You use the management API /metrics/systemLoad resource to request system-load information from an S Series Node. The S Series Node response to this request includes statistics for the processing resources listed below. Each returned statistic is either for a point in time or for a duration of approximately one minute.
- CPUs
- The statistics for CPU usage are:
- Percent of CPU processing power in use on each server module and on the S Series Node as a whole. The S Series Node value is the average of the two server module values.
- Percent of CPU processing time spent waiting for I/O operations to finish or waiting to acquire processing resources on each server module since the server module last started and on the S Series Node as a whole. The S Series Node value is the average of the two server module values.
- Network bandwidth
- The statistics for network-bandwidth usage are:
- For each server module:
- Rate of data transmission on each access network port on each server module during the past minute, in bits per second
- Rate of data reception on each access network port on each server module during the past minute, in bits per second
- Larger of the total rate of data transmission and total rate of data reception on each server module during the past minute, in bits per second
- Total of the bandwidth capacity that can be used for data transmission on each access network port on each server module, in bits per second
- Total of the bandwidth capacity that can be used for data reception on each access network port on each server module, in bits per second
- Larger of the total bandwidth capacity that can be used for data transmission and the total bandwidth capacity that can be used for data reception on each server module, in bits per second
- Percent of the total bandwidth capacity in use for data transmission or data reception on each server module
- For the S Series Node as a whole:
- Sum of the larger of the total rate of data transmission and total rate of data reception on each server module during the past minute, in bits per second.
- Sum of the larger of the total amount of bandwidth that can be used for data transmission and the total amount of bandwidth that can be used for data reception on each server module, in bits per second.
- Percent of the total amount of bandwidth in use for data transmission or data reception on both server modules. The S Series Node value is calculated independently of the values for the two server modules.
- For each server module:
- I/O threads
- The statistics for I/O-thread usage are:
- Number of I/O threads in use on each server module and on the S Series Node as a whole. The S Series Node value is the sum of the two server module values.
- Total number of I/O threads on each server module and on the S Series Node as a whole. The S Series Node value is the sum of the two server module values.
- Percent of I/O threads in use on each server module and on the S Series Node as a whole. The S Series Node value is calculated independently of the values for the two server modules.
- Memory
- The statistics for memory usage are:
- Amount of memory in use on each server module and on the S Series Node as a whole, in bytes. The S Series Node value is the sum of the two server module values.
- Total amount of memory on each server module and on the S Series Node as a whole, in bytes. The S Series Node value is the sum of the two server module values.
- Percent of memory in use on each server module and on the S Series Node as a whole. The S Series Node value is calculated independently of the values for the two server modules.
- Hitachi API for Amazon S3 (S3 compatible API)
- The statistics for usage of the S3 compatible API are, for each of the PUT, GET, DELETE, HEAD, and POST operations:
- Rate of throughput for operations of the given type on each server module and on the S Series Node as a whole during the past minute, in operations per second. The S Series Node value is the sum of the two server module values.
All operations of the applicable type are included in the rate calculation, regardless of whether the operations succeeded or ended with errors.
- Rate of data throughput for operations of the given type on each server module and on the S Series Node as a whole during the past minute, in bytes per second. The S Series Node value is the sum of the two server module values.
All operations of the applicable type are included in the rate calculation, regardless of whether the operations succeeded or ended with errors.
- Percent of operations of the given type that succeeded on each server module and on the S Series Node as a whole. The S Series Node value is calculated independently of the values for the two server modules.
- Percent of operations of the given type that terminated with errors on each server module and on the S Series Node as a whole. The S Series Node value is calculated independently of the values for the two server modules.
- Average latency of the operations of the given type on each server module and on the S Series Node as a whole during the past minute, in milliseconds. The S Series Node value is the sum of the two server module values.
- Rate of throughput for operations of the given type on each server module and on the S Series Node as a whole during the past minute, in operations per second. The S Series Node value is the sum of the two server module values.