Reputation: 1804
I am launching pyspark and I can supply the driver-memory
parameter via command line to specify the maximum memory usage by the driver. In Spark's online documentation, they often just use a value like 1g
or 2g
as examples, but I am not sure if it's legal to use 3300m
or 4500m
as the value.
I think this parameter is related to the jvm's Xmx
parameter which must be a multiple of 1024m which is more of a reason why I am confused.
Does spark's driver memory parameter properly handle something other than a number of gigabytes?
Upvotes: 1
Views: 499
Reputation: 2333
Yes, it works. Looking at the documentation and my previous experience, you can set the driver-memory in mbs also. Eg: 512m
See: http://spark.apache.org/docs/latest/configuration.html
Properties that specify a byte size should be configured with a unit of size. The following format is accepted:
1b (bytes)
1k or 1kb (kibibytes = 1024 bytes)
1m or 1mb (mebibytes = 1024 kibibytes)
1g or 1gb (gibibytes = 1024 mebibytes)
1t or 1tb (tebibytes = 1024 gibibytes)
1p or 1pb (pebibytes = 1024 tebibytes)
Upvotes: 4