Changes between Version 5 and Version 6 of BoincOverview


Ignore:
Timestamp:
Sep 16, 2018, 2:12:33 PM (3 years ago)
Author:
davea
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • BoincOverview

    v5 v6  
    66rather than low turnaround time of individual jobs.
    77It also offers low-level mechanisms for distributed data storage.
     8BOINC has a client/server architecture:
     9the '''server''' distributes jobs,
     10while the '''client''' runs on worker nodes, which execute jobs.
    811
    9 BOINC has a client/server architecture.
    10 The '''server''' distributes jobs.
    11 The '''client''' runs on worker nodes, which execute jobs.
     12BOINC can be used in two ways, depending on where the worker nodes come from:
    1213
    13 BOINC was originally designed for '''volunteer computing''',
    14 where the worker nodes are consumer devices (desktop and laptop computers,
    15 tables, smartphones) volunteered by their owners.
    16 It addresses the various challenges inherent in this environment
    17 (heterogeneity, host churn and unreliability, scale, security, and so on).
     14 * In '''volunteer computing''',
     15  the worker nodes are consumer devices (desktop and laptop computers,
     16  tablets, smartphones) volunteered by their owners.
     17  BOINC addresses the various challenges inherent in this environment
     18  (heterogeneity, host churn and unreliability, scale, security, and so on).
     19 There are a number of volunteer-computing '''BOINC projects'''
     20 such as SETI@home, LHC@home, IBM World Community Grid, and so on.
     21 The BOINC client can be "attached" to one or many of these;
     22 it processes jobs for the projects to which it is attached.
    1823
    19 There are a number of volunteer-computing '''BOINC projects'''
    20 such as SETI@home, LHC@home, IBM World Community Grid, and so on.
    21 The BOINC client can be "attached" to one or many of these;
    22 it processes jobs for the projects to which it is attached.
     24 * BOINC can also be used for '''in-house computing''' within an organization (e.g. a company).
     25  In this case case the worker nodes are
     26  cluster nodes or other organizational computers,
     27  and they are attached to the organization's server.
    2328
    24 BOINC can also be used for '''in-house computing''' within an organization (e.g. a company).
    25 In this case case the worker nodes are
    26 cluster nodes or other organizational computers,
    27 and they are attached to the organization's server.
     29BOINC can run all existing HTC applications,
     30including those that use GPUs and/or multiple CPU cores.
     31It can use virtual machines to run existing Linux applications on Windows and Mac worker nodes.
     32
     33BOINC provides mechanisms for job submission and control, designed for performance at scale.
     34However, it can also be used as a back end for existing
     35job-submission systems such as HTCondor; details are [GridIntegration here].
    2836
    2937BOINC is distributed under the LGPL v3 open-source license.
     
    3543To compute using BOINC, you'll need to set up a BOINC server
    3644and configure your applications to run under BOINC.
    37 Instructions for doing this are [ProjectMain here].
     45Technical documentation is [ProjectMain here].
    3846
    3947If you're doing in-house computing,