Sequential access I/O to unit open for direct access - fortran

I am trying to write a file with direct access as follows:
open(unit=11,access='DIRECT',recl=2*mumps_par%N,file=namr,form='unformatted',convert='big_endian')
write(11,rec=1) (dble(vp(i)),i=1,mumps_par%N)
close(11)
The 2* for the record length is because I am using ifort.
When I run this I get the error:
forrtl: severe (259): sequential-access I/O to unit open for direct access, unit 11 path/to/file
How can I write to this file?

Related

Query on boost interprocess::file_lock on NFS

We have an application which user can run to generate some data at user specified path. This unique output data is generated with respect to one unique input data-set - this input data is provided by the user.
When we initially developed the application, we never anticipated that number of unique input data-set will be large (due to nature of application). Our expectation was number of unique input data-set could be of order of 10 where as one user has this as 1000. So, that particular user started 1000 jobs of our application on grid and all writing data to same path. Note - these 1000 jobs are not fired from our application and rather he spawned 1000 processes of our application on different machines.
Now this lead to some collision and data loss.
To guard against it, I am planning to synchronization using boost::interprocess. This is what I am planning:
// usual processing of input data ...
boost::filesystem::path reportLockFilePath(boost::filesystem::system_complete(userDir));
rerportLockFilePath.append("report.lock");
// if lock file does not exist, create one
if (!boost::filesystem::exists(reportLockFilePath) {
boost::interprocess::named_mutex reportLockMutex(boost::interprocess::open_or_create, "report_mutex");
boost::interprocess::scoped_lock< boost::interprocess::named_mutex > lock(reportLockMutex);
std::ofstream lockStrm(reportLockFilePath.string().c_str());
lockStrm << "## report lock file ##" << std::endl;
lockStrm.flush();
}
boost::interprocess::file_lock reportFileLock(reportLockFilePath.string().c_str());
boost::interprocess::scoped_lock< boost::interprocess::file_lock > lock(reportFileLock);
// usual reporting code that we already have ...
Now, questions are -
If this is correct synchronization for the problem at hand
If this synchronization scheme will work, when jobs are on different machines and path is on NFS
If on NFS etc., this is not going to work, what are the C++ alternatives? I prefer to avoid lower level C functions to avoid race condition due to lock being held when one instance of execution crashes etc.
I just removed the named mutex part (as that was causing problem on few machines due to permission issue - probably related to umask issue discussed in this context in some other post) and replaced with
std::ofstream lockStrm(reportLockFilePath.string().c_str(), std::ios_base::app);
And it worked at least in our internal testing.

