/*! @page wtperf Simulating workloads with wtperf
The WiredTiger distribution includes a tool that can be used to simulate
workloads in WiredTiger, in the directory \c bench/wtperf.
The \c wtperf utility generally has two phases, the populate phase which
creates a database and then populates an object in that database, and a
workload phase, that does some set of operations on the object.
For example, the following configuration uses a single thread to
populate a file object with 500,000 records in a 500MB cache. The
workload phase consists of 8 threads running for two minutes, all
reading from the file.
@code
conn_config="cache_size=500MB"
table_config="type=file"
icount=500000
run_time=120
populate_threads=1
threads=((count=8,reads=1))
@endcode
In most cases, where the workload is the only interesting phase, the
populate phase can be performed once and the workload phase run
repeatedly (for more information, see the wtperf \c create configuration
variable).
The \c conn_config configuration supports setting any WiredTiger
connection configuration value. This is commonly used to configure
statistics with regular reports, to obtain more information from the
run:
@code
conn_config="cache_size=20G,statistics=(fast,clear),statistics_log=(wait=600)"
report_interval=5
@endcode
Note quoting must be used when passing values to Wiredtiger
configuration, as opposed to configuring the \c wtperf utility itself.
The \c table_config configuration supports setting any WiredTiger object
creation configuration value, for example, the above test can be
converted to using an LSM store instead of a B+tree store, with
additional LSM configuration, by changing \c conn_config to:
@code
table_config="lsm=(chunk_size=5MB),type=lsm,os_cache_dirty_max=16MB"
@endcode
More complex workloads can be configured by creating more threads doing
inserts and updates as well as reads. For example, to configure two
inserting threads two threads doing a mixture of inserts, reads and
updates:
@code
threads=((count=2,inserts=1),(count=2,inserts=1,reads=1,updates=1))
@endcode
Example \c wtperf configuration files can be found in the
\c bench/wtperf/runners/ directory.
There are also a number of command line arguments that can be passed
to \c wtperf:
@par -C config
Specify configuration strings for the ::wiredtiger_open function.
This argument is additive to the \c conn_config parameter in the
configuration file.
@par -h directory
Specify a database home directory. The default is \c ./WT_TEST.
@par -m monitor_directory
Specify a directory for all monitoring related files.
The default is the database home directory.
@par -O config_file
Specify the configuration file to run.
@par -o config
Specify configuration strings for the \c wtperf program.
This argument will override settings in the configuration file.
@par -T config
Specify configuration strings for the WT_SESSION::create function.
This argument is additive to the \c table_config parameter in the
configuration file.
@section monitor Monitoring wtperf
Like all WiredTiger applications, the \c wtperf command can be configured
with statistics logging, and the resulting output displayed using the
\c wtstats visualization tool. For more information, see @ref wtstats.
In addition to statistics logging, \c wtperf can monitor performance and
operation latency times. Monitoring is enabled using the \c sample_interval
configuration. For example to record information every 10 seconds, set the
following on the command line or add it to the \c wtperf configuration file:
@code
sample_interval=10
@endcode
Enabling monitoring causes \c wtperf to create a file \c monitor in the
database home directory (or another directory as specified using the
\c -m option to \c wtperf).
The same visualization tool, \c wtstats, can be used to view a combined
chart with both the \c monitor output and the statistics logging output
at the same time.
The following example shows how to run the \c medium-btree.wtperf configuration
with monitoring enabled, and then generate a graph.
@code
# Change into the WiredTiger directory.
cd wiredtiger
# Configure and build WiredTiger if not already built.
./configure && make
# Remove and re-create the run directory.
rm -rf WTPERF_RUN && mkdir WTPERF_RUN
# Run the medium-btree.wtperf workload, sampling performance every 5 seconds.
bench/wtperf/wtperf \
-h WTPERF_RUN \
-o sample_interval=5 \
-O bench/wtperf/runners/medium-btree.wtperf
# Use the visualization tool to create HTML graph output; the output file is
# named wtstats.html.
python tools/wtstats/wtstats.py WTPERF_RUN/monitor
# Possible alternatives if statistics logging also enabled:
# python tools/wtstats/wtstats.py WTPERF_RUN/monitor WTPERF_RUN/WiredTigerStat*
# python tools/wtstats/wtstats.py WTPERF_RUN
@endcode
The python command creates a file named \c wtstats.html in the current
working directory. You can open the generated HTML document in your browser
and see the generated statistics.
@section config Wtperf configuration options
The following is a list of the currently available \c wtperf
configuration options:
\if START_AUTO_GENERATED_WTPERF_CONFIGURATION
DO NOT EDIT: THIS PART OF THE FILE IS GENERATED BY dist/s_docs.
\endif
@par async_threads (unsigned int, default=0)
number of async worker threads
@par checkpoint_interval (unsigned int, default=120)
checkpoint every interval seconds during the workload phase.
@par checkpoint_stress_rate (unsigned int, default=0)
checkpoint every rate operations during the populate phase in the populate thread(s), 0 to disable
@par checkpoint_threads (unsigned int, default=0)
number of checkpoint threads
@par conn_config (string, default="create")
connection configuration string
@par close_conn (boolean, default=true)
properly close connection at end of test. Setting to false does not sync data to disk and can result in lost data after test exits.
