17.05 - I/O Hard Limits Columns - Teradata Database

Teradata Vantageā„¢ - Resource Usage Macros and Tables

prodname
Advanced SQL Engine
Teradata Database
vrm_release
17.00
17.05
created_date
June 2020
category
Administration
User Guide
featnum
B035-1099-170K

The Input/Output Token Allocations (IOTA) columns provide the percentage of disk throughput that is used.

Column Name Mode Description Data Type
FullPotentialIota count Sum of the full potential input or output token allocations from all devices attached to the node. This field is reported only to the AMP vproc rows.

The input or output token allocations are not limited by the I/O_COD setting.

The data source for this field is located in the all_wd_full_potential_iota field in the /proc/tdmeter/limit_stats file.

FLOAT
CodPotentialIota count Sum of the potential input or output token allocations from all devices attached to the node for this workload. This field is reported only to the AMP vproc rows.

The input or output token allocations are accumulated only when an I/O is pending on a device, and they are limited by the I/O_COD setting.

The data source for this field is located in the wd_potential_iota field in the /proc/tdmeter/limit_stats file.

FLOAT
UsedIota count Sum of the used input or output token allocations from all devices attached to the node for this workload. This field is reported only to AMP vproc rows.

The data source for this field is located in the wd_used_iota field in the /proc/tdmeter/limit_stats file.

FLOAT
IoThrottleCount count Number of times an I/O was throttled. FLOAT
IoThrottleTime count Total I/O Throttle Time in milliseconds. FLOAT
IoThrottleTimeMax max Maximum I/O Throttle Time in milliseconds. FLOAT
IoThrottleCntZeroIotas count

Number of times an I/O was throttled due to no IOTA tokens available in the bucket.

This data is included in the IoThrottleCount.

FLOAT
IoThrottleCntInsuffIotas count

Number of times an I/O was throttled due to insufficient IOTA tokens available in the bucket. There were some IOTA tokens in the bucket, but not enough.

This data is included in the IoThrottleCount.

FLOAT
IoThrottleCntInsuffIotasHL count

Number of times an I/O was throttled due to insufficient IOTA tokens available in the bucket for Hard Limits. There were some IOTA tokens in the bucket, but not enough.

This data is included in the IoThrottleCount.

FLOAT
IoThrottleCntMaxQD count

Number of times an I/O was throttled due to the queue depth to the storage being at maximum.

This data is not part of IoThrottleCount.

FLOAT