For large objects, it will be more efficient to broadcast it. If your array
is small it won't really matter. How many centers do you have? Unless you
are finding that you have very large tasks (and Spark will print a warning
about this), it could be okay to just reference it directly.
On Wed,
The reason is that some operators get pipelined into a single stage.
rdd.map(XX).filter(YY) - this executes in a single stage since there is no
data movement needed in between these operations.
If you call toDeubgString on the final RDD it will give you some
information about the exact lineage.
Yep - that's correct. As an optimization we save the shuffle output and
re-use if if you execute a stage twice. So this can make A:B tests like
this a bit confusing.
- Patrick
On Friday, August 22, 2014, Nieyuan qiushuiwuh...@gmail.com wrote:
Because map-reduce tasks like join will save
might land in Spark 1.2,
is that being tracked in https://issues.apache.org/jira/browse/SPARK-1823
or is there another ticket I should be following?
Thanks!
Andrew
On Tue, Aug 5, 2014 at 3:39 PM, Patrick Wendell pwend...@gmail.com
wrote:
Hi Jens,
Within a partition things will spill
Hi All,
I want to invite users to submit to the Spark Powered By page. This page
is a great way for people to learn about Spark use cases. Since Spark
activity has increased a lot in the higher level libraries and people often
ask who uses each one, we'll include information about which
Yeah - each batch will produce a new RDD.
On Wed, Aug 27, 2014 at 3:33 PM, Soumitra Kumar
kumar.soumi...@gmail.com wrote:
Thanks.
Just to double check, rdd.id would be unique for a batch in a DStream?
On Wed, Aug 27, 2014 at 3:04 PM, Xiangrui Meng men...@gmail.com wrote:
You can use RDD
Changing this is not supported, it si immutable similar to other spark
configuration settings.
On Wed, Sep 3, 2014 at 8:13 PM, 牛兆捷 nzjem...@gmail.com wrote:
Dear all:
Spark uses memory to cache RDD and the memory size is specified by
spark.storage.memoryFraction.
One the Executor starts,
I would say that the first three are all used pretty heavily. Mesos
was the first one supported (long ago), the standalone is the
simplest and most popular today, and YARN is newer but growing a lot
in activity.
SIMR is not used as much... it was designed mostly for environments
where users had
I am happy to announce the availability of Spark 1.1.0! Spark 1.1.0 is
the second release on the API-compatible 1.X line. It is Spark's
largest release ever, with contributions from 171 developers!
This release brings operational and performance improvements in Spark
core including a new
[moving to user@]
This would typically be accomplished with a union() operation. You
can't mutate an RDD in-place, but you can create a new RDD with a
union() which is an inexpensive operator.
On Fri, Sep 12, 2014 at 5:28 AM, Archit Thakur
archit279tha...@gmail.com wrote:
Hi,
We have a use
Hey SK,
Yeah, the documented format is the same (we expect users to add the
jar at the end) but the old spark-submit had a bug where it would
actually accept inputs that did not match the documented format. Sorry
if this was difficult to find!
- Patrick
On Fri, Sep 12, 2014 at 1:50 PM, SK
Yeah that issue has been fixed by adding better docs, it just didn't make
it in time for the release:
https://github.com/apache/spark/blob/branch-1.1/make-distribution.sh#L54
On Thu, Sep 11, 2014 at 11:57 PM, Zhanfeng Huo huozhanf...@gmail.com
wrote:
resolved:
./make-distribution.sh --name
If each partition can fit in memory, you can do this using
mapPartitions and then building an inverse mapping within each
partition. You'd need to construct a hash map within each partition
yourself.
On Tue, Sep 16, 2014 at 4:27 PM, Akshat Aranya aara...@gmail.com wrote:
I have a use case where
guess I would need to
override mapPartitions() directly within my RDD. Right?
On Tue, Sep 16, 2014 at 4:57 PM, Patrick Wendell pwend...@gmail.com wrote:
If each partition can fit in memory, you can do this using
mapPartitions and then building an inverse mapping within each
partition. You'd need
Hey Grzegorz,
EMR is a service that is not maintained by the Spark community. So
this list isn't the right place to ask EMR questions.
