Sparklens is a profiling tool for Spark with a built-in Spark scheduler simulator. Its primary goal is to make it easy to understand the scalability limits of Spark applications. It helps in understanding how efficiently a given Spark application is using the compute resources provided to it. Maybe your application will run faster with more executors and may be it wont. Sparklens can answer this question by looking at a single run of your application.
It helps you narrow down to few stages (or driver, or skew or lack of tasks) which are limiting your application from scaling out and provides contextual information about what could be going wrong with these stages. Primarily it helps you approach spark application tuning as a well defined method/process instead of something you learn by trial and error, saving both developer and compute time.
http://sparklens.qubole.com is a reporting service built on top of Sparklens. This service was built to lower the pain of sharing and discussing Sparklens output. Users can upload the Sparklens JSON file to this service and retrieve a global sharable link. The link delivers the Sparklens report in an easy-to-consume HTML format with intuitive charts and animations. It is also useful to have a link for easy reference for yourself, in case some code changes result in lower utilization or make the application slower.
- Estimated completion time and estimated cluster utilisation with different numbers of executors
Executor count 31 ( 10%) estimated time 87m 29s and estimated cluster utilization 92.73%
Executor count 62 ( 20%) estimated time 47m 03s and estimated cluster utilization 86.19%
Executor count 155 ( 50%) estimated time 22m 51s and estimated cluster utilization 71.01%
Executor count 248 ( 80%) estimated time 16m 43s and estimated cluster utilization 60.65%
Executor count 310 (100%) estimated time 14m 49s and estimated cluster utilization 54.73%
Given a single run of a Spark application, Sparklens can estimate how your application will perform given any arbitrary number of executors. This helps you understand the ROI on adding executors.
- Job/stage timeline which shows how the parallel stages were scheduled within a job. This makes it easy to visualise the DAG with stage dependencies at the job level.
07:05:27:666 JOB 151 started : duration 01m 39s
[ 668 |||||||||||||||||||||| ]
[ 669 ||||||||||||||||||||||||||| ]
[ 673 ]
[ 674 |||| ]
[ 675 ||||||| ]
[ 676 |||||||||||||| ]
[ 677 ||||||| ]
[ 678 | ]
[ 679 |||||||||||||||||||||||||||||||]
*Lots of interesting per-stage metrics like Input, Output, Shuffle Input and Shuffle Output per stage. OneCoreComputeHours available and used per stage to discover inefficient stages.
Total tasks in all stages 189446
Per Stage Utilization
Stage-ID Wall Task Task IO% Input Output ----Shuffle----- -WallClockTime- --OneCoreComputeHours--- MaxTaskMem
Clock% Runtime% Count Input | Output Measured | Ideal Available| Used%|Wasted%
0 0.00 0.00 2 0.0 254.5 KB 0.0 KB 0.0 KB 0.0 KB 00m 04s 00m 00s 05h 21m 0.0 100.0 0.0 KB
1 0.00 0.01 10 0.0 631.1 MB 0.0 KB 0.0 KB 0.0 KB 00m 07s 00m 00s 08h 18m 0.2 99.8 0.0 KB
2 0.00 0.40 1098 0.0 2.1 GB 0.0 KB 0.0 KB 5.7 GB 00m 14s 00m 00s 16h 25m 3.2 96.8 0.0 KB
3 0.00 0.09 200 0.0 0.0 KB 0.0 KB 5.7 GB 2.3 GB 00m 03s 00m 00s 04h 35m 2.6 97.4 0.0 KB
4 0.00 0.03 200 0.0 0.0 KB 0.0 KB 2.3 GB 0.0 KB 00m 01s 00m 00s 01h 13m 2.9 97.1 0.0 KB
7 0.00 0.03 200 0.0 0.0 KB 0.0 KB 2.3 GB 2.7 GB 00m 02s 00m 00s 02h 27m 1.7 98.3 0.0 KB
8 0.00 0.03 38 0.0 0.0 KB 0.0 KB 2.7 GB 2.7 GB 00m 05s 00m 00s 06h 20m 0.6 99.4 0.0 KB
Internally, Sparklens has the concept of an analyzer which is a generic component for emitting interesting events. The following analyzers are currently available:
- AppTimelineAnalyzer
- EfficiencyStatisticsAnalyzer
- ExecutorTimelineAnalyzer
- ExecutorWallclockAnalyzer
- HostTimelineAnalyzer
- JobOverlapAnalyzer
- SimpleAppAnalyzer
- StageOverlapAnalyzer
- StageSkewAnalyzer
We are hoping that Spark experts the world over will help us with ideas or contributions to extend this set. Similarly, Spark users can help us in finding what is missing here by raising challenging tuning questions.
Use the following arguments to spark-submit
or spark-shell
:
--packages qubole:sparklens:0.3.1-s_2.11
--conf spark.extraListeners=com.qubole.sparklens.QuboleJobListener
You can choose not to run sparklens inside the app, but at a later time. Run your app as above with additional configuration parameters:
--packages qubole:sparklens:0.3.1-s_2.11
--conf spark.extraListeners=com.qubole.sparklens.QuboleJobListener
--conf spark.sparklens.reporting.disabled=true
This will not run reporting, but instead create a Sparklens JSON file for the application which is
stored in the spark.sparklens.data.dir directory (by default, /tmp/sparklens/). Note that this will be stored on HDFS by default. To save this file to s3, please set spark.sparklens.data.dir to s3 path. This data file can now be used to run Sparklens reporting independently, using spark-submit
command as follows:
./bin/spark-submit --packages qubole:sparklens:0.3.1-s_2.11 --class com.qubole.sparklens.app.ReporterApp qubole-dummy-arg <filename>
<filename>
should be replaced by the full path of sparklens json file. If the file is on s3 use the full s3 path. For files on local file system, use file:// prefix with the local file location. HDFS is supported as well.