Streaming MutationGroups into Spanner

I'm trying to stream MutationGroups into spanner with SpannerIO.
The goal is to write new MuationGroups every 10 seconds, as we will use spanner to query near-time KPI's.
When I don't use any windows, I get the following error:
Exception in thread "main" java.lang.IllegalStateException: GroupByKey cannot be applied to non-bounded PCollection in the GlobalWindow without a trigger. Use a Window.into or Window.triggering transform prior to GroupByKey.
at org.apache.beam.sdk.transforms.GroupByKey.applicableTo(GroupByKey.java:173)
at org.apache.beam.sdk.transforms.GroupByKey.expand(GroupByKey.java:204)
at org.apache.beam.sdk.transforms.GroupByKey.expand(GroupByKey.java:120)
at org.apache.beam.sdk.Pipeline.applyInternal(Pipeline.java:537)
at org.apache.beam.sdk.Pipeline.applyTransform(Pipeline.java:472)
at org.apache.beam.sdk.values.PCollection.apply(PCollection.java:286)
at org.apache.beam.sdk.transforms.Combine$PerKey.expand(Combine.java:1585)
at org.apache.beam.sdk.transforms.Combine$PerKey.expand(Combine.java:1470)
at org.apache.beam.sdk.Pipeline.applyInternal(Pipeline.java:537)
at org.apache.beam.sdk.Pipeline.applyTransform(Pipeline.java:491)
at org.apache.beam.sdk.values.PCollection.apply(PCollection.java:299)
at org.apache.beam.sdk.io.gcp.spanner.SpannerIO$WriteGrouped.expand(SpannerIO.java:868)
at org.apache.beam.sdk.io.gcp.spanner.SpannerIO$WriteGrouped.expand(SpannerIO.java:823)
at org.apache.beam.sdk.Pipeline.applyInternal(Pipeline.java:537)
at org.apache.beam.sdk.Pipeline.applyTransform(Pipeline.java:472)
at org.apache.beam.sdk.values.PCollection.apply(PCollection.java:286)
at quantum.base.transform.entity.spanner.SpannerProtoWrite.expand(SpannerProtoWrite.java:52)
at quantum.base.transform.entity.spanner.SpannerProtoWrite.expand(SpannerProtoWrite.java:20)
at org.apache.beam.sdk.Pipeline.applyInternal(Pipeline.java:537)
at org.apache.beam.sdk.Pipeline.applyTransform(Pipeline.java:491)
at org.apache.beam.sdk.values.PCollection.apply(PCollection.java:299)
at quantum.entitybuilder.pipeline.EntityBuilderPipeline$Write$SpannerWrite.expand(EntityBuilderPipeline.java:388)
at quantum.entitybuilder.pipeline.EntityBuilderPipeline$Write$SpannerWrite.expand(EntityBuilderPipeline.java:372)
at org.apache.beam.sdk.Pipeline.applyInternal(Pipeline.java:537)
at org.apache.beam.sdk.Pipeline.applyTransform(Pipeline.java:491)
at org.apache.beam.sdk.values.PCollection.apply(PCollection.java:299)
at quantum.entitybuilder.pipeline.EntityBuilderPipeline.main(EntityBuilderPipeline.java:122)
:entityBuilder FAILED
Because of the error above I assume the input collection needs to be windowed and triggered, as SpannerIO uses a GroupByKey (this is also what I need for my use case):
...
.apply("1-minute windows", Window.<MutationGroup>into(FixedWindows.of(Duration.standardMinutes(1)))
.triggering(Repeatedly.forever(AfterProcessingTime
.pastFirstElementInPane()
.plusDelayOf(Duration.standardSeconds(10))
).orFinally(AfterWatermark.pastEndOfWindow()))
.discardingFiredPanes()
.withAllowedLateness(Duration.ZERO))
.apply(SpannerIO.write()
.withProjectId(entityConfig.getSpannerProject())
.withInstanceId(entityConfig.getSpannerInstance())
.withDatabaseId(entityConfig.getSpannerDb())
.grouped());
When I do this, I get the following exceptions during runtime:
java.lang.IllegalArgumentException: Attempted to get side input window for GlobalWindow from non-global WindowFn
org.apache.beam.sdk.transforms.windowing.PartitioningWindowFn$1.getSideInputWindow(PartitioningWindowFn.java:49)