- Patrick
On Thu, Sep 18, 2014 at 3:19 AM, Grzegorz Białek
grzegorz.bia...@codilime.com wrote:
Hi,
I would like to run Spark application on Amazon EMR. I have
IIRC - the random is seeded with the index, so it will always produce
the same result for the same index. Maybe I don't totally follow
though. Could you give a small example of how this might change the
RDD ordering in a way that you don't expect? In general repartition()
will not preserve the
Spark will need to connect both to the hive metastore and to all HDFS
nodes (NN and DN's). If that is all in place then it should work. In
this case it looks like maybe it can't connect to a datanode in HDFS
to get the raw data. Keep in mind that the performance might not be
very good if you are
Hey Ryan,
I've found that filing issues with the Scala/Typesafe JIRA is pretty
helpful if the issue can be fully reproduced, and even sometimes
helpful if it can't. You can file bugs here:
https://issues.scala-lang.org/secure/Dashboard.jspa
The Spark SQL code in particular is typically the
It shows the amount of memory used to store RDD blocks, which are created
when you run .cache()/.persist() on an RDD.
On Wed, Oct 22, 2014 at 10:07 PM, Haopu Wang hw...@qilinsoft.com wrote:
Hi, please take a look at the attached screen-shot. I wonders what's the
Memory Used column mean.
I
Hey Cheng,
Right now we aren't using stable API's to communicate with the Hive
Metastore. We didn't want to drop support for Hive 0.12 so right now
we are using a shim layer to support compiling for 0.12 and 0.13. This
is very costly to maintain.
If Hive has a stable meta-data API for talking to
Hey Jim,
There are some experimental (unstable) API's that support running jobs
which might short-circuit:
https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/SparkContext.scala#L1126
This can be used for doing online aggregations like you are
describing. And in one
The doc build appears to be broken in master. We'll get it patched up
before the release:
https://issues.apache.org/jira/browse/SPARK-4326
On Tue, Nov 11, 2014 at 10:50 AM, Alessandro Baretta
alexbare...@gmail.com wrote:
Nichols and Patrick,
Thanks for your help, but, no, it still does not
Hi There,
Because Akka versions are not binary compatible with one another, it
might not be possible to integrate Play with Spark 1.1.0.
- Patrick
On Tue, Nov 11, 2014 at 8:21 AM, Akshat Aranya aara...@gmail.com wrote:
Hi,
Sorry if this has been asked before; I didn't find a satisfactory
It looks like you are trying to directly import the toLocalIterator
function. You can't import functions, it should just appear as a
method of an existing RDD if you have one.
- Patrick
On Thu, Nov 13, 2014 at 10:21 PM, Deep Pradhan
pradhandeep1...@gmail.com wrote:
Hi,
I am using Spark 1.0.0
Hi Judy,
Are you somehow modifying Spark's classpath to include jars from
Hadoop and Hive that you have running on the machine? The issue seems
to be that you are somehow including a version of Hadoop that
references the original guava package. The Hadoop that is bundled in
the Spark jars should
/Preconitions.checkArgument:(ZLjava/lang/Object;)V
50: invokestatic #502// Method
org/spark-project/guava/common/base/Preconitions.checkArgument:(ZLjava/lang/Object;)V
On Wed, Nov 26, 2014 at 11:08 AM, Patrick Wendell pwend...@gmail.com wrote:
Hi Judy,
Are you somehow
I recently posted instructions on loading Spark in Intellij from scratch:
https://cwiki.apache.org/confluence/display/SPARK/Useful+Developer+Tools#UsefulDeveloperTools-BuildingSparkinIntelliJIDEA
You need to do a few extra steps for the YARN project to work.
Also, for questions like this that
to bypass the error.
This was caused by a local change, so no impact on the 1.2 release.
-Original Message-
From: Patrick Wendell [mailto:pwend...@gmail.com]
Sent: Wednesday, November 26, 2014 8:17 AM
To: Judy Nash
Cc: Denny Lee; Cheng Lian; u...@spark.incubator.apache.org
Subject: Re
Thanks for flagging this. I reverted the relevant YARN fix in Spark
1.2 release. We can try to debug this in master.
On Thu, Dec 4, 2014 at 9:51 PM, Jianshi Huang jianshi.hu...@gmail.com wrote:
I created a ticket for this:
https://issues.apache.org/jira/browse/SPARK-4757
Jianshi
On Fri,
Yeah the main way to do this would be to have your own static cache of
connections. These could be using an object in Scala or just a static
variable in Java (for instance a set of connections that you can
borrow from).
- Patrick
On Thu, Dec 4, 2014 at 5:26 PM, Tobias Pfeiffer t...@preferred.jp
Hey Manoj,
One proposal potentially of interest is the Spark Kernel project from
IBM - you should look for their. The interface in that project is more
of a remote REPL interface, i.e. you submit commands (as strings)
and get back results (as strings), but you don't have direct
programmatic
, 2014 at 12:57 AM, Patrick Wendell pwend...@gmail.com wrote:
The second choice is better. Once you call collect() you are pulling
all of the data onto a single node, you want to do most of the
processing in parallel on the cluster, which is what map() will do.
