GenBrowse application/DB server

[Anna Battenhouse]

Hello -

We have been maintaining a partial mirror of the UCSC Genome Browser for a number of years, including only a few organisms (human, mouse, yeast) and excluding much of the wgEncode data that takes so much disk space.

We are now in a position to implement a full UCSC Genome Browser mirror, and have a solution in mind for the file server portion. This dedicated file server would have about ~100 TB available storage for goldenPath and gbdb files, and would be NFS mounted from the application/DB server via 10 gig Ethernet.

However I am looking for some advice regarding an application/DB server. This machine would run the MySQL DB for the mirror, so would need fast local disk with capacity for all the standard UCSC DB tables (~6 TB last time I checked) as well as DB tables for non-public tracks created by us. The server would also host multiple GenBrowse web applications, each with a different view of the underlying database (via local trackDb configurations) so that labs can browse their own data tracks.

Can you suggest specifications for the kind of app/DB server I've outlined? And could you provide specs on the machines you currently use for the standard UCSC Genome Browser?

Thanks much,

Anna

--

Anna Battenhouse | Associate Research Scientist | Iyer Lab | Institute for Cellular & Molecular Biology | Center for Systems and Synthetic Biology | The University of Texas at Austin | 2500 Speedway, MBB 3.106 | Austin, TX 78712-0159 | (512) 232-2632 (o) | (512) 587-4159 (m) | abatte...@utexas.edu

[Matthew Speir]

Hi Anna,

Thank you for your question about creating the hardware needed to create a full mirror of the UCSC Genome Browser.

This README file from our source tree describes our hardware setup:

http://genome-source.cse.ucsc.edu/gitweb/?p=kent.git;a=blob_plain;f=src/product/README.requirements

You can use this a reference point for the hardware that your mirror may require.

I hope this is helpful. If you have any further questions, please reply to genome...@soe.ucsc.edu. All messages sent to that address are archived on a publicly-accessible Google Groups forum. If your question includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.

Matthew Speir
UCSC Genome Bioinformatics Group

--