Hadoop and Spark request

To use Hadoop and Spark on Bridges, please complete this form. We will contact you within one business day once we receive your request.



Powered by ChronoForms - ChronoEngine.com

New on Bridges

GPUs to be allocated separately
Read more

Upgraded scratch file system installed
Read more

Hadoop and Spark Request Instructions

Programming: Python, Java and Scala are all available.

Framework: Request Hadoop, Spark, or both. If you will use Spark, you can also use HDFS if you choose.

Nodes: Bridges supports Hadoop and Spark on its RSM (128GB) nodes.  You will have the use of all 28 cores on the nodes you request.  You will also be allocated all the local disk associated with those nodes.

Time: Request the amount of time you need in hours.  You will be charged 28 SUs for each hour (1 SU = 1 core-hour).

Specific requests: Be sure to describe any special needs you have, such as a specific Hadoop implementation.  

Questions:  Contact bridges@psc.edu with any questions.