[Rd] Reduce memory peak when serializing to raw vectors

Martinez de Salinas, Jorge jorge.martinez-de-salinas at hp.com
Tue Mar 17 23:09:21 CET 2015


I've been doing some tests using serialize() to a raw vector:

	df <- data.frame(runif(50e6,1,10))
	ser <- serialize(df,NULL)

In this example the data frame and the serialized raw vector occupy ~400MB each, for a total of ~800M. However the memory peak during serialize() is ~1.2GB:

	$ cat /proc/15155/status |grep Vm
	VmHWM:	 1207792 kB
	VmRSS:	  817272 kB

We work with very large data frames and in many cases this is killing R with an "out of memory" error.

This is the relevant code in R 3.1.3 in src/main/serialize.c:2494

	InitMemOutPStream(&out, &mbs, type, version, hook, fun);
	R_Serialize(object, &out);
	val =  CloseMemOutPStream(&out);

The serialized object is being stored in a buffer pointed by out.data. Then in CloseMemOutPStream() R copies the whole buffer to a newly allocated SEXP object (the raw vector that stores the final result):

	PROTECT(val = allocVector(RAWSXP, mb->count));
	memcpy(RAW(val), mb->buf, mb->count);

Before calling free_mem_buffer() the process is using ~1.2GB (the original data frame + the serialization buffer + final serialized raw vector). 

One possible solution would be to allocate a buffer for the final raw vector and store the serialization result directly into that buffer. This would bring the memory peak down from ~1.2GB to ~800MB.


More information about the R-devel mailing list