Hadoop操作手冊（影印版）

Preface
1．Introduction
2．HDFS
Goals and Motivation
Design
Daemons
Reading and Writing Data
The Read Path
The Write Path
Managing Filesystem Metadata
Namenode High Availability
Namenode Federation
Access and Integration
Command—Line Tools
FUSE
REST Support
3．MapReduce
The Stages of MapReduce
Introducing Hadoop MapReduce
Daemons
When It All Goes Wrong
YARN
4．Planning a Hadoop Cluster
Picking a Distribution and Version of Hadoop
Apache Hadoop
Cloudera’S Distribution Including Apache Hadoop
What Should I Use?
Hardware Selection
Master Hardware Selection
Worker Hardware Selection
Cluster Sizing
Blades，SANs，and Virtualization
Operating System Selection and Preparation
Deployment Layout
Software
Hostnames．DNS．and Identmcation
Users，Groups，and Privileges
Kernel Tuning
vm．swappiness
vm．overcommit_memory
Disk Configuration
Choosing a Filesystem
Mount Options
Network Design
Network Usage in Hadoop：A Review
1 Gb versus 10 Gb Networks
Typical Network Topologies
5．Installation andConfiguration
Installing Hadoop
Apache Hadoop
CDH
Configuration：An 0verview
The Hadoop XML Configuration Files
Environment Variables and Shell Scripts
Logging Configuration
HDFS
Identification and Location
Optimization and Tuning
Formatting the Namenode
Creating a／tmp Directory
Namenode High Availability
Fencing Options
Basic Configuration
Automatic Failover Configuration
Format and Bootstrap the Namenodes
Namenode Federation
MapReduce
Identification and Location
Optimization and Tuning
Rack Topology
Security
6．Identity，Authentication，and Authorization
Identity
Kerberos and Hadoop
Kerberos：A Refresher
Kerberos Support in Hadoop
Authorization
HDFS
MapReduce
Other Tools and Systems
Tying It Together
7．ResojJrceManagement
What Is Resource Management?
HDFS Quotas
MapReduce Schedulers
The FIFO Scheduler
The Fair Scheduler
The Capacity Scheduler
The Future
8．ClusterMaintenance
Managing Hadoop Processes
Starting and Stopping Processes with Into Scripts
Starting and Stopping Processes Manually
HDFS Maintenance Tasks
Adding a Datanode
Decommissioning a Datanode
Checking Filesystem Integrity with fsck
Balancing HDFS Block Data
Dealing with a Failed Disk
MapReduce Maintenance Tasks
Adding a Tasktracker
Decommissioning a Tasktracker
Killing a MapReduce Job
Killing a MapReduce Task
Dealing with a Blacklisted Tasktracker
9．Troubleshooting
Differential Diagnosis Applied to Systems