IBM Hardware Performance Monitor (hpm)

1
IBM Hardware Performance Monitor (hpm)

• NPACI Parallel Computing Workshop February 5,
  2002 at SDSC

2
What is Performance?

• Where is time spent and how is time spent?
• MIPS Millions of Instructions Per Second
• not necessarily indicative of the amount of
  useful work done
• MFLOPS Millions of Floating-Point Operations
  Per Second
• A better metric for numerically intensive codes,
  but different platforms measure Flops
  differently, and Flops is not completely
  indicative of useful work done
• Run time/CPU time
• The only true measure of code performance!
  accounts for algorithmic improvements to code.
  Can be converted to cycles.
• Counting cycles means
• Estimate how many cycles your loop(s) should take
• Compare to measured times(converted to cycles)
  and tune the code to narrow the difference

3
What is a Performance Monitor?

• Provides detailed processor/system data
• Processor Monitors
• Typically a group of registers
• Special purpose registers keep track of
  programmable events
• Non-intrusive counts result in accurate
  measurement of processor events
• Typical Events counted are Instruction, floating
  point instr, cache misses, etc.
• System Level Monitors
• Can be h/w or s/w
• Intended to measure system activity
• Examples
• bus monitor measures memory traffic, can analyze
  cache coherency issues in multiprocessor system
• Network monitor measures network traffic, can
  analyze web traffic internally and externally

4
Hardware Counter Motivations

• To understand execution behavior of application
  code
• Why not use software?
• Strength simple, GUI interface
• Weakness large overhead, intrusive, higher
  level, abstraction and simplicity
• How about using a simulator?
• Strength control, low-level, accurate
• Weakness limit on size of code, difficult to
  implement
• When should we directly use hardware counters?
• Software and simulators not available or not
  enough
• Strength non-intrusive, instruction level
  analysis, moderate control, very accurate, low
  overhead
• Weakness not typically reusable, OS kernel
  support

5
Problem Set

• Should we collect all events all the time?
• No. not necessary and wasteful
• What counts should be used?
• Safe to say gather only what you need

• Cycles
• Committed Instructions
• Loads
• Stores
• L1/L2 misses
• L1/L2 stores

• Committed fl pt instr
• Branches
• Branch misses
• TLB misses
• Cache misses

6
POWER3 Architecture

7
IBM HPM Toolkit

• High Performance Monitor
• Developed for performance measurement of
  applications running on IBM Power3 systems. It
  consists of
• An utility (hpmcount)
• An instrumentation library (libhpm)
• A graphical user interface (hpmviz).
• Requires PMAPI kernel extensions to be loaded
• Works on IBM 630 and 604e processors

8
HPM Count

• Utilities for performance measurement of
  application
• Extra logic inserted to the processor to count
  specific events
• Updated at every cycle
• Provide a summary output at the end of the
  execution
• Wall clock time
• Resource usage statistics
• Hardware performance counters information
• Derived hardware metrics

9
HPM Usage HW Event Categories

• EVENT SET 1
• Cycles
• Inst. Completed
• TLB misses
• Stores completed
• Loads completed
• FPU0 ops
• FPU1 ops
• FMAs executed 

EVENT SET 2 Cycles Inst. Completed TLB
misses Stores dispatched L1 store misses Loads
dispatched L1 load misses LSU idle 
EVENT SET 3 Cycles Inst. dispatched Inst.
Completed Cycles w/  0 inst. completed I cache
misses FXU0 ops FXU1 ops FXU2 ops
EVENT SET 4 Cycles Loads dispatched L1 load
misses L2 load misses Stores dispatched L2 store
misses Comp. unit waiting on load LSU idle
floating point performance and usage of floating
point units
performance and usage of level 1 instruction cache
usage of level 2 data cache and branch prediction
data locality and usage of level 1 data cache
10
HPM for Whole Program using HPMCOUNT

• Installed in /usr/local/apps/hpm,
  /usr/local/apps/HPM_V2.3
• Environment setting
• setenv LIBHPM_EVENT_SET 1 (2,3,4)
• setenv MP_LABELIO YES -gt to correlate each
  line of output with corresponding task
• setenv MP_STDOUTMODE -gttaskID(e.g. 0) to
  discard output from other tasks
• Usage
• poe hpmcount ./a.out -nodes 1 -tasks_per_node
  1 -rmpool 1 -s ltsetgt -e ev,ev -h
• -h displays a help message
• -e ev0,ev1, list of event numbers, separated by
  commas
• evltigt corresponds to event selected for counter
  ltIgt
• -s predefined set of envets

