Benefits of Using The Cloud

Working in the cloud is exceptionally scalable and versatile; you only use as much as you need, whether that’s in terms of storage space or processing cores. Cloud-based data is easily read by massively parallel processes, expediting results. When you’re done, resources disappear! You don’t have idle resources sitting around collecting dust.

Don’t be intimidated by the cloud! Scale your analyses using the data on ISB-CGC. If you’ve conducted bioinformatics before using the command line or SQL, this will be just as easy (if not easier) and we are also here to help. Email feedback@isb-cgc.org or visit our Community Notebooks page for guides and tutorials.

Most bioinformaticians today are likely accustomed to using the high performance compute (HPC) resources provided by their institution to conduct high-throughput bioinformatics analyses. Here’s a breakdown on how the Google Cloud Platform compares to your institution’s HPC resources.

Your University’s HPC Resource

Google Cloud Platform

Operating Systems

Linux, Windows

Virtual machines can run Linux and Windows

Compute

Virtual machines not determined by you

You can sign up with you own virtual machines*

Storage

Block Storage

  • Small storage is available in your home directory (usually around 1TB)

  • Some Scratch storage that is often deleted after a certain amount of time

  • Storage is usually a shared resource

Block Storage & Object Storage

  • Each virtual machine instance has a single boot persistent disk with a default size of 10GB that can be adjusted up to 64TB*

  • For storage that needs IO, consider persistent disks

  • Google Cloud Storage (GCS) buckets are the most flexible and economical storage option

  • You can save objects to GCS buckets including images, videos, blobs, and unstructured data

Pricing

Depends on the institution:

  • Institution provides basic HPC resources for researchers free of charge

  • PIs requiring larger-scale resources must purchase clusters and storage space

Pay as you go

  • You pay for the compute resources and storage that you use*

Do you have to wait?

Yes

  • Resources are shared among users

  • Scheduler systems used to schedule jobs based on resource availability

No

  • Once you’ve set up a Google Cloud Platform account, you can spin up a virtual machine and begin computing quickly

Is machine powerful enough?

Yes and no, depends on what you’re trying to do; often it’s a no

Compute resources and storage are unlimited, but you have to pay for it*

Accessing Cancer Genomics Data

Typically not stored on the HPC; you have to download to your local machine

Data is stored on the cloud

How to connect

Log in using Secure Shell protocol (SSH)

Log in using Secure Shell protocol (SSH)

*Be careful of costs, See the Cost Management page for more information.

Have feedback or corrections? Please email us at feedback@isb-cgc.org.