I stumbled upon this article:
http://blog.cloudera.com/blog/2011/04/simple-moving-average-secondary-sort-and-mapreduce-part-3/
which mentions how to calculate moving average using Hadoop.
Please notice that all the records for a KEY should be sorted and then reduced. Now assume that the records for a particular KEY are spread across all the shards of Mongo cluster. In such case, would it be possible to calculate the moving average?
I understand that Mongo does the map reduce at each node. The prime requirement to solve this problem is to make sure all the emits for a map be reduced in a single reduce phase. If that's the case, then Mongo Map Reduce will never be able to solve such problems. Is there some basic misunderstanding?
Also, with billions of rows, and petabytes of data, why is it that Hadoop Reduce phase doesn't crash out of memory, since it has to deal with at least several TBs of mapped data.
Related
I'm seeking a way to control sharded collection migration thresholds in mongodb. These thresholds are described at https://docs.mongodb.com/manual/core/sharding-balancer-administration/#sharding-migration-thresholds
What I see in those values is that they have tuned the migration thresholds for roughly 10% of the chunk counts for small numbers of chunks (0-20: 2, 20-80: 4, 80+: 8). Above that, it's locked at 8 chunks: just 8 chunk counts being different between shard members will trigger a migration activity.
For our collections having high activity rates and large bodies of data, this causes balancing thrash - there is almost always a difference of 8 chunks, all the time. With high transaction rates on a sharded collection, there are a range of perfectly-acceptable causes of temporary imbalance (which I won't go into here). When we shut off the balancer, small temporary imbalances are often then corrected organically as activity across the cluster shifts. With the balancer turned on, by the time it finishes one migration, another (or many in parallel) triggers right away.
With the thresholds locked down like this, our larger collections thrash all the time - consuming IOPS and network bandwidth that we would really like to use in other ways. These tiny migrations have no practical benefit, either: if we're talking about a large collection, then 8 chunks can be a vanishingly small quantity of data relative to any real workload. So we're spending a lot of energy moving lots of small snippets around for zero effective benefit.
I would love to find a config file setting that - at a minimum - allows me to redefine those values. Even better would be to force a fractional policy, like 10% of the number of chunks in the collection. I don't see any controls of this type in the mongo documentation, but could be missing it.
Failing that, I'll have to spin up on the code and retool it myself to build from source, so I'm hoping someone has already solved this and I just can't see where to control it. Thanks in advance!
AFAIK, in case of Relational Database on MPP hardware, the key to performance is a correct data distribution. While Dimensional Modeling is about query flexibility, you don't even know how the data will be queried (shuffled) in future.
For example, you have MPP Data Warehouse (Greenplum, Redshift, Synapse Analytics). For example, in 1-2 years, you expect your fact table will grow up to 10 billion of rows and you'll have 15-30 dimension tables of 10s millions of rows. How the data should be distributed accross DW nodes? Is there any common techniques? Like shard fact table and replicate dimension tables. Or should I minimize node amount in MPP DW?
I can bring specific use case, but I believe that the question arise from my misunderstanding of how Dimensional Modeling could be paired with scaling out.
One technique I’ve seen applied with success in the past is: segment the fact table (e.g., by mod’ing the date key), and distribute all dimensions across all nodes. That way all joins can be done locally.
Note that even with large dimensions, their total size on disk should be a small fraction of the total needed for the fact table.
So, I understand that in general one should use coalesce() when:
the number of partitions decreases due to a filter or some other operation that may result in reducing the original dataset (RDD, DF). coalesce() is useful for running operations more efficiently after filtering down a large dataset.
I also understand that it is less expensive than repartition as it reduces shuffling by moving data only if necessary. My problem is how to define the parameter that coalesce takes (idealPartionionNo). I am working on a project which was passed to me from another engineer and he was using the below calculation to compute the value of that parameter.