Ideally you'd try to summarize
I'm happy to announce the availability of Spark 1.2.0! Spark 1.2.0 is
the third release on the API-compatible 1.X line. It is Spark's
largest release ever, with contributions from 172 developers and more
than 1,000 commits!
This release brings operational and performance improvements in Spark
Is it sufficient to set spark.hadoop.validateOutputSpecs to false?
http://spark.apache.org/docs/latest/configuration.html
- Patrick
On Wed, Dec 24, 2014 at 10:52 PM, Shao, Saisai saisai.s...@intel.com wrote:
Hi,
We have such requirements to save RDD output to HDFS with saveAsTextFile
like
: Patrick Wendell [mailto:pwend...@gmail.com]
Sent: Thursday, December 25, 2014 3:22 PM
To: Shao, Saisai
Cc: user@spark.apache.org; d...@spark.apache.org
Subject: Re: Question on saveAsTextFile with overwrite option
Is it sufficient to set spark.hadoop.validateOutputSpecs to false?
http
What do you mean when you say the overhead of spark shuffles start to
accumulate? Could you elaborate more?
In newer versions of Spark shuffle data is cleaned up automatically
when an RDD goes out of scope. It is safe to remove shuffle data at
this point because the RDD can no longer be
Hey Eric,
I'm just curious - which specific features in 1.2 do you find most
help with usability? This is a theme we're focusing on for 1.3 as
well, so it's helpful to hear what makes a difference.
- Patrick
On Sun, Dec 28, 2014 at 1:36 AM, Eric Friedman
eric.d.fried...@gmail.com wrote:
Hi
Akhil,
Those are handled by ASF infrastructure, not anyone in the Spark
project. So this list is not the appropriate place to ask for help.
- Patrick
On Sat, Jan 17, 2015 at 12:56 AM, Akhil Das ak...@sigmoidanalytics.com wrote:
My mails to the mailing list are getting rejected, have opened a
It should appear in the page for any stage in which accumulators are updated.
On Wed, Jan 14, 2015 at 6:46 PM, Justin Yip yipjus...@prediction.io wrote:
Hello,
From accumulator documentation, it says that if the accumulator is named, it
will be displayed in the WebUI. However, I cannot find
The map will start with a capacity of 64, but will grow to accommodate
new data. Are you using the groupBy operator in Spark or are you using
Spark SQL's group by? This usually happens if you are grouping or
aggregating in a way that doesn't sufficiently condense the data
created from each input
Hey Jerry,
I think standalone mode will still add more features over time, but
the goal isn't really for it to become equivalent to what Mesos/YARN
are today. Or at least, I doubt Spark Standalone will ever attempt to
manage _other_ frameworks outside of Spark and become a general
purpose
I think there is a minor error here in that the first example needs a
tail after the seq:
df.map { row =
(row.getDouble(0), row.toSeq.tail.map(_.asInstanceOf[Double]))
}.toDataFrame(label, features)
On Wed, Feb 11, 2015 at 7:46 PM, Michael Armbrust
mich...@databricks.com wrote:
It sounds like
You may need to add the -Phadoop-2.4 profile. When building or release
packages for Hadoop 2.4 we use the following flags:
-Phadoop-2.4 -Phive -Phive-thriftserver -Pyarn
- Patrick
On Thu, Mar 5, 2015 at 12:47 PM, Kelly, Jonathan jonat...@amazon.com wrote:
I confirmed that this has nothing to
We don't support expressions or wildcards in that configuration. For
each application, the local directories need to be constant. If you
have users submitting different Spark applications, those can each set
spark.local.dirs.
- Patrick
On Wed, Mar 11, 2015 at 12:14 AM, Jianshi Huang
Hi All,
I'm happy to announce the availability of Spark 1.3.0! Spark 1.3.0 is
the fourth release on the API-compatible 1.X line. It is Spark's
largest release ever, with contributions from 172 developers and more
than 1,000 commits!
Visit the release notes [1] to read about the new features, or
Hey Jim,
Thanks for reporting this. Can you give a small end-to-end code
example that reproduces it? If so, we can definitely fix it.
- Patrick
On Tue, Mar 24, 2015 at 4:55 PM, Jim Carroll jimfcarr...@gmail.com wrote:
I have code that works under 1.2.1 but when I upgraded to 1.3.0 it fails to
I think we need to just update the docs, it is a bit unclear right
now. At the time, we made it worded fairly sternly because we really
wanted people to use --jars when we deprecated SPARK_CLASSPATH. But
there are other types of deployments where there is a legitimate need
to augment the classpath
is exactly the issue: on my master node UI
at the bottom I can see the list of Completed Drivers all with ERROR
state...