@par compact (boolean, default=false)
post-populate compact for LSM merging activity
@par compression (string, default="none")
compression extension. Allowed configuration values are: 'none', 'lz4', 'snappy', 'zlib', 'zstd'
@par create (boolean, default=true)
do population phase; false to use existing database
@par database_count (unsigned int, default=1)
number of WiredTiger databases to use. Each database will execute the workload using a separate home directory and complete set of worker threads
@par drop_tables (boolean, default=false)
Whether to drop all tables at the end of the run, and report time taken to do the drop.
@par icount (unsigned int, default=5000)
number of records to initially populate. If multiple tables are configured the count is spread evenly across all tables.
@par idle_table_cycle (unsigned int, default=0)
Enable regular create and drop of idle tables, value is the maximum number of seconds a create or drop is allowed before flagging an error. Default 0 which means disabled.
@par index (boolean, default=false)
Whether to create an index on the value field.
@par insert_rmw (boolean, default=false)
execute a read prior to each insert in workload phase
@par key_sz (unsigned int, default=20)
key size
@par log_partial (boolean, default=false)
perform partial logging on first table only.
@par log_like_table (boolean, default=false)
Append all modification operations to another shared table.
@par min_throughput (unsigned int, default=0)
notify if any throughput measured is less than this amount. Aborts or prints warning based on min_throughput_fatal setting. Requires sample_interval to be configured
@par min_throughput_fatal (boolean, default=false)
print warning (false) or abort (true) of min_throughput failure.
@par max_latency (unsigned int, default=0)
notify if any latency measured exceeds this number of milliseconds. Aborts or prints warning based on min_throughput_fatal setting. Requires sample_interval to be configured
@par max_latency_fatal (boolean, default=false)
print warning (false) or abort (true) of max_latency failure.
@par pareto (unsigned int, default=0)
use pareto distribution for random numbers. Zero to disable, otherwise a percentage indicating how aggressive the distribution should be.
@par populate_ops_per_txn (unsigned int, default=0)
number of operations to group into each transaction in the populate phase, zero for auto-commit
@par populate_threads (unsigned int, default=1)
number of populate threads, 1 for bulk load
@par random_range (unsigned int, default=0)
if non zero choose a value from within this range as the key for insert operations
@par random_value (boolean, default=false)
generate random content for the value
@par range_partition (boolean, default=false)
partition data by range (vs hash)
@par read_range (unsigned int, default=0)
scan a range of keys after each search
@par readonly (boolean, default=false)
reopen the connection between populate and workload phases in readonly mode. Requires reopen_connection turned on (default). Requires that read be the only workload specified
@par reopen_connection (boolean, default=true)
close and reopen the connection between populate and workload phases
@par report_interval (unsigned int, default=2)
output throughput information every interval seconds, 0 to disable
@par run_ops (unsigned int, default=0)
total read, insert and update workload operations
@par run_time (unsigned int, default=0)
total workload seconds
@par sample_interval (unsigned int, default=0)
performance logging every interval seconds, 0 to disable
@par sample_rate (unsigned int, default=50)
how often the latency of operations is measured. One for every operation, two for every second operation, three for every third operation etc.
@par sess_config (string, default="")
session configuration string
@par session_count_idle (unsigned int, default=0)
number of idle sessions to create. Default 0.
@par table_config (string, default="key_format=S,value_format=S,type=lsm,exclusive=true, allocation_size=4kb,internal_page_max=64kb,leaf_page_max=4kb, split_pct=100")
table configuration string
@par table_count (unsigned int, default=1)
number of tables to run operations over. Keys are divided evenly over the tables. Cursors are held open on all tables. Default 1, maximum 99999.
@par table_count_idle (unsigned int, default=0)
number of tables to create, that won't be populated. Default 0.
@par threads (string, default="")
workload configuration: each 'count' entry is the total number of threads, and the 'insert', 'read' and 'update' entries are the ratios of insert, read and update operations done by each worker thread; If a throttle value is provided each thread will do a maximum of that number of operations per second; multiple workload configurations may be specified per threads configuration; for example, a more complex threads configuration might be 'threads=((count=2,reads=1)(count=8,reads=1,inserts=2,updates=1))' which would create 2 threads doing nothing but reads and 8 threads each doing 50% inserts and 25% reads and updates. Allowed configuration values are 'count', 'throttle', 'update_delta', 'reads', 'inserts', 'updates', 'truncate', 'truncate_pct' and 'truncate_count'. There are also behavior modifiers, supported modifiers are 'ops_per_txn'
@par transaction_config (string, default="")
WT_SESSION.begin_transaction configuration string, applied during the populate phase when populate_ops_per_txn is nonzero
@par table_name (string, default="test")
table name
@par truncate_single_ops (boolean, default=false)
Implement truncate via cursor remove instead of session API
@par value_sz_max (unsigned int, default=1000)
maximum value size when delta updates are present. Default disabled
@par value_sz_min (unsigned int, default=1)
minimum value size when delta updates are present. Default disabled
@par value_sz (unsigned int, default=100)
value size
@par verbose (unsigned int, default=1)
verbosity
@par warmup (unsigned int, default=0)
How long to run the workload phase before starting measurements
\if STOP_AUTO_GENERATED_WTPERF_CONFIGURATION
DO NOT EDIT: THIS PART OF THE FILE IS GENERATED BY dist/s_docs.
\endif
*/