I was running an application on AWS EMR-Spark. Here, is the spark-submit job;-
Arguments : spark-submit --deploy-mode cluster --class com.amazon.JavaSparkPi s3://spark-config-test/SWALiveOrderModelSpark-1.0.assembly.jar s3://spark-config-test/2017-08-08
AWS uses YARN for resource management. I was looking at the metrics (screenshot below), and have a doubt regarding the YARN 'container' metrics.
Here, the container allocated is shown as 2. However, I was using 4 nodes (3 slave + 1 master),all 8 cores CPU. So, how are only 2 container allocated?
A couple of thing you need to do. First of all, you need to set the following configuration in capacity-scheduler.xml
"yarn.scheduler.capacity.resource-calculator":"org.apache.hadoop.yarn.util.resource.DominantResourceCalculator"
otherwise YARN will not use all the cores you specify. Secondly, you need to actually specify the number of executors you need, and also the number of cores you need and the amount of memory you want allocated on executors (and possibly on the driver as well, if you either have many shuffle partitions or if you collect data to the driver).
YARN is designed to manage clusters running many different jobs at the time, so it will not per default assign all ressources to a single job, unless you force it to by setting the above mentioned setting. Furthermore, the default setting for Spark are also not sufficient for most jobs and you need to set them explicitly. Please have a read through this blog post to get a better understanding of how to tune spark settings for optimal performance.
Related
There are parameters that decide the maximum, minimum and total of the memory and cpu that yarn can allocate via containers
example:
yarn.nodemanager.resource.memory-mb
yarn.scheduler.maximum-allocation-mb
yarn.scheduler.minimum-allocation-mb
yarn.nodemanager.resource.cpu-vcores
yarn.scheduler.maximum-allocation-vcores
yarn.scheduler.minimum-allocation-vcores
There are also spark side parameters that seemingly would control similar kind of allocations:
spark.executor.instances
spark.executor.memory
spark.executor.cores
etc
What happens when the two set of parameters are infeasible according to the bounds set by the other. For example: What if yarn.scheduler.maximum-allocation-mb is set to 1G and the spark.executor.memory is set to 2G? Similar conflicts and infeasibilities could be imagined for the other parameters as well.
What happens in such cases? And, what is the suggested way to set these parameters?
When running Spark on YARN, each Spark executor runs as a YARN container
So take spark.executor.memory as an example:
If spark.executor.memory is 2G and yarn.scheduler.maximum-allocation-mb is 1G, then your container will be OOM killer
If spark.executor.memory is 2G and yarn.scheduler.minimum-allocation-mb is 4G, then your container is much bigger than needed by the Spark application
Suggestions for setting parameters depend on your hardware resources and other services running on this machine. You can try to use the default value first, and then make adjustments by monitoring machine resources
This excellent https://community.cloudera.com/t5/Support-Questions/Yarn-container-size-flexible-to-satisfy-what-application-ask/m-p/115458 and Difference between `yarn.scheduler.maximum-allocation-mb` and `yarn.nodemanager.resource.memory-mb`? should give you the basics. Additionally, here is a good SO-related answer Spark on YARN resource manager: Relation between YARN Containers and Spark Executors.
TL;DR
As you are not talking about Kubernetes, then YARN as Resource / Cluster Mgr allocates Executors with needed resouces, based on Spark params / defaults that are allocated based on those YARN params for the Containers.
1 Container = 1 Executor. Some state incorrectly, 1 Container N Executors, not so.
There is minimum allocation and max allocation of resources, based on those YARN params. So,YARN will provide Executors with some wastage of resources, if it can - or of restricted size.
If non-dynamic YARN resource allocation, then Apps start with less resources, else there will be a wait to get all resources, and those acquired are not available for others.
There is also a fair scheduler for more smooth, uniform throughput for many concurrent apps.
We are currently running parallel Spark jobs on an EMR cluster using HadoopActivity task from Datapipeline. By default, the newer versions of EMR clusters sets spark dynamic allocation to true which will increase/ reduce the number of executors required based on the load. So do we need to set any other property along with spark-submit e.g. number of cores, executor memory etc. or its best to have EMR cluster handle it dynamically?
This always depends of how you application is working. I can give you an good example of how I work here. For the Data Scientists in general they use the default configuration and it works pretty well due to they use Jupyter here to run their models. The only thing that we setup that can be useful for you is the conf spark.dynamicAllocation.minExecutors this allow to setup at least two or one worker for the job. To not be without any executor. That is what we do with the Data Scientists.
But, EMR has one specific type of configuration for each type of machine you choose. So in general it is optimized for the most common activities. But sometimes you need to change according your request, if you need more memory and less cores for skewed data that is better to change.
