From 0ed819d781e83cdfeb3faa491132b8eea48613ca Mon Sep 17 00:00:00 2001 From: David K Jackson Date: Mon, 11 Sep 2023 15:40:09 +0100 Subject: [PATCH] typo --- README.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index 7ddca705..b7f8cdd9 100644 --- a/README.md +++ b/README.md @@ -11,22 +11,22 @@ There are two main pipelines: and the daemons which automatically start these pipelines. -Processing is performed as a appropriate for the entire run, for each lane in -the sequencing flowcell, or each tagged library (within a pool on the flowcell). +Processing is performed as appropriate for the entire run, for each lane in the +sequencing flowcell, or each tagged library (within a pool on the flowcell). ## Batch Processing and Dependency Tracking with LSF or wr -With this system, all of a pipeline's job for its steps are submitted for +With this system, all of a pipeline's jobs for its steps are submitted for execution to the LSF, or wr, batch/job processing system as the pipeline is -initialised. As such, a _submitted_ pipeline does not have a orchestration +initialised. As such, a _submitted_ pipeline does not have an orchestration script or daemon running: managing the runtime dependencies of jobs within an instance of a pipeline is delegated to the batch/job processing system. How is this done? The job representing the start point of a graph is submitted to LSF, or wr, in a suspended state and is resumed once all other jobs have been submitted thus ensuring that the execution starts only if all steps are -successfully submitted to LSF, or wr. If an error occurs at any point, all -submitted jobs, apart from the start job, are killed. +successfully submitted to LSF, or wr. If an error occurs at any point during job +submission, all submitted jobs, apart from the start job, are killed. ## Pipeline Creation