com.google.cloud.dataflow.worker.StreamingModeExecutionContext$StepContext.issueSideInputFetch(StreamingModeExecutionContext.java:631)
com.google.cloud.dataflow.worker.StreamingModeExecutionContext$UserStepContext.issueSideInputFetch(StreamingModeExecutionContext.java:683)
com.google.cloud.dataflow.worker.StreamingSideInputFetcher.storeIfBlocked(StreamingSideInputFetcher.java:182)
com.google.cloud.dataflow.worker.StreamingSideInputDoFnRunner.processElement(StreamingSideInputDoFnRunner.java:71)
com.google.cloud.dataflow.worker.SimpleParDoFn.processElement(SimpleParDoFn.java:323)
com.google.cloud.dataflow.worker.util.common.worker.ParDoOperation.process(ParDoOperation.java:43)
com.google.cloud.dataflow.worker.util.common.worker.OutputReceiver.process(OutputReceiver.java:48)
com.google.cloud.dataflow.worker.SimpleParDoFn$1.output(SimpleParDoFn.java:271)
org.apache.beam.runners.core.SimpleDoFnRunner.outputWindowedValue(SimpleDoFnRunner.java:219)
org.apache.beam.runners.core.SimpleDoFnRunner.access$700(SimpleDoFnRunner.java:69)
org.apache.beam.runners.core.SimpleDoFnRunner$DoFnProcessContext.output(SimpleDoFnRunner.java:517)
org.apache.beam.runners.core.SimpleDoFnRunner$DoFnProcessContext.output(SimpleDoFnRunner.java:505)
org.apache.beam.sdk.values.ValueWithRecordId$StripIdsDoFn.processElement(ValueWithRecordId.java:145)
After investigating further it appears to be due to the .apply(Wait.on(input)) in SpannerIO: It has a global side input which does not seem to work with my fixed windows, as the docs of Wait.java state:
If signal is globally windowed, main input must also be. This typically would be useful
* only in a batch pipeline, because the global window of an infinite PCollection never
* closes, so the wait signal will never be ready.
As a temporary workaround I tried the following:
add a GlobalWindow with triggers instead of fixed windows:
.apply("globalwindow", Window.<MutationGroup>into(new GlobalWindows())
.triggering(Repeatedly.forever(AfterProcessingTime
.pastFirstElementInPane()
.plusDelayOf(Duration.standardSeconds(10))
).orFinally(AfterWatermark.pastEndOfWindow()))
.discardingFiredPanes()
.withAllowedLateness(Duration.ZERO))
This results in writes to spanner only when I drain my pipeline. I have the impression the Wait.on() signal is only triggered when the Global windows closes, and doesn't work with triggers.
Disable the .apply(Wait.on(input)) in SpannerIO:
This results in the pipeline getting stuck on the view creation which
is described in this SO post:
SpannerIO Dataflow 2.3.0 stuck in CreateDataflowView.
When I check the worker logs for clues, I do get the following warnings:
logger: "org.apache.beam.sdk.coders.SerializableCoder"
message: "Can't verify serialized elements of type SpannerSchema have well defined equals method. This may produce incorrect results on some PipelineRunner
logger: "org.apache.beam.sdk.coders.SerializableCoder"
message: "Can't verify serialized elements of type BoundedSource have well defined equals method. This may produce incorrect results on some PipelineRunner"
Note that everything works with the DirectRunner and that I'm trying to use the DataflowRunner.
Does anyone have any other suggestions for things I can try to get this running? I can hardly imagine that I'm the only one trying to stream MutationGroups into spanner.
Thanks in advance!
Currently, SpannerIO connector is not supported with Beam Streaming. Please follow this Pull Request which adds streaming support for spanner IO connector.

tensorflow places softmax op on cpu instead of gpu

I have a tensorflow model with multiple inputs and several layers, and a final softmax layer. The model is trained in Python (using the Keras framework), then saved and inference is done using a C++ program that facilitates a CMake build of TensorFlow (following basically those instructions: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/cmake).
In python (tensorflow-gpu) all ops use the GPU (using log_device_placement):
out/MatMul: (MatMul): /job:localhost/replica:0/task:0/gpu:0
2017-12-04 14:07:38.005837: I C:\tf_jenkins\home\workspace\rel-in\M\windows-gpu\PY\35\tensorflow\core\common_runtime\simple_placer.cc:872] out/MatMul: (MatMul)/job:localhost/replica:0/task:0/gpu:0
out/BiasAdd: (BiasAdd): /job:localhost/replica:0/task:0/gpu:0
2017-12-04 14:07:38.006201: I C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\35\tensorflow\core\common_runtime\simple_placer.cc:872]
out/BiasAdd: (BiasAdd)/job:localhost/replica:0/task:0/gpu:0
out/Softmax: (Softmax): /job:localhost/replica:0/task:0/gpu:0
2017-12-04 14:07:38.006535: I C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\35\tensorflow\core\common_runtime\simple_placer.cc:872] out/Softmax: (Softmax)/job:localhost/replica:0/task:0/gpu:0
To save the graph, the freeze_graph script is used (the script producing the log above loads again the freezed graph in .pb format).
When I use the C++ program and load the freezed graph (following closely the LoadGraph() function in https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/label_image/main.cc - ReadBinaryProto() and session->Create()), and log again the device placements, I find that the Softmax is placed on CPU (all others ops are on GPU):
dense_6/MatMul: (MatMul): /job:localhost/replica:0/task:0/device:GPU:0
dense_6/BiasAdd: (BiasAdd): /job:localhost/replica:0/task:0/device:GPU:0
dense_6/Relu: (Relu): /job:localhost/replica:0/task:0/device:GPU:0
out/MatMul: (MatMul): /job:localhost/replica:0/task:0/device:GPU:0
out/BiasAdd: (BiasAdd): /job:localhost/replica:0/task:0/device:GPU:0
out/Softmax: (Softmax): /job:localhost/replica:0/task:0/device:CPU:0
This placement is also confirmed by high CPU/low GPU utilization, and also apparent from profiling the application. The data type of the out layer is float32 (out/Softmax -> (<tf.Tensor 'out/Softmax:0' shape=(?, 1418) dtype=float32>,)).
Further investigation revealed:
Creating the softmax-op in C++ and placing it on GPU explicitly throws this error message:
Cannot assign a device for operation 'tsoftmax': Could not satisfy explicit device specification '/device:GPU:0' because no supported kernel for GPU devices is available.
A call to tensorflow::LogAllRegisteredKernels() showed also that Softmax is only available for CPU!
The build directory contains many files related to "softmax" (e.g. `tf_core_gpu_kernels_generated_softmax_op_gpu.cu.cc.obj.Release.cmake). Don't know how to check every compilation step, though.
when I look into the "tf_core_gpu_kernels.lib" (one can open a .lib with 7Z ;)), there are files like "tf_core_gpu_kernels_generated_softmax_op_gpu.cu.cc.lib" - so I believe there is nothing wrong with compiling the kernels itself
But: inspecting the "tensorflow.dll" (Dependency Walker) shows that only CPU kernels for Softmax are included (there are functions like const tensorflow::SoftmaxOp<struct Eigen::ThreadPoolDevice,double>, but no functions with GPU such as const tensorflow::SoftplusGradOp<struct Eigen::GpuDevice,float>).
Setup: Tensorflow 1.3.0, Windows 10, GPU: NVidia GTX 1070 (8GB RAM, memory utilization also very low).
I found a workaround - the workaround is to include the tf_core_gpu_kernels.lib in some of the steps (create_def_file.py). More details here: GitHub Issue 15254

Retrieve data from file written in FORTRAN during program run

I am trying to write a series of values for time (real values) into a dat file in FORTRAN. This is a part of an MPI code and the code runs for a long time. So I would like to extract data at every time step and print it into a file and read the file any time during the execution of the program. Currently, the problem I am facing is, the values of time are not written into the file until the program ends. I have put the open statement before the do loop and the close statement after the end of do loop.
The parts of my code look like:
open(unit=57,file='inst.dat')
do loop starts
.
.
.
write(57,*) time
.
.
.
end do
close(57)
try call flush(unit). Check your compiler docs as this is i think an extension.
You mention MPI: For parallel codes I think you need to give each thread its own file/unit,
or take other measures to avoid conflicts.
From Gfortran manual:
Beginning with the Fortran 2003 standard, there is a FLUSH statement that should be preferred over the FLUSH intrinsic.
The FLUSH intrinsic and the Fortran 2003 FLUSH statement have identical effect: they flush the runtime library's I/O buffer so that the data becomes visible to other processes. This does not guarantee that the data is committed to disk.
On POSIX systems, you can request that all data is transferred to the storage device by calling the fsync function, with the POSIX file descriptor of the I/O unit as argument (retrieved with GNU intrinsic FNUM). The following example shows how:
! Declare the interface for POSIX fsync function
interface
function fsync (fd) bind(c,name="fsync")
use iso_c_binding, only: c_int
integer(c_int), value :: fd
integer(c_int) :: fsync
end function fsync
end interface
! Variable declaration
integer :: ret
! Opening unit 10
open (10,file="foo")
! ...
! Perform I/O on unit 10
! ...
! Flush and sync
flush(10)
ret = fsync(fnum(10))
! Handle possible error
if (ret /= 0) stop "Error calling FSYNC"
How about closing the file after every time step (assuming a reasonable amount of time elapses between time steps)?
do loop starts
.
.
!Note: an if statement should wrap the following so that it is
!only called by one processor.
open(unit=57,file='inst.dat')
write(57,*) time
close(57)
.
.
end do
Alternatively if the time between time steps is short, writing the data after blocks of 10, 100, ... iterations may be more efficient.

