TensorHive is an open source tool for monitoring and managing computing resources across multiple hosts. It solves the most common problems and nightmares about accessing and sharing your AI-oriented infrastructure across multiple, often competing users.
It's designed with simplicity, flexibility and configuration-friendliness in mind.
GPU Reservation calendar
Each column represents all reservation events for a GPU on a given day. In order to make a new reservation simply click and drag with your mouse, select GPU(s), add some meaningful title, optionally adjust time range.
If there are many hosts and GPUs in our infrastructure, you can use our simplified, horizontal calendar to quickly identify empty time slots and filter out already reserved GPUs. image
From now on, only your processes are eligible to run on reserved GPU(s). TensorHive periodically checks if some other user has violated it. He will be spammed with warnings on all his PTYs, emailed every once in a while, additionally admin will also be notified (it all depends on the configuration).
|Terminal warning||Email warning||Admin warning|
Infrastructure monitoring dashboard
Accessible infrastructure can be monitored in the Nodes overview tab. Sample screenshot:
Here you can add new watches, select metrics and monitor ongoing GPU processes and its' owners.
Thanks to the
Task execution module, you can define commands for tasks you want to run on any configured nodes.
You can manage them manually or set spawn/terminate date.
Commands are run within
screen session, so attaching to it while they are running is a piece of cake.
It provides a simple, but flexible (framework-agnostic) command templating mechanism that will help you automate multi-node trainings.
Additionally, specialized templates help to conveniently set proper parameters for chosen well known frameworks:
In the examples
directory, you will find sample scenarios of using the
Task execution module for various
frameworks and computing environments.
TensorHive requires that users who want to use this feature must append TensorHive's public key to their
~/.ssh/authorized_keys on all nodes they want to connect to.
Our goal is to provide solutions for painful problems that ML engineers often have to struggle with when working with remote machines in order to run neural network trainings.
You should really consider using TensorHive if anything described in profiles below matches you:
- You're an admin, who is responsible for managing a cluster (or multiple servers) with powerful GPUs installed.
- :angry: There are more users than resources, so they have to compete for it, but you don't know how to deal with that chaos
- :ocean: Other popular tools are simply an overkill, have different purpose or require a lot of time to spend on reading documentation, installation and configuration (Grafana, Kubernetes, Slurm)
- :penguin: People using your infrastructure expect only one interface for all the things related to training models (besides terminal): monitoring, reservation calendar and scheduling distributed jobs
- :collision: Can't risk messing up sensitive configuration by installing software on each individual machine, prefering centralized solution which can be managed from one place
- You're a standalone user who has access to beefy GPUs scattered across multiple machines.
- :part_alternation_mark: You want to be able to determine if batch size is too small or if there's a bottleneck when moving data from memory to GPU - charts with metrics such as
mem_usedare great for this purpose
- :date: Visualizing names of training experiments using calendar helps you track how you're progressing on the project
- :snake: Launching distributed trainings is essential for you, no matter what the framework is
- :dizzy_face: Managing a list of training commands for all your distributed training experiments drives you nuts
- :zzz: Remembering to manually launch the training before going sleep is no fun anymore
Advantages of TensorHive
:zero: Dead-simple one-machine installation and configuration, no
:one: Users can make GPU reservations for specific time range in advance via reservation mechanism
:arrow_right: no more frustration caused by rules: "first come, first served" or "the law of the jungle".
:two: Users can prepare and schedule custom tasks (commands) to be run on selected GPUs and hosts
:arrow_right: automate and simplify distributed trainings - "one button to rule them all"
:three: Gather all useful GPU metrics, from all configured hosts in one dashboard
:arrow_right: no more manual logging in to each individual machine in order to check if GPU is currently in use or not
For more details, check out the full list of features.
- All nodes must be accessible via SSH, without password, using SSH Key-Based Authentication (How to set up SSH keys - explained in Quickstart section)
- Only NVIDIA GPUs are supported (relying on
- Currently TensorHive assumes that all users who want to register into the system must have identical UNIX usernames on all nodes configured by TensorHive administrator (not relevant for standalone developers)
- (optional) We recommend installing TensorHive on a separate user account (for example
tensorhive) and adding this user to the
pip install tensorhive
(optional) For development purposes we encourage separation from your current python packages using e.g. virtualenv, Anaconda.
git clone https://github.com/roscisz/TensorHive.git && cd TensorHive pip install -e .
TensorHive is already shipped with newest web app build, but in case you modify the source, you can can build it with
make app (currently on
master branch). For more useful commands see our Makefile.
Build tested with
Node v14.15.1 and
init command will guide you through basic configuration process:
You can check connectivity with the configured hosts using the
(optional) If you want to allow your UNIX users to set up their TensorHive accounts on their own and run distributed
Task execution plugin, use the
key command to generate the SSH key for TensorHive:
Now you should be ready to launch a TensorHive instance:
Web application and API Documentation can be accessed via URLs highlighted in green (Ctrl + click to open in browser).
You can fully customize TensorHive behaviours via INI configuration files (which will be created automatically after
~/.config/TensorHive/main_config.ini ~/.config/TensorHive/mailbot_config.ini ~/.config/TensorHive/hosts_config.ini
- [x] :mag_right: Monitor metrics on each host
- [x] :tm: Nvidia GPUs
- [x] :pager: CPU, RAM
- [ ] :open_file_folder: HDD
- [x] :customs: Protection of reserved resources
- [x] :warning: Send warning messages to terminal of users who violate the rules
- [x] :mailbox_with_no_mail: Send e-mail warnings
- [x] :closed_lock_with_key: Grant users and groups access to specific GPUs at given time schedules
- [ ] :bomb: Kill unwanted processes
- [X] :rocket: Task execution and scheduling
- [x] :old_key: Execute any command in the name of a user
- [x] :alarm_clock: Schedule spawn and termination
- [x] :repeat: Synchronize process status
- [x] :factory: Use
screencommand as backend - user can easily attach to running task
- [x] :skull: Remote process interruption, termination and kill
- [x] :floppy_disk: Save stdout to disk
- [x] :watch: Track wasted (idle) time during reservation
- [x] :hocho: Gather and calculate average gpu and mem utilization
- [ ] :loudspeaker: Remind user when his reservation starts and ends
- [ ] :incoming_envelope: Send e-mail if idle for too long
- [x] :chart_with_downwards_trend: Configurable charts view
- [x] Metrics and active processes
- [ ] Detailed hardware specification
- [x] :calendar: Calendar view
- [x] Allow making reservations for selected GPUs
- [x] Edit reservations
- [x] Cancel reservations
- [x] Attach jobs to reservation
- [x] :baby_symbol: Task execution
- [x] Create parametrized tasks and assign to hosts, automatically set
- [x] Buttons for task spawning/scheduling/termination/killing actions
- [x] Fetch log produced by running task
- [x] Group actions (spawn, schedule, terminate, kill selected)
- [x] Create parametrized tasks and assign to hosts, automatically set
- [ ] :straight_ruler: Detailed hardware specification panel (CPU clock speed, RAM, etc.)
- [ ] :penguin: Admin panel
- [ ] User banning
- [ ] Accept/reject reservation requests
- [ ] Modify rules on-the-fly (without restarting)
- [ ] Show popups to users (something like message of the day -