Nalini Ganapati

Results 56 comments of Nalini Ganapati

@droazen, yes for luster, we should pass the `--genomicsdb-shared-posixfs-optimizations` especially with GenomicsDBImport.

Thanks @spatel-gfb. The consolidate argument is what will give you the query performance as the consolidation happens in memory when running `GenotypeGVCFs` or `SelectVariants` otherwise. Working on the fix for...

Modifying what I wrote earlier, got confused with another issue. I am not familiar with Lustre and Lustre configuration. Did the excessive file locking from Lustre(FUTEX_WAIT_PRIVATE?) go away with `--genomicsdb-shared-posixfs-optimizations`?...

> The test contains the argument --intervals chrX:1000000-5000000, but I'm not sure why that would be an issue. This is from `htslib::faidx_fetch_seq_into_buffer` because the reference for the test does not...

The only parallelism exposed by GenomicsDBImport is via `max-num-intervals-to-import-in-parallel` and that is at the granularity of threads in the same process. Not sure there is any parallelism in GenotypeGVCFs. The...

@shuaiwang2, can you please paste your entire command to `gatk GenotypGVCfs`? And the error section from running it? Can you also paste your command to `gatk GenomicsDBImport`? Thanks.

Almost looks like there is a buffer overrun somewhere. Most of our testing has been on `nfs` and have not encountered a tcache(thread local cache) issue. Is `gpfs` available as...

@danagibbon thanks for this pointer. What versions of gatk have you seen this error on?

Thanks @danagibbon, I may know what the issue is. `hdfs` support in GenomicsDB still relies on JVM/Java 11 and we had some workarounds with thread local caches from a while...

@danagibbon, here is the branch - https://github.com/broadinstitute/gatk/tree/ng_remove_hdfs_support. Can you build gatk from this branch and try it out please? If the problem still exists, can you attach the core dump...