tt_chip_count

Name

Prometheus Metric Name

tt_chip_count

Metric Path (tt-telemetry)

Schema:

{hostname}/tt_chip_count

Example path:

bh-glx-c09u02/tt_chip_count

Description

The number of Tenstorrent chips discovered on the host according to registered PCI devices. The value is read from PCIDevice::enumerate_devices when chip metrics are created after a successful driver initialization. It is refreshed on later collection cycles.

This is a host-level metric (not scoped to a tray or chip path). It is only emitted when device telemetry collection is enabled and UMD initialization succeeds. The metric is removed when the driver is torn down (for example during a warm reset) and is recreated on the next successful initialization.

Values

Type: Unsigned Integer

Units: None

Allowable values: Non-negative integer representing the total pci chip count (for example 8 on an eight-chip host).

Related Metrics

Metric	Description
tt_expected_chip_count	Expected chip count on Galaxy hosts (32); compare against `tt_chip_count` to detect a mismatch.
tt_driver_initialized	Whether the device driver is initialized. Chip count is assessed independently of whether or not driver initialization succeeds.
tt_pcie_link_alive	Per-chip PCIe liveness (only created for MMIO-capable Wormhole and Blackhole chips).
tt_noc_alive	Per-chip NOC liveness (only created for MMIO-capable Wormhole and Blackhole chips).

Prometheus Labels

Label Name	Value
hostname	The host from which the metric was collected.
hall	The datacenter hall where the host is located. Sourced from the Factory System Descriptor (FSD).
aisle	The datacenter aisle where the host is located. Sourced from the Factory System Descriptor (FSD).
rack	The rack number where the host is located. Sourced from the Factory System Descriptor (FSD).
shelf_u	The shelf U position in the rack where the host is located. Sourced from the Factory System Descriptor (FSD).