11
Derived Hardware Metrics

• Hardware counters provide only raw counts
• 8 counters on Power3
• Enough info for generation of derived metrics on
  each execution
• Derived Metrics
• Floating point rate
• Computational Intensity
• Instruction per load / store
• Load/store per data cache misses
• Cache hit rate
• Loads per load miss
• Stores per store miss
• Loads per TLB miss
• FMA
• Branches Misspredicted

12
HPMCOUNT Output (Event1)

• ---usage statistics---
• Total amount of time in user mode
  141.130000 seconds
• Total amount of time in system mode
  36.300000 seconds
• Maximum resident set size
  25516 Kbytes
• Average shared memory use in text segment
  1978356 Kbytessec
• Average unshared memory use in data segment
  357949904 Kbytessec
• Number of page faults without I/O activity
  6750
• Number of page faults with I/O activity
  81
• Number of times process was swapped out 0
• Number of times file system performed INPUT
  0
• Number of times file system performed OUTPUT 0
• Number of IPC messages sent
  0
• Number of IPC messages received
  0
• Number of signals delivered
  0
• Number of voluntary context switches
  266907
• Number of involuntary context switches
  2128527

13
HPMCOUNT (Event1 continued)

• ---Resource statistics---
• Wall Clock Time 35.099596 seconds
• Total time in user mode 54.0518473203182
  seconds
• Average duration 0.0146248
• Standard deviation 0.0112495
• Exclusive duration 0.191238 seconds
• PM_CYC (Cycles)
  20271809159
• PM_INST_CMPL (Instructions completed)
  14974657747
• PM_TLB_MISS (TLB misses)
  4474101
• PM_ST_CMPL (Stores completed)
  2687036544
• PM_LD_CMPL (Loads completed)
  5220888450
• PM_FPU0_CMPL (FPU 0 instructions)
  2581927160
• PM_FPU1_CMPL (FPU 1 instructions)
  519835526
• PM_EXEC_FMA (FMAs executed)
  792849657

14
HPMCOUNT (Event1 continued)

• Utilization rate
  153.988
• Avg number of loads per TLB miss
  1166.913
• Load and store operations
  7907.925 M
• Instructions per load/store
  1.894
• MIPS
  426.633
• Instructions per cycle
  0.739
• HW Float points instructions per Cycle
  0.153
• Floating point instructions FMAs
  3894.612 M
• Float point instructions FMA rate
  110.959 Mflip/s
• FMA percentage
  40.715
• Computation intensity
  0.492

15
HPM for Part of Program using LIBHPM

• Instrumentation of performance library for
  performance measurement of Fortran, C and C
  applications
• Collects information and performs summarization
  during run-time, generate performance file for
  each task
• Use the same set of hardware counters events used
  by hpmcount
• User can specify an event set with the file
  libHPMevents
• For each instrumented point in a program, libhpm
  provides output
• Total count
• Total duration (wall clock time)
• Hardware performance counters information
• Hardware derived metrics
• Supports
• multiple instrumentation points, nested
  instrumentation
• OpenMP and thread applications
• Multiple calls to an instrumented point

16
LIBHPM Functions

• C C
• hpmInit(taskID)
• hpmTerminate(taskID)
• hpmStart(instID)
• hpmStop(instID)
• hpmTstart(instID)
• hpmTstop(instID)

• Fortran
• f_hpminit(taskID)
• f_hpmterminate(taskID)
• f_ hpmstart(instID)
• f_ hpmstop(instID)
• f_ hpmtstart(instID)
• f_ hpmtstop(instID)

17
Using LIBHPM - C

• Declaration
• include libhpm.h
• C usage
• MPI_Comm_rank( MPI_COMM_WORLD, taskID)
• hpmInit(taskID,hpm_test)
• hpmStart(1,outer call)
• code segment to be timed
• hpmStop( 1)
• hpmTerminate(taskID)
• Compilation
• mpcc_r -I/usr/local/apps/HPM_V2.3/include -O3
  -lhpm_r -lpmapi -lm -qarchpwr3 -qstrict
  -qsmpomp -L/usr/local/apps/HPM_V2.3/lib
  hpm_test.c -o hpm_test.x

