Computational Requirements
Warning
Details on this page not complete
Requirements
The following indicates the necessary computational requirements to learn this course. Instructors and self-learners can choose to learn using only the sample data provided for this course or with a new full genome. The requirements for each will be different.
For the Course Sample Data Only
Error
ADD DETAILS HERE
For a Full Genome
Error
ADD DETAILS HERE
Options
The following describes several ways to access the computational resources necessary for this course. Instructors are encouraged to choose the method that best suits the needs of their students and the current compute infrastructure you have available.
The course material will provide instructions throughout for usage of each of the following three options:
Praxis AI
This course is offered via a cloud-based service called Praxis AI. Instructors and students who do not have ecomputational resources onsite for students. Praxis AI integrates educational material with computational resources, conference and collaboration services. The course requires a student fee, equivalent for a text book, for access to these services and provides predictive costs for offering the course. Once the course is completed students can continue to access course content via this site as a free reference.
Advantages
Compute infrastructure is already available.
You can upscale the amount of computing power as needed.
Course material and compute are available within the same systems.
Praxis AI is a great solution for instructors that do not have ready access to compute.
Limitations
A fee is required, but at a rate similar to most textbooks.
The base fee should cover the costs for teaching this course with the sample data but may not be sufficient for an entire genome.
Workstations or Compute Cluster
For instructors and students with access to a high-performance workstations or an institutional compute cluster, we have created Docker images that contain all of the necessary software required for this class. The Computational infrastructure must have a Docker service or a Singularity service installed to use these images.
Note
A docker image is a minimal self-contained UNIX/Linux based-operating system pre-configured with a desired set of software.
For learners that have access to an institutional compute cluster you may need to request installation of Singularity if it is not already present.
Note
Instructors: while there are multiple job schedulers available for compute clusters, this course currently only provides instructions for the SLURM scheduler. If your cluster does not use SLURM you may have to adapt the course material to use the scheduler that your cluster provides.
Students: If you are self-training and you have access to a cluster but it does not use SLURM you will need to do a bit of searching to find the corresponding commands for submitting and managing jobs on your cluster.
Advantages
The software needed for the course is ready to go. No installation is required.
You can use your own compute infrastructure.
Students receive training on compute infrastructure they may be using for other purposes in the future.
No extra costs. Use the compute you already have.
Everyone uses the same software versions in the same environment despite working on different machines.
Limitations
Unlike Praxis AI, the course material is not embedded in an interface along with the computational access. Students must connect separately to their respective infrastructure.
You must ensure that the compute power is sufficient for the course.
A bit more time is required to train students to use Singularity and Docker.
Set it up Yourself
While Praxis AI and or the pre-configured Docker images make learning easiest, we recognize that some folks prefer to learn the nitty-gritty of bioinformatics software installation. Within the lessons we will provide instructions.
Advantages
Students learn how to install software.
Limitations
It is more time consuming
Students must have the ability to install software on the machine on which they work.