Many SAS procedures have NTHREADS option that allows one to fine-tune computer resource allocation and can have dramatic impact on processing times.  As far as I know GLIMMIX is not one of them.   In running bootstrap simulations, I am running the same GLIMMIX model 500 times and, with approximately 4 minutes per bootstrap Replication, processing time ends up 2,000 minutes or well over one day.   Given that computer resource utilization is minimal during the whole run, I tried to split the 500 bootstrap samples into 10 parallel runs with 50 Replications in each.  To my surprise, when running exactly the same GLIMMIX model with 10 parallel processes, the per-replication processing times increased from 4 minutes to about 18 minutes.  Computer resource utilization did not show anything remotely close to reaching the limits of the powerful Windows machine I am using (64 core CPU, 128GB or RAM, NvME Hard Discs).   In summary, by parallelizing the process, I cut the total run time in half but that's not anywhere near close to 10 times corresponding to 10 parallel runs of the models.   Can anyone help me understand why this is?   More importantly, how do I optimize the workflow.  That is, would 50 parallel runs be faster than 10 or would 5 be better?  I can, of course, just try but the process is very time-consuming.   Thanks for your thoughts. ... View more

I am trying to read a JSON file with 30 arrays.  Here's the code:   filename in "Source.json"; filename map 'WorkingMap.Map'; libname in json map=Map automap=reuse; libname j json "Source.json" map=Map;   By default this syntax creates a separate table for each array.  For the example below, the code above creates Proc1-Proc4 columns in a table PROC with Ordinal_Root link to the main ROOT dataset.   "proc": [ 2800, 2980, 4090, 4170 ], I have to merge multiple files into one for my purposes and also aggregate multiple columns into one.  On top of that, I am processing dozens of JSONS that need to be stacked to there's a ton of re-merging going on. Is there an option or a mapping method/trick to read arrays as simple text strings?  For the PROC example above, rather than a dedicated separate table with multiple columns for each value of the array, I would like a single column that contains them all in one cell just like in the JSON: [ 2800, 2980, 4090, 4170 ] Need help!!!  ... View more

Need help to import a JSON schema into an empty SAS dataset.  I have the schema file but no data yet.  Tried this syntax but get an ERROR: Invalid Physical Name, ERROR: in the LIBNAMD statement.   filename map "C:\JSON_Schema.json"; libname JSONlib json map=map; Direct read of the JSON schema as a data file creates a separate dataset for each field in the schema: libname JSONdata json 'c:\JSON_Schema.json'; Is getting a preview of the eventual SAS dataset file structure even possible? ... View more

I am running multiple I/O heavy programs in SAS batch mode.  I have 10-core CPU with HT and two storage disks.  Is there a way I can direct one job to be done on HDisk1 and the second on HDisk2?  Any direction is appreciated. ... View more

Need to run multiple parallel MCMC processes.  Models currently take between 8 and 21 hours to complete one-by-one.  Looking for any advice in terms of software OPTIONS settings such as THREADS, MEMSIZE, etc.   Also hardware: what should I prioritize -- get more or faster RAM, single thread CPU speed, more cores/threads, more virtual memory, faster I/O as in would upgrading from SATA to M2 interface.   Any help or references much appreciated. ... View more