// DEFINE OPTIMAL PARTITION NUMBER
implicit val NO_OF_EXECUTOR_INSTANCES = sc.getConf.getInt("spark.executor.instances", 5)
implicit val NO_OF_EXECUTOR_CORES = sc.getConf.getInt("spark.executor.cores", 2)
val idealPartionionNo = NO_OF_EXECUTOR_INSTANCES * NO_OF_EXECUTOR_CORES * REPARTITION_FACTOR
This is then used with a partitioner object:
val partitioner = new HashPartitioner(idealPartionionNo)
but also used with:
RDD.filter(x=>x._3<30).coalesce(idealPartionionNo)
Is this the right approach? What is the main idea behind the idealPartionionNo value computation? What is the REPARTITION_FACTOR? How do I generally work to define that?
Also, since YARN is responsible for identifying the available executors on the fly is there a way of getting that number (AVAILABLE_EXECUTOR_INSTANCES) on the fly and use that for computing idealPartionionNo (i.e. replace NO_OF_EXECUTOR_INSTANCES with AVAILABLE_EXECUTOR_INSTANCES)?
Ideally, some actual examples of the form:
Here 's a dataset (size);
Here's a number of transformations and possible reuses of an RDD/DF.
Here is where you should repartition/coalesce.
Assume you have n executors with m cores and a partition factor equal to k
then:
The ideal number of partitions would be ==> ???
Also, if you can refer me to a nice blog that explains these I would really appreciate it.
In practice optimal number of partitions depends more on the data you have, transformations you use and overall configuration than the available resources.
If the number of partitions is too low you'll experience long GC pauses, different types of memory issues, and lastly suboptimal resource utilization.
If the number of partitions is too high then maintenance cost can easily exceed processing cost. Moreover, if you use non-distributed reducing operations (like reduce in contrast to treeReduce), a large number of partitions results in a higher load on the driver.
You can find a number of rules which suggest oversubscribing partitions compared to the number of cores (factor 2 or 3 seems to be common) or keeping partitions at a certain size but this doesn't take into account your own code:
If you allocate a lot you can expect long GC pauses and it is probably better to go with smaller partitions.
If a certain piece of code is expensive then your shuffle cost can be amortized by a higher concurrency.
If you have a filter you can adjust the number of partitions based on a discriminative power of the predicate (you make different decisions if you expect to retain 5% of the data and 99% of the data).
In my opinion:
With one-off jobs keep higher number partitions to stay on the safe side (slower is better than failing).
With reusable jobs start with conservative configuration then execute - monitor - adjust configuration - repeat.
Don't try to use fixed number of partitions based on the number of executors or cores. First understand your data and code, then adjust configuration to reflect your understanding.
Usually, it is relatively easy to determine the amount of raw data per partition for which your cluster exhibits stable behavior (in my experience it is somewhere in the range of few hundred megabytes, depending on the format, data structure you use to load data, and configuration). This is the "magic number" you're looking for.
Some things you have to remember in general:
Number of partitions doesn't necessarily reflect
data distribution. Any operation that requires shuffle (*byKey, join, RDD.partitionBy, Dataset.repartition) can result in non-uniform data distribution. Always monitor your jobs for symptoms of a significant data skew.
Number of partitions in general is not constant. Any operation with multiple dependencies (union, coGroup, join) can affect the number of partitions.
Your question is a valid one, but Spark partitioning optimization depends entirely on the computation you're running. You need to have a good reason to repartition/coalesce; if you're just counting an RDD (even if it has a huge number of sparsely populated partitions), then any repartition/coalesce step is just going to slow you down.
Repartition vs coalesce
The difference between repartition(n) (which is the same as coalesce(n, shuffle = true) and coalesce(n, shuffle = false) has to do with execution model. The shuffle model takes each partition in the original RDD, randomly sends its data around to all executors, and results in an RDD with the new (smaller or greater) number of partitions. The no-shuffle model creates a new RDD which loads multiple partitions as one task.
Let's consider this computation:
sc.textFile("massive_file.txt")
.filter(sparseFilterFunction) // leaves only 0.1% of the lines
.coalesce(numPartitions, shuffle = shuffle)
If shuffle is true, then the text file / filter computations happen in a number of tasks given by the defaults in textFile, and the tiny filtered results are shuffled. If shuffle is false, then the number of total tasks is at most numPartitions.
If numPartitions is 1, then the difference is quite stark. The shuffle model will process and filter the data in parallel, then send the 0.1% of filtered results to one executor for downstream DAG operations. The no-shuffle model will process and filter the data all on one core from the beginning.