18
Using LIBHPM - Fortran

• Declaration
• include f_hpm.h
• Fortran usage
• CALL MPI_COMM_RANK( MPI_COMM_WORLD, taskid,
  ierr )
• call f_hpminit(taskID)
• call f_hpmstart(instID)
• code segment to be timed
• call f_hpmstop(instID)
• call f_hpmterminate(taskID)
• CALL MPI_FINALIZE(ierr)
• Compilation
• mpxlf_r -I/usr/local/apps/HPM_V2.3/include
  -qsuffixcppf -O3 -qarchpwr3 -qstrict -qsmpomp
  -L/usr/local/apps/HPM_V2.3/lib -lhpm_r -lpmapi
  -lm hpm_test.f -o hpm_test.x

19
Using LIBHPM - Threads

• call f_hpminit(taskID)
• //do
• call f_hpmtstart(10)
• do_work
• call f_hpmtstop(10)
• end //do
• //do
• call f_hpmtstart(20my_thread_ID )
• do_work
• call f_hpmtstop(20my_thread_ID )
• end //do
• call f_hpmterminate(taskID)

20
HPM Code in C
include ltmpi.hgt include ltstdio.hgt include
"libhpm.h" define n 10000 main(int argc, char
argv) int taskID,i,numprocs double
an,bn,cn MPI_Init(argc,argv) MPI_Comm_si
ze(MPI_COMM_WORLD,numprocs) MPI_Comm_rank(MPI_CO
MM_WORLD,taskID) hpmInit(taskID,"hpm_test") hpm
Start(1,section 1") for(i1iltn1i) aii
bin-1 hpmStop(1)
hpmStart(2, "section 2") for(i2iltn1i)
ciaibiai/bi hpmStop(2) hpmTermin
ate(taskID) MPI_Finalize()

21
HPM Code in Fortran

• program hpm_test
• parameter (n10000)
• integer taskID,ierr,numtasks
• dimension a(n),b(n),c(n)
• include "mpif.h"
• include "f_hpm.h"
• call MPI_INIT(ierr)
• call MPI_COMM_RANK(MPI_COMM_WORLD,taskID,ier
  r)
• call MPI_COMM_SIZE(MPI_COMM_WORLD,numtasks,i
  err)
• call f_hpminit(taskID,"hpm_test")
• call f_hpmstart(5,section1")
• do i1,n
• a(i)real(i)
• b(i)real(n-i)
• enddo

call f_hpmstop(5) call
f_hpmterminate(taskID) call
MPI_FINALIZE(ierr) end
22
Compiling and Linking

• FF mpxlf_r
• HPM_DIR /usr/local/apps/HPM_V2.3
• HPM_INC -I(HPM_DIR)/include
• HPM_LIB -L(HPM_DIR)/lib -lhpm_r -lpmapi -lm
• FFLAGS -qsuffixcppf -O3 -qarchpwr3 -qstrict
  -qsmpomp
• Note -qsuffixcppf is only required for
  Fortran code with .f
• hpm_test.x hpm_test.f
• (FF) (HPM_INC) (FFLAGS) hpm_test.f (HPM_LIB)
  -o hpm_test.x

23
HPMVIZ

• takes as input the performance files generated by
  libhpm
• Usage
•  gt hpmviz ltperformance files(.viz)gt
• define a range of values considered satisfactory
• Red below predefined as minimum recommended
  value
• Green above the threshold value
• HPMVIZ left pane of the window
• displays for each instrumented point, identified
  by its label, the inclusive duration, exclusive,
  and count.
• HPMVIZ right pane of the window
• shows the corresponding source code which can be
  edited and saved.
• The metrics windows
• display the task ID, Thread ID, count, exclusive
  duration, inclusive duration, and the derived
  hardware metrics.

24
HPMVIZ
25
IBM SP HPM Toolkit Summary

• A complete problem set
• Derived metrics
• Analysis of error message
• Analyze derived metrics
• HPMCOUNT very accurate with low overhead,
  non-intrusive, general view for whole program
• LIBHPM same sets as hpmcount, for part of
  program
• HPMVIZ easier to view the hardware counters
  information and derived metrics

26
HPM References

• HPM README file in /usr/local/apps/HPM_V2.3
• Online Documentation
• http//www.sdsc.edu/SciApps/IBM_tools/hpm.html