Column

Description

System

The system being measured.

Machine Type

The machine type

CPU Type

CPU type (CP, zIIP etc.)

CPI

Cycles per Instruction. The average number of machine cycles to execute 1 instruction.

MIPS

Millions of Instructions Per Second. The number of instructions executed per second of CPU time.

L1MP

Level 1 Miss Per 100 Instructions. The number of times required data or instructions were not in the processor level 1 cache, per 100 instructions.

RNI

Relative Nest Intensity. A relative measurement of the cost of  sourcing data and instructions from cache and memory, when they were not in the Level 1 cache. Calculated using processor specific formulas published by IBM.

Workload Hint

LSPR workload hint. Derived from the L1MP and RNI using the table published by IBM. Can be Low, Average or High.

L15P

Percent of level 1 misses sourced from the level 1.5 cache (z10).

L2P

Percent of level 1 misses sourced from the level 2 cache (z196/zEC12/z13).

L2LP

Percent of level 1 misses sourced from the level 2 cache on the same book. (z10)

L2RP

Percent of level 1 misses sourced from the level 2 cache on a different book. (z10)

L3P

Percent of level 1 misses sourced from the level 3 cache (z196/zEC12/z13).

L4LP

Percent of level 1 misses sourced from the level 4 cache on the same book (z196/zEC12/z13).

L4RP

Percent of level 1 misses sourced from the level 4 cache on a different book (z196/zEC12/z13).

MEMP

Percent of level 1 misses sourced from memory.

Prb State

Percentage of instructions executed in problem state.

Finite CPI

Estimated Cycles Per Instruction resulting from a finite level 1 cache i.e. the cycles to source data from other cache levels or memory, per instruction.

Complexity CPI

Estimated Cycles Per Instruction if there was infinite level 1 processor cache i.e. the cycles per instruction attributable to complexity of the instructions being executed.

Parameter

Description

System

Select the system to display.

Type

Subtype

113

2