Steps to take
Consider your downstream operations. If you're just using this dataset once, then you probably don't need to repartition at all. If you are saving the filtered RDD for later use (to disk, for example), then consider the tradeoffs above. It takes experience to become familiar with these models and when one performs better, so try both out and see how they perform!
As others have answered, there is no formula which calculates what you ask for. That said, You can make an educated guess on the first part and then fine tune it over time.
The first step is to make sure you have enough partitions. If you have NO_OF_EXECUTOR_INSTANCES executors and NO_OF_EXECUTOR_CORES cores per executor then you can process NO_OF_EXECUTOR_INSTANCES*NO_OF_EXECUTOR_CORES partitions at the same time (each would go to a specific core of a specific instance).
That said this assumes everything is divided equally between the cores and everything takes exactly the same time to process. This is rarely the case. There is a good chance that some of them would be finished before others either because of locallity (e.g. the data needs to come from a different node) or simply because they are not balanced (e.g. if you have data partitioned by root domain then partitions including google would probably be quite big). This is where the REPARTITION_FACTOR comes into play. The idea is that we "overbook" each core and therefore if one finishes very quickly and one finishes slowly we have the option of dividing the tasks between them. A factor of 2-3 is generally a good idea.
Now lets take a look at the size of a single partition. Lets say your entire data is X MB in size and you have N partitions. Each partition would be on average X/N MBs. If N is large relative to X then you might have very small average partition size (e.g. a few KB). In this case it is usually a good idea to lower N because the overhead of managing each partition becomes too high. On the other hand if the size is very large (e.g. a few GB) then you need to hold a lot of data at the same time which would cause issues such as garbage collection, high memory usage etc.
The optimal size is a good question but generally people seem to prefer partitions of 100-1000MB but in truth tens of MB probably would also be good.
Another thing you should note is when you do the calculation how your partitions change. For example, lets say you start with 1000 partitions of 100MB each but then filter the data so each partition becomes 1K then you should probably coalesce. Similar issues can happen when you do a groupby or join. In such cases both the size of the partition and the number of partitions change and might reach an undesirable size.
I am building a recommendation system for retail purposes. I use python and Spark.
I am trying to subtract all user product combinations of my predictions which also occur in the ratings (so I only predict the values of products users never bought before).
Those 2 RDD's are pretty large and are giving me memory issues on 28gb per worker node (3 nodes) when I do
filter_predictions = predictions.subtractByKey(user_boughtproduct)
When I read the documentation of Spark subtractByKey is optimal when using 1 large and 1 small rdd.
I cannot make the user_boughtproduct smaller (unless I loop it), but I could make.
filter_predictions = predictions.join(user_nonBoughtProduct)
Any thoughts on which of them is faster or best practice? Or another cleaner solution.
subtractByKey pushes filters after co-grouping and doesn't have to touch right values so it should be slightly more efficient than using outer join an filter after flattening.
If you use Spark 2.0+ and records can be encoded using Dataset encoders you can consider leftanti join but depending on the rest of your code cost of moving the data can negate benefits of an optimized execution.
Finally if you can accept potential data loss then building Bloom filter on the right RDD and use it to filter the left one can give really good result without shuffling.
I have to check performance of various clustering algos using different performance operators in rapidminer. For that I want to know the following things:
what does cluster number index value shows which is output of cluster count performance operator?
what does small and large value of avg within cluster distance and avg. within centroid distance mean in terms of good and bad clustering?
I also want to check other indexes value like Dunn index,Jaccard index, Fowlkes–Mallows for various clustering algos. but rapidminer don't have any operator for this, what to do for that. I don't have experience with R.
I have copied part of the answer I gave on the Rapid-I forum
The cluster number index is the count of clusters - pointless you might say but when used with DBSCAN, it can be quite interesting http://rapidminernotes.blogspot.co.uk/2010/12/counting-clusters.html
The avg within cluster and centroid distances are hard to interpret - one thing to search for is "elbow criterion" in this context. As the number of clusters varies, note how the validity measure changes and look for an "elbow" that marks the point where the natural progression of the measure dominates the structure.
R has many validity measures and it's worth investing some time because you can always call the R process from RapidMiner which makes it easier to work out what is going on.