Hardware Requirements for
                              Helix GitSwarm

                              Version 2015.3

Introduction

    This document describes the hardware requirements for Helix GitSwarm
    release 2015.3.

Supported Operating System Platforms

    * Ubuntu 12.04 or 14.04
    * CentOS 6.7 or 7

Supported Git Fusion Releases

    * Git Fusion 2015.2, or newer.

Storage

    The necessary hard drive space largely depends on the size of the repos
    you want to store in GitSwarm. As a rule of thumb you should have at
    least twice as much free space as all your repos combined take up. You
    need twice the storage because GitSwarm satellites contain an extra copy
    of each repo.

    Note that if you are mirroring projects to Helix Server using the locally
    provisioned Helix Git Fusion Server, you will want at least four times as
    much free space as all your repos combined.

    If you want to be flexible about growing your hard drive space in the
    future, consider mounting it using LVM so you can add more hard drives
    when you need them.

    Apart from a local hard drive, you can also mount a volume that supports
    the network file system (NFS) protocol. This volume might be located on
    a file server, a network attached storage (NAS) device, a storage area
    network (SAN) or on an Amazon Web Services (AWS) Elastic Block Store
    (EBS) volume.

    If you have enough RAM and a recent CPU, the speed of GitSwarm is mainly
    limited by hard drive seek times. Having a fast drive (7,200 RPM and up)
    or a solid state drive (SSD) will improve the responsiveness of GitSwarm.

CPU

    For production use, it is recommended GitSwarm be run on a dedicated server.
    The Helix Server and Helix Git Fusion products should ideally be installed
    on their own independent machines. In that configuration:

    * 1 core supports up to 100 users, but performance may suffer as all workers
      and background jobs running on the same core
    * 2 cores is the recommended number of cores and supports up to 500 users
    * 4 cores supports up to 2,000 users
    * 8 cores supports up to 5,000 users
    * 16 cores supports up to 10,000 users
    * 32 cores supports up to 20,000 users
    * 64 cores supports up to 40,000 users

    By default, GitSwarm will attempt to automatically provision an instance
    of Helix Server and Helix Git Fusion all on the local system. When running
    all components on the same machine, we suggest a minimum of 4 cores.

    Note: A minimum of 3 unicorn workers is required for concurrent usage of
          the system. Generally we recommend having unicorn workers set as
          CPU cores + 1 = unicorn workers.

Memory

    You need at least 2GB of addressable memory (RAM + swap) to install and
    use GitSwarm! With less memory, GitSwarm will give strange errors during
    the reconfigure run and 500 errors during usage.

    For production use, it is recommended GitSwarm be run on a dedicated server.
    The Helix Server and Helix Git Fusion products should ideally be installed
    on their own independent machines. In that configuration:

    * 512MB RAM + 1.5GB of swap is the absolute minimum, but we strongly
      advise against this amount of memory.
    * 1GB RAM + 1GB swap supports up to 100 users
    * 2GB RAM is the recommended memory size and supports up to 100 users
    * 4GB RAM supports up to 1,000 users
    * 8GB RAM supports up to 2,000 users
    * 16GB RAM supports up to 4,000 users
    * 32GB RAM supports up to 8,000 users
    * 64GB RAM supports up to 16,000 users
    * 128GB RAM supports up to 32,000 users

    By default, GitSwarm will attempt to automatically provision an instance
    of Helix Server and Helix Git Fusion all on the local system. When running
    all components on the same machine, we suggest a minimum of 4GB of RAM and
    4GB of swap.

    Note: The 25 workers of Sidekiq will show up as separate processes in
          your process overview (such as top or htop) but they share the same
          RAM allocation since Sidekiq is a multithreaded application.

END