|
Summary: A tool for running performance benchmarks on the space in various scenarios.
OverviewThe benchmark example provides a good tool for running performance benchmarks on the cache in various scenarios. This program performs a loop of write/put and read/get or take/remove operations from a space according to a different set of parameters. The result is the average time it took to perform the operations. You can define a sampling rate to allow you to track the intermediate performance while the benchmark example is running. You may also dump the results into a file to be converted into graphs and analyzed using a spreadsheet and analysis tools. You can find the benchmark in the <GigaSpaces Root>\examples\Advanced\Benchmark directory. Typing run.bat -h in the console displays:
This print-out is displayed below: Option arguments in square brackets [] are required;
Option arguments in triangular brackets <> are optional.
Setup:
---------
-objecttype Defines the object type that will be written to the space.
Usage: -objecttype ext
Other possible values are:
entry - Object implements net.jini.core.entry.Entry interface.
pojo - Plain Java object
uid - Object extends com.j_spaces.core.client.MetaDataEntry using the ClientUIDHandler.
fifo - Object extends com.j_spaces.core.client.MetaDataEntry and sets FIFO=true.
ser - Object implements net.jini.core.entry.Entry and java.io.Serializable interfaces.
Contains additional field with a complex list, therefore should NOT be compared with the following objects: entry, pojo, uid, fifo, jms
ext - Object implements net.jini.core.entry.Entry and java.io.Externalizable interfaces.
Contains additional field with a complex list, therefore should NOT be compared with the following objects: entry, pojo, uid, fifo, jms
pojo-ext - Plain Java object implements java.io.Externalizable interface.
Contains additional field with a complex list, therefore should NOT be compared with the following objects: entry, pojo, uid, fifo, jms
fifo-ext - Object extends com.j_spaces.core.client.MetaDataEntry implements java.io.Externalizable interface and sets FIFO=true.
Contains additional field with a complex list, therefore should NOT be compared with the following objects: entry, pojo, uid, fifo, jms
uid-ext - Object extends com.j_spaces.core.client.MetaDataEntry
implements java.io.Externalizable interface and using the ClientUIDHandler. Contains additional field with a complex list,
therefore should NOT be compared with the following objects: entry, pojo, uid, fifo, jms
mde-ext - Object extends com.j_spaces.core.client.MetaDataEntry implements java.io.Externalizable interface.
Contains additional field with a complex list, therefore should NOT be compared with the following objects: entry, pojo, uid, fifo, jms
jms - com.j_spaces.jms.GSSimpleMessageImpl, a basic JMS message which extends com.j_spaces.core.client.MetaDataEntry
implements java.io.Externalizable and javax.jms.Message interfaces.
For basic JMS send or sync/async receive, pass -write for JMS send, -take for JMS sync receive or -notify for JMS async receive.
-clean Clean space before benchmark starts
-url [url] Connection url; If none provided the one defined as part of the
script file will be used
-f [FileName] dump results into file. Works with -showrate option. Default File name is BenchMarkResult{Date}_{Time}.
-pause Pauses once connection with space is established.
-exit Benchmark will hang once finished. When used with embedded space, space will still be alive.
Operations:
-------------
-write perform first operation as write/put
-read perform second operation as read/get
-update perform second operation as update (with uid)
-take perform second/third operation as take/remove
-all perform all operations - first, second and then
third
-execute [ first/second ] enables execution of only first or second operation.
-bench perform first-*second-second-third operation with uid
Options:
----------
-i <number of iterations> number of iterations; default is 1000
-lease [time in ms] lease timeout in milliseconds of the write/put operation
-t [time in ms] timeout in milliseconds of the read/take/get/remove operations
-s [size in bytes] set byte size of the entry object.
-content changes the content value each iteration
-tr [number of threads] number of threads performing each operation
-rangefirst [from-to] will perform first operation with ids/keys between range
-rangesecond [from-to] will perform second operation with ids/keys between range
-repeatfirst [repeats] repeat times of first operation
-repeatsecond [repeats] repeat times of second operation
-m [batch size] specifies the batch size when performing batch operations
-rand <number of iterations> will randomize second operation ids/keys in a range; default is by -i
-tx [nth-iteration] tx will be committed every nth-iteration; if zero, as specified by -i
-delaywrite [t ms] delays first operation every t milliseconds
-delayreadtake [t ms] delays second & third operation every t milliseconds
-writerate [Max Rate msg/sec] will limit the first operation rate (msg in 1 second) for all threads
-notify registers for notifications on all events
-parallel parallel execution of the operations
-usesinglespace This parameter indicates if operation will be performed using a single direct space proxy (and not clustered proxy)
-returnlease This parameter indicates if write operation will return a lease object
Transactions:
--------------
-tx 2000 Perform operation under local transaction, commit every 2000 operations.
-dtx 1000 Perform operation under distributed transaction, commit every 1000 operations.
Topologies:
-------------
-cache Turn on local cache mode; default is false
-map Map API - first/second/third operations as: put, get, remove.
-map -cache Map API with local cache.