Using Spark(1.6.1) standalone master, I need to run multiple applications on same spark master.
All application submitted after first one, keep on holding 'WAIT' state always. I also observed, the one running holds all cores sum of workers.
I already tried limiting it by using SPARK_EXECUTOR_CORES but its for yarn config, while I am running is "standalone master". I tried running many workers on same master but every time first submitted application consumes all workers.
I was having same problem on spark standalone cluster.
What I got is, Somehow it is utilising all the resources for one single job. We need to define the resources so that their will be space to run other job as well.
Below is the command I am using to submit spark job.
bin/spark-submit --class classname --master spark://hjvm1:6066 --deploy-mode cluster --driver-memory 500M --conf spark.executor.memory=1g --conf spark.cores.max=1 /data/test.jar
A crucial parameter for running multiple jobs in parallel on a Spark standalone cluster is spark.cores.max. Note that spark.executor.instances,
num-executors and spark.executor.cores alone won't allow you to achieve this on Spark standalone, all your jobs except a single active one will stuck with WAITING status.
Spark-standalone resource scheduling:
The standalone cluster mode currently only supports a simple FIFO
scheduler across applications. However, to allow multiple concurrent
users, you can control the maximum number of resources each
application will use. By default, it will acquire all cores in the
cluster, which only makes sense if you just run one application at a
time. You can cap the number of cores by setting spark.cores.max ...
I am assuming you run all the workers on one server and try to simulate a cluster. The reason for this assumption is that if otherwise you could use one worker and master to run Standalone Spark cluster.
The executor cores are something completely different compared to the normal cores. To set the number of executors you will need YARN to be turned on as you earlier said. The executor cores are the number of Concurrent tasks as executor can run (when using hdfs it is advisable to keep this below 5) [1].
The number of cores you want to limit to make the workers run are the “CPU cores”. These are specified in the configuration of Spark 1.6.1 [2]. In Spark there is the option to set the amount of CPU cores when starting a slave [3]. This happens with -c CORES, --cores CORES . Which defines the total CPU cores to allow Spark applications to use on the machine (default: all available); only on worker.
The command to start Spark would be something like this:
./sbin/start-all.sh --cores 2
Hope this helps
In the configuration settings add this line to "./conf/spark-env.sh " this file.
export SPARK_MASTER_OPTS="-Dspark.deploy.defaultCores=1"
maximum cores now will limit to 1 for the master.
if multiple spark application is running then it will use only one core for the master. By then defining the amount of workers and give the workers the setting:
export SPARK_WORKER_OPTS="-Dspark.deploy.defaultCores=1"
Each worker has then one core as well. Remember this has to be set for every worker in the configuration settings.
We (an engineering team) are running an EMR cluster with YARN and Spark. What is typically happening is that when one user submits a heavy memory intensive job, it grabs all the YARN available memory and then all the subsequent users submitted jobs have to wait for that memory to clear (I know that autoscaling will solve this problem to a certain extent and we are looking into that, but we would like to avoid a single user occupying all the memory even when the cluster is autoscaled to it's full limits).
Is there a way to configure YARN such that any application (Spark or otherwise) may not occupy more than, say 75% of available memory?
Thanks
According to the documentation, you can manage the amount of memory allocated to an executor using the parameter: spark.executor.memory
I have pretty low configuration testing machine for my data pipelines developed in Spark. I will use only one AWS t2.large instance, which has only 2 CPUs and 8 GB of RAM.
I need to run 2 spark streaming jobs, as well as leave some memory and CPU power for occasionally testing batch jobs.
So I have master and one worker, which are on the same machine.
I have some general questions:
1) How many executors can run per one worker? I know that default is one, but does it make sense to change this?
2) Can one executor execute multiple applications, or one executor is dedicated only to one application?
3) Is a way to make this work, to set memory that application can use in configuration file, or when I create spark context?
Thank you
How many executors can run per one worker? I know that default is one, but does it make sense to change this?
It makes sense only in case you have enough resources. Say, on a machine with 24 GB and 12 cores it's possible to run 3 executors if you're sure that 8 GB is enough for one executor.
Can one executor execute multiple applications, or one executor is dedicated only to one application?
Nope, every application starts their own executors.
Is a way to make this work, to set memory that application can use in configuration file, or when I create spark context?
I'm not sure I understand the question, but there are 3 ways to provide configuration for applications
file spark-defaults.conf, but don't forget to turn on to read default properties, when you create new SparkConf instance.
providing system properties through -D, when you run the application or --conf if that's spark-submit or spark-shell. Although for memory options there are specific parameters like spark.executor.memory or spark.driver.memory and others to be used.
provides the same options through new SparkConf instance using its set methods.