Name: Matmul GPU Case 1 Cache-OFF Contact Person: cristian.tatu@bsc.es Access Level: public License Agreement: Apache2 Platform: COMPSs 3.3 Machine: Minotauro-MN4

Matmul running on the GPU without Cache. Launched using 32 GPUs (16 nodes). Performs C = A @ B Where A: shape (320, 56_900_000) block_size (10, 11_380_000)             B: shape (56_900_000, 10)   block_size (11_380_000, 10)             C: shape (320, 10)                block_size (10, 10) Total dataset size 291 ...

Name: K-Means GPU Cache OFF Contact Person: cristian.tatu@bsc.es Access Level: public License Agreement: Apache2 Platform: COMPSs Machine: Minotauro-MN4

K-Means running on GPUs. Launched using 32 GPUs (16 nodes). Parameters used: K=40 and 32 blocks of size (1_000_000, 1200). It creates a block for each GPU. Total dataset shape is (32_000_000, 1200). Version dislib-0.9

Average task execution time: 194 seconds

Name: K-Means GPU Cache ON Contact Person: cristian.tatu@bsc.es Access Level: public License Agreement: Apache2 Platform: COMPSs Machine: Minotauro-MN4

K-Means running on the GPU leveraging COMPSs GPU Cache for deserialization speedup. Launched using 32 GPUs (16 nodes). Parameters used: K=40 and 32 blocks of size (1_000_000, 1200). It creates a block for each GPU. Total dataset shape is (32_000_000, 1200). Version dislib-0.9

Average task execution time: 16 seconds

Name: Dislib Distributed Training - Cache ON Contact Person: cristian.tatu@bsc.es Access Level: public License Agreement: Apache2 Platform: COMPSs Machine: Minotauro-MN4

PyTorch distributed training of CNN on GPU and leveraging COMPSs GPU Cache for deserialization speedup. Launched using 32 GPUs (16 nodes). Dataset: Imagenet Version dislib-0.9 Version PyTorch 1.7.1+cu101

Average task execution time: 36 seconds

Name: Dislib Distributed Training - Cache OFF Contact Person: cristian.tatu@bsc.es Access Level: public License Agreement: Apache2 Platform: COMPSs Machine: Minotauro-MN4

PyTorch distributed training of CNN on GPU. Launched using 32 GPUs (16 nodes). Dataset: Imagenet Version dislib-0.9 Version PyTorch 1.7.1+cu101

Average task execution time: 84 seconds

HiC scaffolding pipeline

Snakemake pipeline for scaffolding of a genome using HiC reads using yahs.

Prerequisites

This pipeine has been tested using Snakemake v7.32.4 and requires conda for installation of required tools. To run the pipline use the command:

snakemake --use-conda --cores N

where N is number of cores to use. There are provided a set of configuration and running scripts for exectution on a slurm queueing system. After configuring the cluster.json file run:

./run_cluster ...

Purge dups

This snakemake pipeline is designed to be run using as input a contig-level genome and pacbio reads. This pipeline has been tested with snakemake v7.32.4. Raw long-read sequencing files and the input contig genome assembly must be given in the config.yaml file. To execute the workflow run:

snakemake --use-conda --cores N

Or configure the cluster.json and run using the ./run_cluster command

HiC contact map generation

Snakemake pipeline for the generation of .pretext and .mcool files for visualisation of HiC contact maps with the softwares PretextView and HiGlass, respectively.

Prerequisites

This pipeine has been tested using Snakemake v7.32.4 and requires conda for installation of required tools. To run the pipline use the command:

snakemake --use-conda

There are provided a set of configuration and running scripts for exectution on a slurm queueing system. After configuring ...

Post-genome assembly quality control workflow using Quast, BUSCO, Meryl, Merqury and Fasta Statistics. Updates November 2023. Inputs: reads as fastqsanger.gz (not fastq.gz), and assembly.fasta. New default settings for BUSCO: lineage = eukaryota; for Quast: lineage = eukaryotes, genome = large. Reports assembly stats into a table called metrics.tsv, including selected metrics from Fasta Stats, and read coverage; reports BUSCO versions and dependencies; and displays these tables in the workflow ...