Thanks,
Oleg
-Original Message-
From: Patrick Wendell [mailto:pwend...@gmail.com]
Sent: Monday, February 23, 2015 12:59 AM
To: Oleg Shirokikh
Cc: user
Added - thanks! I trimmed it down a bit to fit our normal description length.
On Mon, Jan 5, 2015 at 8:24 AM, Thomas Stone tho...@prediction.io wrote:
Please can we add PredictionIO to
https://cwiki.apache.org/confluence/display/SPARK/Powered+By+Spark
PredictionIO
http://prediction.io/
I've added it, thanks!
On Fri, Feb 20, 2015 at 12:22 AM, Emre Sevinc emre.sev...@gmail.com wrote:
Hello,
Could you please add Big Industries to the Powered by Spark page at
https://cwiki.apache.org/confluence/display/SPARK/Powered+By+Spark ?
Company Name: Big Industries
URL:
If you invoke this, you will get at-least-once semantics on failure.
For instance, if a machine dies in the middle of executing the foreach
for a single partition, that will be re-executed on another machine.
It could even fully complete on one machine, but the machine dies
immediately before
The source code should match the Spark commit
4aaf48d46d13129f0f9bdafd771dd80fe568a7dc. Do you see any differences?
On Fri, Mar 27, 2015 at 11:28 AM, Manoj Samel manojsamelt...@gmail.com wrote:
While looking into a issue, I noticed that the source displayed on Github
site does not matches the
I think we have a version of mapPartitions that allows you to tell
Spark the partitioning is preserved:
https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/rdd/RDD.scala#L639
We could also add a map function that does same. Or you can just write
your map using an
Hey Yiannis,
If you just perform a count on each name, date pair... can it succeed?
If so, can you do a count and then order by to find the largest one?
I'm wondering if there is a single pathologically large group here that is
somehow causing OOM.
Also, to be clear, you are getting GC limit
Hey Jonathan,
Are you referring to disk space used for storing persisted RDD's? For
that, Spark does not bound the amount of data persisted to disk. It's
a similar story to how Spark's shuffle disk output works (and also
Hadoop and other frameworks make this assumption as well for their
shuffle
Hi Deepak - please direct this to the user@ list. This list is for
development of Spark itself.
On Sun, Apr 26, 2015 at 12:42 PM, Deepak Gopalakrishnan
dgk...@gmail.com wrote:
Hello All,
I'm trying to process a 3.5GB file on standalone mode using spark. I could
run my spark job succesfully on
Hi All,
I'm happy to announce the Spark 1.3.1 and 1.2.2 maintenance releases.
We recommend all users on the 1.3 and 1.2 Spark branches upgrade to
these releases, which contain several important bug fixes.
Download Spark 1.3.1 or 1.2.2:
http://spark.apache.org/downloads.html
Release notes:
...@gmail.com wrote:
Ok so it is the case that small shuffles can be done without hitting any
disk. Is this the same case for the aux shuffle service in yarn? Can that be
done without hitting disk?
On Wed, Jun 10, 2015 at 9:17 PM, Patrick Wendell pwend...@gmail.com wrote:
In many cases the shuffle
In many cases the shuffle will actually hit the OS buffer cache and
not ever touch spinning disk if it is a size that is less than memory
on the machine.
- Patrick
On Wed, Jun 10, 2015 at 5:06 PM, Corey Nolet cjno...@gmail.com wrote:
So with this... to help my understanding of Spark under the
Hi All,
I'm happy to announce the availability of Spark 1.4.0! Spark 1.4.0 is
the fifth release on the API-compatible 1.X line. It is Spark's
largest release ever, with contributions from 210 developers and more
than 1,000 commits!
A huge thanks go to all of the individuals and organizations
Hi All,
I'm happy to announce the Spark 1.4.1 maintenance release.
We recommend all users on the 1.4 branch upgrade to
this release, which contain several important bug fixes.
Download Spark 1.4.1 - http://spark.apache.org/downloads.html
Release notes -
Hi All,
Today, I'm happy to announce SparkHub
(http://sparkhub.databricks.com), a service for the Apache Spark
community to easily find the most relevant Spark resources on the web.
SparkHub is a curated list of Spark news, videos and talks, package
releases, upcoming events around the world,
101 - 162 of 162 matches
Mail list logo