Inside the HP-UX operating system

Prerequisites

• HP-UX system and network administration I (H3064S) or HP-UX for experienced UNIX system administrators (H5875S)
• Highly recommended: At least one year's experience in HP-UX administration

Audience

• Experienced HP-UX system administrators and programmers

Ways to save

Course Objectives

• Describe LVM architecture, structures and operations
• Describe the I/O subsystem and the HP-UX 11i v3 mass storage stack
• Outline the main steps in system initialization
• Describe the role of the HP-UX kernel and give an overview of kernel subsystems
• Explain how the PA-RISC or Integrity system architecture provides processing and memory resources
• Describe how kernel services are requested
• Describe how processes are managed
• Outline the issues with a multiprocessor system
• Describe how memory is managed
• Describe the methods used for interprocess communication
• Describe how files are managed and the use of the Unified File Cache

Benefits to you

• Gain the knowledge you need to understand, support and optimize your HP-UX system

Next Steps

• HP-UX performance and tuning (H4262S)
• HP-UX troubleshooting (H4264S)

Course outline

• Introduction
• Introduce the concepts and roles of the HP-UX kernel
• Introduce the major kernel data structures to be covered in the class
• Introduce the kernel debugging tools used to explore these structures
• System architecture
• Describe the architecture and their characteristics of the systems used to run HP-UX, particularly the concept behind cell based systems
• Briefly describe the characteristics of PA-RISC
• Describe the Itanium product line, with particular focus on the memory management and caching components
• Describe the interrupt handling mechanism of the processors
• Process Management
• Start by defining programs, processes and threads
• Describe the different threading models
• Examine the systems calls that make up the life cycle of processes, from a process management prospective
• Describe the context switching a state saving mechanisms
• Examine the system call gateway
• Look at the different virtual address space options available for processes running on HP-UX
• Describe the states and substates of processes and threads
• Explore the behavior of the scheduler
• Examine the major kernel functions involved in thread scheduling
• Describe the context switching state saving mechanisms
• Multiprocessor Systems
• Explore the need for multiprocessor environments, what they provide and what they do not
• Describe the major data structures involved in managing the MP environment
• Explain the different possible contention situations and the way in which the kernel avoids them: Locking, Spin locks, Semaphores
• Interprocess Communications
• Look at the implementation and data structures used for
• Signals
• System V IPC (semaphores, message queues and shared memory)
• POSIX IPC
• Filesystems
• Introduce the original Bell labs filesystem as a way of introducing filesystem concepts and problems
• Briefly describe the operation of the HFS filesystem
• Describe the major disk structures used by the Veritas filesystem
• Explore the kernel data structures used in file access and filesystem management
• Cover the upper layer, filesystem independent structures
• The vnode/vfs layer
• The Filesystem-dependent structures for both HFS and VxFS
• Look at the new unified file cache
• Memory management
• Describe the mechanisms to translate between virtual and physical addresses
• Examine the hardware independent management of the physical memory resources of the system including the free space
• Explore the kernel data structures involved in managing a processes address space
• Describe the operation of the process management system calls from a memory management perspective
• Examine the mechanisms used to bring data into memory
• Describe how the system responds to memory pressure
• Describe the data structures involved in swap space management
• Examine the management of the kernels memory space
• Volume management
• Describe the operation of the Logical Volume Manager
• Its overall architectures
• The disc based structures
• The kernel based datastructures
• The IO Subsystem
• Describe the addressing of IO devices
• Compare the modes of IO operation: direct IO, DMA
• Describe IO system call flow
• Describe the use of the device switch tables
• Examine the General IO environment and context dependent IO modules
• Explore the major IO related data structures
• Describe the changes in the 11i v3 mass storage stack
• System Initialization
• For both PA-RISC and Itanium servers, cover the different phases of system startup
• Firmware layers (PAL, SAL & EFI for IPF, PDC & BCH for PA-RISC)
• Operating systems loaders and secondary loaders
• Early kernel initialization
• Later initialization phases
• Building and initializing major data structures
• Starting the IO environment
• Mapping IO drivers and devices
• Kernel Services
• Describe how user programs access kernel functions
• Describe the callout service