You can also upload a Sparklens JSON data file to http://sparklens.qubole.com to see this report as an HTML page.
You can also run Sparklens on a previously run spark-app using an event history file, (similar to
running via sparklens-json-file
above) with another option specifying that is file is an
event history file. This file can be in any of the formats the event history files supports, i.e. text, snappy, lz4
or lzf. Note the extra source=history
parameter in this example:
./bin/spark-submit --packages qubole:sparklens:0.3.1-s_2.11 --class com.qubole.sparklens.app.ReporterApp qubole-dummy-arg <filename> source=history
It is also possible to convert an event history file to a Sparklens json file using the following command:
./bin/spark-submit --packages qubole:sparklens:0.3.1-s_2.11 --class com.qubole.sparklens.app.EventHistoryToSparklensJson qubole-dummy-arg <srcDir> <targetDir>
EventHistoryToSparklensJson is designed to work on local file system only. Please make sure that the source and target directories are on local file system.
sbt compile
sbt package
sbt clean
You will find the Sparklens jar in the target/scala-2.11
directory. Make sure the Scala and Java versions correspond to those required by your Spark cluster. We have tested it with Java 7/8,
Scala 2.11.8 and Spark versions 2.0.0 and onwards.
Once you have the Sparklens JAR available, add the following options to your spark-submit
command line:
--jars /path/to/sparklens_2.11-0.3.1.jar
--conf spark.extraListeners=com.qubole.sparklens.QuboleJobListener
You could also add this to your cluster's spark-defaults.conf so that it is automatically available for all applications.
It is possible to use Sparklens in your development cycle using notebooks. Sparklens keeps lots of information in-memory. To make it work with notebooks, it tries to minimize the amount of memory by keeping limited history of jobs executed in Spark.
- Add this as the first cell
QNL = sc._jvm.com.qubole.sparklens.QuboleNotebookListener.registerAndGet(sc._jsc.sc())
import time
def profileIt(callableCode, *args):
if (QNL.estimateSize() > QNL.getMaxDataSize()):
QNL.purgeJobsAndStages()
startTime = long(round(time.time() * 1000))
result = callableCode(*args)
endTime = long(round(time.time() * 1000))
time.sleep(QNL.getWaiTimeInSeconds())
print(QNL.getStats(startTime, endTime))
- Wrap your code in some python function say myFunc
profileIt(myFunc)
As you can see this is not the only way to use it from Python. The core function is: QNL.getStats(startTime, endTime)
Another way to use this tool, so that we don’t need to worry about objects going out of scope is:
Create the QNL object as part of the first paragraph For every piece of code that requires profiling:
if (QNL.estimateSize() > QNL.getMaxDataSize()):
QNL.purgeJobsAndStages()
startTime = long(round(time.time() * 1000))
<-- Your Python code here -->
endTime = long(round(time.time() * 1000))
time.sleep(QNL.getWaiTimeInSeconds())
print(QNL.getStats(startTime, endTime))
QNL.purgeJobsAndStages()
is responsible for making sure that the tool doesn’t use too much memory.
It removes historical information, throwing away data about old stages to keep the memory usage
by the tool modest.
- Add this as the first cell
import com.qubole.sparklens.QuboleNotebookListener
val QNL = new QuboleNotebookListener(sc.getConf)
sc.addSparkListener(QNL)
- Anywhere you need to profile the code:
QNL.profileIt {
//Your code here
}
It is important to realize that QNL.profileIt
takes a block of code as input. Hence any variables declared in this
part are not accessible after the method returns. Of course it can refer to other code/variables in scope.
The way to go about using this tool with notebooks is to have only one cell in the profiling scope. The moment you are happy with the results, just remove the profiling wrapper and execute the same cell again. This will ensure that your variables come back in scope and are accessible to next cell. Also note that, the output of the tool in notebooks is little different from what you would see in command line. This is just to make the information concise. We will be making this part configurable.
- Introduction to Sparklens
- Video from meetup: Concepts behind Sparklens
- Slides from meetup
- Video from Fifth Elephant Conference
- Video from Spark AI Summit London 2018
- [03/20/2018] Version 0.1.1 - Sparklens Core
- [04/06/2018] Version 0.1.2 - Package name fixes
- [08/07/2018] Version 0.2.0 - Support for offline reporting
- [01/10/2019] Version 0.2.1 - Stability fixes
- [05/10/2019] Version 0.3.0 - Support for handling parallel Jobs
- [05/10/2019] Version 0.3.1 - Fixed JSON parsing issue with Spark 2.4.0 and above
We haven't given this much thought. Just raise a PR and if you don't hear from us, shoot an email to [email protected] to get our attention.
Please use the GitHub issues for the Sparklens project to report issues or raise feature requests. If you can code, better raise a PR.