Index of modules

Represents blocks from a single datanode as (index) ranges start..end

Keys are indexes!

block ranges can be merged and split

a set of block ranges

The clustername

Returns this as factory for inclusion into Netplex_main.startup

Returns this as factory for inclusion into Netplex_main.startup

Shared memory for transmitting data from the main process to the I/O processes and vice versa

The directory containing the store.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Authentication in server context

Configuration of mapred servers

Defining the map/reduce job

Access to record fields

Filesystem abstraction

Utility library for record-based I/O

Background reader for Mapred_io

Run a map/red job

Extract job configuration, and marshalling

Execute a scheduled job by submitting tasks to servers

Main program for the task server and job controller, for this single job only

Registered functions (low-level)

Scheduler

Sorter classes

Splits a line into key and value.

Support for streaming

Executes this task

Factory for the processor "mapred_task_server"

Copy files to and from the task nodes

Representation of tasks

Map/reduce toolkit

Block allocation algorithm

Block allocation algorithm

Management of authentication tickets

Support module for blocklists

Already allocated blocks of this inode.

Fast RPC to the blockmap server via camlboxes

Two-phase commit protocol

This monitor ensures that only certain commits can be done in parallel.

Configuration of name nodes

The implementation of the namenode component for the coordinator node

Manage tickets and push ticket secrets to the datanodes

Getting news about datastores and their availability

Datastore types

Namenode access routines to database

The dentrycache checks whether directory metadata is still up to date.

Election of the coordinator

The inodecache checks whether inode metadata is still up to date.

Returns this as factory for inclusion into Netplex_main.startup

Management of datastores, i.e.

Push updates to the slaves (client)

The implementation of the namenode component for slave nodes

State manager: Provide the logical view of the state, and forward changes to the database and to the nameslaves (2-phase commit)

Users and groups

Authentication in server context

Condition variables

Utility functions for database accesses

Manager for potentially persisting objects

Plasma Key Value API

Helpers for analyzing blocklists

Client access to the Plasma Filesystem

Configuration of clients

Buffer for read and write

The inodecache stores the metadata found in inodes, namely the inodeinfo struct, and subsets of the block list.

Accessing PlasmaFS via the Netfs.stream_fs interface

Some C functions

Random numbers

Management of large buffers incl.

Users and groups

Platform-specific functions

Exports a hash and cmp function for Mapred_sorters.generic_sorter.

Exports a hash and cmp function for Mapred_sorters.generic_sorter.