-target [space name] Space name to perform second operations on (in a cluster configuration)
-hashtable use java.util.Hashtable API- first/second/third operations as: put,get, remove
Statistics:
-------------
-showrate <iteration cycle> global throughput will be displayed every iteration cycle; default is 1000
-showthreadrate <iteration cycle> thread throughput will be displayed every iteration cycle; default is 1000
-stress [repeat times] runs the entire benchmark as many as 'repeat times' as stated
Examples:
------------
-url -url rmi://localhost/container/space
protocol://[host]:<port>/[container name]/[space name]
-f -f resultsfile.xsl -showrate
dump results into resultsfile.xsl
-execute -execute first
will perform only write/put operations
-execute -execute second
will perform only read operations
-execute -execute second -take
will perform only take operations
-bench -bench -map
will perform put,put(update),get,remove with uid
-bench -bench
will perform write,update,read,take with uid
-lease -i 1000 -lease 20000 -execute first
perform 1st operation with entry lease of 2 sec
-t -i 1000 -t 20000 -execute second
perform 2nd operation with timeout of 2 sec
-s -s 1000
define 1K size entries each
-content -i 1000 -content -s 1000
defines an entry with a changing content of 1K
-tr -i 1000 -tr 4
will perform 1000 iterations by each of the 4 threads
-rangefirst -rangefirst 1000-2000
will write/put ids/keys from 1000 to 2000
-rangesecond -rangesecond 1000-2000
will read/take/get/remove ids/keys from 1000 to 2000
-repeatfirst -i 1000 -repeatfirst 10
will write/put 10 times 1000 entries
-repeatsecond -i 1000 -repeatsecond 10
will read/get or take/remove 10 times 1000 entries
-m -m 100 -i 1000
will use 10 batches of 100 to perform operations on 1000
entries.
-rand -i 1000 -rand
will randomize second operation ids/keys from 0 to 1000
-rand -i 10000 -rand 40000
will randomize between ids/keys in the range of 0-40000
-rand -rangesecond 10000-25000 -rand
will randomize between 10000-25000
-writerate -i 100000 -writerate 50000 -tr 4
4 threads will write 100000 entries each with a maximum
TP rate of 50000 msg/sec
-notify -notify -i 1000 -tr 5
will start 5 threads that will write and notify 1000 entries
-parallel -parallel -i 1000 -tr 5
will start 5 parallel threads, each performing 1000 operations
-usesinglespace -usesinglespace -i 100000
will write 100000 objects to specific single cluster space
-returnlease -returnlease -i 100000
will write 100000 objects with returning Lease object
-dtx-manager-url -dtx 1000 dtx-manager-url thunder1:3733
Use distributed transaction. Commit every 1000 operations.
Transaction Manager registered on LUS that is running on machine thunder1:3733
-cache -cache
will use local cache to remote space.
-map -map -all -i 1000
will put/get/remove 1000 entries.
-map -map -cache -i 1000 -repeatsecond 4
will put 1000 entries, first get from space, successive
gets from local cache.
-target -target rmi://host:port/container/space_name
will perform second operation this target space
-showrate -showrate 10000
will show TP every 10000 iterations
-showthreadrate -showthreadrate 10000
will show each thread-TP every 10000 iterations
-stress -map -all -stress 10
will run 10 cycles of map api put/get/remove
More ExamplesThe following example uses the JavaSpaces API, writes 1000 Entries into the space, 1K each, and reads them back into the client. run -read -objecttype entry -i 1000 -s 1024 The following example uses the JavaSpaces API, writes 1000 POJOs into the space, 1K each, and takes them from the space. A remote cache is accessed. run -take -objecttype pojo -i 1000 -s 1024 The following example uses the JavaSpaces API, writes 1000 Entries into the space in FIFO mode, 1K each, and reads them back into the client three times. run -read -objecttype fifo -i 1000 -s 1024 -rt 3 The following example uses the Map API, puts 1000 Entries into the space, 1K each, and gets them back into the client. run -map -read -i 1000 -s 1024 The following example uses the Map API, put 1000 Entries into the space, 1K each, and removes them from the space. A remote space is accessed. run -map -take -i 1000 -s 1024 The following example uses the Map API, puts 1000 Entries into the space, 1K each, and gets them back into the client three times. run -map -read -i 1000 -s 1024 -rt 3 The following example uses the Map API, puts 1,000,000 Entries into the space, gets them back, and removes them from the space. Throughput is displayed every 100,000 operations. run -map -all -i 1000000 -showrate 100000 The following example uses the JMS API, sends 1,000,000 JMS messages into the space (MessageProducer.send()), and receives them asynchronously (MessageListener.onMessage()) with notifications. run -i 1000000 -notify -objecttype jms The following example uses the JMS API, sends 1,000,000 JMS messages into the space (MessageProducer.send()), and receives them synchronously (MessageConsumer.receive()). Throughput is displayed every 100,000 operations. run -i 1000000 -take -objecttype jms -showrate 100000 |
Previous release
The returnlease, clusteredoperation, and dtx-manager-url parameters are new in GigaSpaces version 6.0.2 and onwards.
The rangefirst and rangesecond parameters are supported only with single operations (and not with multiple/batch operations). 
Add GigaSpaces wiki search to your browser search engines! (works on Firefox 2 and Internet Explorer 7)
