Skip to main content

Run high power workload on all connected chips

Project description

TT-BURNIN

Tenstorrent Burnin (TT-Burnin) is a command line utility to run a high power consumption workload on TT devices.

Official Repository

https://github.com/tenstorrent/tt-burnin/

Getting started

Build and editing instruction are as follows -

Building from Git

After cloning the repo, install and source rust for the luwen library

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
source "$HOME/.cargo/env"

Upgrade pip to the latest and install tt-burnin

pip3 install --upgrade pip
pip3 install .

Optional - for TT-Tools developers

Generate and source a python3 environment

python3 -m venv .venv
source .venv/bin/activate
pip3 install --upgrade pip

For users who would like to edit the code without re-building, install burnin in editable mode.

pip3 install --editable .

Usage

Command line arguments

usage: tt-burnin [-h] [-v] [--reset_file reset_config.json]

Getting Help!

Running tt-burnin with the -h, --help flag should bring up something that looks like this

usage: tt-burnin [-h] [-v] [--reset_file reset_config.json]

Tenstorrent Burnin (TT-Burnin) is a command line utility to run a high power consumption workload on TT devices.

optional arguments:
  -h, --help            show this help message and exit
  -v, --version         show program's version number and exit
  --reset_file reset_config.json
                        Provide a custom reset json file for the host.Generate a default reset json file with the -g option with tt-smi.

Running tt-burnin

After building run tt-burnin to run the program.

TT-Burnin performs the following steps when running:

  1. Reset the boards on the host to get them into a known good state
  2. Start the power hungry workload on all boards
  3. Output a realtime telemetry command line widget to monitor the devices
  4. After user hits "enter" to stop the workload, another reset is performed to bring the boards back to known good state

A full run of burnin should look like -

$ tt-burnin

 Detected Chips: 3
┏━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Pci Dev ID ┃ Board Type ┃ Device Series ┃ Board Number    ┃ Coordinates  ┃
┡━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ 0          │ grayskull  │ e75           │ 100007311523010 │ N/A          │
│ 1          │ wormhole   │ n300 L        │ 10001451170801d │ [0, 0, 0, 0] │
│ N/A        │ wormhole   │ n300 R        │ 10001451170801d │ [1, 0, 0, 0] │
└────────────┴────────────┴───────────────┴─────────────────┴──────────────┘
 Resetting devices on host... 
 Re-initializing boards after reset.... 
 Detected Chips: 3

 Starting TT-Burnin workload on all boards. WARNING: Opening SMI might cause unexpected behavior 
                                                                                                                                                               
┏━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┓
┃ ID ┃ Core Voltage (V) ┃ Core Current (A) ┃ AICLK (MHz) ┃ Power (W)     ┃ Core Temp (°C) ┃
┡━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━┩
│ 0  │ 0.74 / 0.84      │  73.0 / 170.0    │  653 / 1000 │  54.0 /  56.0 │ 41.3 / 75.0    │
│ 1  │ 0.75 / 0.95      │ 110.0 / 160.0    │  872 / 1000 │  84.0 /  85.0 │ 37.9 / 75.0    │
│ 2  │ 0.75 / 0.95      │ 110.0 / 160.0    │  885 / 1000 │  85.0 /  85.0 │ 33.4 / 75.0    │
└────┴──────────────────┴──────────────────┴─────────────┴───────────────┴────────────────┘
 Press Enter to STOP TT-Burnin on all boards...

 Stopping TT-Burnin workload on all boards. 

 Resetting devices on host... 
 Re-initializing boards after reset.... 
 Detected Chips: 3

License

Apache 2.0 - https://www.apache.org/licenses/LICENSE-2.0.txt

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

tt_burnin-0.2.4.tar.gz (230.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

tt_burnin-0.2.4-py3-none-any.whl (221.7 kB view details)

Uploaded Python 3

File details

Details for the file tt_burnin-0.2.4.tar.gz.

File metadata

  • Download URL: tt_burnin-0.2.4.tar.gz
  • Upload date:
  • Size: 230.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for tt_burnin-0.2.4.tar.gz
Algorithm Hash digest
SHA256 25a4ce8713821a5e0d67066919ab4f644fe34e935c58bde1f24b94eb851a4bd3
MD5 c9807b26c0c93e581cbdf20318d2e71a
BLAKE2b-256 fe152a59ca4a3ec8e29f7451e270895a82f0fb82728e62bf60de6c4f0cabf3b9

See more details on using hashes here.

Provenance

The following attestation bundles were made for tt_burnin-0.2.4.tar.gz:

Publisher: release.yml on tenstorrent/tt-burnin

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file tt_burnin-0.2.4-py3-none-any.whl.

File metadata

  • Download URL: tt_burnin-0.2.4-py3-none-any.whl
  • Upload date:
  • Size: 221.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for tt_burnin-0.2.4-py3-none-any.whl
Algorithm Hash digest
SHA256 77e2b7b66640dc33a5031c7761047e6ef6f7e5c5ee0f21bee0ef2c911b455dc2
MD5 62d8293177c9ce9ee8b3f8c173c52667
BLAKE2b-256 a73be3b65859a3ba48f3e737f2e308dc23efd95891750a66cd6f4cbc3becdee8

See more details on using hashes here.

Provenance

The following attestation bundles were made for tt_burnin-0.2.4-py3-none-any.whl:

Publisher: release.yml on tenstorrent/tt-burnin

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page