streaming fast changing data between servers

I have server 1 which is generating a large amount of data, e.g there are files that are constantly being updated, on the time scale of milliseconds.
I would like to get these files onto another server, using C++ or standard Linux methods.
Currently, I have been doing this by compressing the files every second and using scp to transfer them, and unpacking on the other server.
However, the latency of this is very high and I can't break sub 1 second times with this.
Can anybody suggest the methods I can use to move the data with lower latency?
Just an idea, i don't know if it'll work for your particular situation:
Write two programs. One that runs on the server on which you files are being updated, and it monitors the changes with inotify. The other program runs on the second server and maintains a tcp connection with the first one. Whenever the first program detects a change, it sends the changed part of the file to the second program, which can apply the change to its own copy of the file.
Also, if the first server is not actually generating data for those file, but is reading it from the newtwork, it would be a good idea to just multicast the stream of data to both servers.
On Linux you can use DRBD and a cluster file system like GFS2 to have some partition transparently replicated between the two servers.
Another option, would be to use rsync.
A Perl script that uses inotify to detect changes on the filesystem and rsync over SSH to resynchronize the remote copies:
#!/usr/bin/perl
use strict;
use warnings;
use Net::OpenSSH;
use Linux::Inotify2;
use Time::HiRes qw(sleep);
my $usage = "Usage:\n $0 local_dir [user\#]host remote_dir\n\n";
#ARGV == 3 or die $usage;
my ($local, $host, $remote) = #ARGV;
-d $local or die $usage;
my $ssh = Net::OpenSSH->new($host);
$ssh->error and die "unable to connect to remote host: " . $ssh->error;
my $inotify = Linux::Inotify2->new;
$inotify->watch ($local, IN_MODIFY|IN_MOVED_TO);
$ssh->rsync_put({verbose => 1, glob => 1}, "$local/*", $remote);
while (1) {
my #events = $inotify->read or die "read error: $!";
my %changed;
$changed{"$local/$_->{name}"} = 1 for #events;
$ssh->rsync_put({verbose => 1}, keys %changed, $remote);
sleep 0.1;
}
update: #user788171: in response to your question:
It may be or may be not, there are too many unknowns:
how many files do actually change
their size
how large are the changes relative to full file size
network bandwidth and latency
available CPU power in both sides
etc.
But trying it is cheap so I suggest you do it, if it is not enough, then you can try to identify bottlenecks and try to eliminate them.
For instance, rsync is a talky protocol, very sensible to network latency, so, if your files are small, scp may produce better results. Or you could keep a local copy of the last version transmitted locally for every file and send just deltas. If CPU is the bottleneck, rewrite it in C++, eliminate SSH, etc.
And if anyway, that approach turns to be a dead end, then, you can still...
Do it at the OS level, using DRDB or some other transparent replication mechanism. You can even try to implement it yourself using FUSE.
Modify your main application to write a log of the changes that can be streamed easyly to the other side.