Project

General

Profile

Crunch » History » Version 21

Ward Vandewege, 11/06/2017 07:50 PM

1 14 Ward Vandewege
h1. Crunch - container orchestration
2 1 Anonymous
3 17 Ward Vandewege
Arvados has a robust container orchestration system called 'Crunch', which executes CWL workflows while maintaining provenance and reproducibility.
4 1 Anonymous
5
h2. Design Goals
6
7
Notable design goals and features include:
8
9 20 Ward Vandewege
* Scale horizontally to support very large workloads
10 6 Tom Clegg
* Integrate with [[Keep]] and git repositories to maintain provenance
11
* Use off-the-shelf software tools in distributed computations
12
* Efficient over a wide range of problem sizes
13
* Maximum flexibility of programming language choice
14
* Maximum flexibility of execution environment
15 18 Ward Vandewege
* Low entry barrier for bioinformaticians
16 21 Ward Vandewege
* Isolate workloads by running all jobs inside "Docker":http://docker.org containers
17 1 Anonymous
18 15 Ward Vandewege
h2. Benefits of Crunch
19 1 Anonymous
20 15 Ward Vandewege
Although some of the workflow and provenance features in Arvados could theoretically be implemented using Hadoop MapReduce, there are distinct benefits to Crunch:
21 1 Anonymous
22 16 Ward Vandewege
* *Provenance and Reproducibility* - Like Keep, the Arvados distributed file system, Crunch is designed to automatically track the origin of result data. It can also efficiently reproduce complex workflows and comparing workflows to one another.
23 6 Tom Clegg
24 15 Ward Vandewege
* *Performance* - Most genomics problems are embarrassingly parallel and can benefit from horizontal scaling. In the cloud, Crunch can deliver cost-effective performance for genomics related analyses by automatically adjusting the available compute resources to the workload.
25 6 Tom Clegg
26 15 Ward Vandewege
* *Standardization* - "Common Workflow Language (CWL)":http://commonwl.org is the workflow description standard in bioinformatics. It is the native workflow language in Crunch.