging technique.
  • Windows event log.
  • WCF tracing.

After completing this module, students will be able to:

  • Use diagnostics to find common causes for cluster issues.
  • Understand the most common scenarios related to MPI and SOA job submission.
  • Use tracing to find problems associated with WCF.

Module 10: Cluster Maintenance

This module overviews the typical maintenance requirements for a Windows HPC Server based cluster.

Lessons

  • Cluster Manager charts and reports.
  • Performance maintenance: MPI ping-pong, Lizard, and uSane.
  • Synchronizing software patches to the cluster.
  • Headnode maintenance.
  • SQL Server maintenance.
  • Compute and broker node maintenance.
  • Automating maintenance runs.

After completing this module, students will be able to:

  • User the charts and reports feature to better understand cluster utilization
  • Find performance-related issues by using MPI ping-pong and uSane tools.
  • Efficiently maintain the cluster running by configuring the head node.

Before attending this course, students must have:

  • Experience with Windows Server 2008.

Course Reviews

No reviews found for this course.

Be the first to write a review