ARCHIVED: On XSEDE, what is Gordon (SDSC)?

This content has been archived, and is no longer maintained by Indiana University. Information here may no longer be accurate, and links may no longer be available or reliable.

On the Extreme Science and Engineering Discovery Environment (XSEDE), Gordon (gordon.sdsc.xsede.org) is an Appro International digital compute service hosted by San Diego Supercomputer Center (SDSC).

Built specifically to support data-intensive computing, Gordon features:

1,024 compute nodes, each equipped with two 8-core 2.6 GHz Intel Xeon Sandy Bridge processors and 64 GB of DDR3-1333 memory
64 I/O nodes, each equipped with two 6-core 2.67 GHz Intel Xeon Westmere processors, 48 GB of DDR3-1333 memory, and 16 Intel 710 Series 300 GB flash-based solid-state storage devices (SSDs), providing 4.4 TB of usable fast flash memory per node

Most "native" compute nodes on Gordon are equipped with a single 300 GB flash-based SSD, providing 280 GB of usable fast flash memory per node. For compute jobs that need more flash storage, Gordon also provides four "bigflash" nodes and 32 vSMP nodes:

Bigflash nodes: These "native" compute nodes each have access to a separate 4.4 TB flash-based SSD; for more, see ARCHIVED: On Gordon (SDSC), what are the "bigflash" compute nodes?
vSMP nodes: Using a virtualization solution designed by ScaleMP, each vSMP "supernode" aggregates 32 compute nodes and two I/O nodes, providing access to two 4.4 TB RAID 0 devices, each built from 16 300 GB flash-based SSDs.

The Data Oasis (SDSC) is mounted on Gordon, providing users access to Lustre-based shared scratch space and allocated project space.

For detailed specifications and user information, see SDSC Gordon User Guide in the XSEDE User Portal.

For more about XSEDE compute, advanced visualization, storage, and special purpose systems, see the Resources Overview, Systems Monitor, and User Guides. For scheduled maintenance windows, outages, and other announcements related to XSEDE digital services, see User News.

This document was developed with support from National Science Foundation (NSF) grants 1053575 and 1548562. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF.