This publication describes how to use the Cray Network Queuing Environment (NQE). NQE is a software product that lets you submit, monitor, and control batch jobs for execution on Network Queuing System (NQS) server nodes in the NQE cluster.
The Network Load Balancer (NLB) uses the system load information received from NQS server nodes to offer NQS an ordered list of nodes to run a request; NQS uses the list to distribute the request.
The NQE database provides an alternate mechanism for distributing work. Requests are submitted and stored centrally. The NQE scheduler examines each request and determines when and where the request is run.
The File Transfer Agent (FTA) provides asynchronous and synchronous file transfer. You can queue your transfers so that they are retried if a network link fails.
This manual contains the following chapters:
Chapter 1, “Seeing the Big Picture”, provides an overview of NQE components and basic functions.
Chapter 2, “Preparing to Use NQE”, describes which environment variables you must set to use NQE, how to set up NQE database authorization, and how NQS authorizes you to use client commands in the group of execution nodes in the NQE cluster.
Chapter 3, “Creating Batch Requests”, describes how to create a request and describes basic options you can use.
Chapter 4, “Submitting Requests”, describes using the NQE GUI or command-line interface to submit requests, using the NLB default queue, submitting requests to the NQE database, using request attributes, and using basic options.
Chapter 5, “Customizing Requests”, describes how to use limits, password prompting, alternative user names, and several miscellaneous options in your requests.
Chapter 6, “Working with Output Files”, describes how to customize where your output is delivered and how to find it if it did not go where you expected it to go.
Chapter 7, “Communicating with Requests”, describes how to monitor your output when your request is executing and how to write messages to executing requests.
Chapter 8, “Using Job Dependency”, describes how to use the cevent(1) command to make events in script files or requests interdependent.
Chapter 9, “Customizing Your Environment”, describes how to use environment variables to customize your NQS and NQE environments, and how to configure NQE displays.
Chapter 10, “Monitoring Requests”, describes how to use the NQE GUI Status window and the cqstatl(1) and qstat(1) commands to view request status.
Chapter 11, “Monitoring Queues”, describes the cqstatl(1) and qstat(1) command options available for viewing queue information.
Chapter 12, “Deleting Requests”, describes how to delete a request.
Chapter 13, “Signaling Requests”, describes how to signal a request.
Chapter 14, “Transferring Files”, describes how to use the ftua(1) and rft(1) commands to transfer files.
Chapter 15, “Monitoring Machine Load”, describes how to use the NQE GUI Load window to monitor system status.
Chapter 16, “Solving Problems”, provides troubleshooting information.
This manual also includes the following appendixes and glossary:
Appendix A, “Man Page List”, provides a list of all online user-level man pages.
Appendix B, “Command Line Interface Tutorial”, provides sample exercises on how to submit, monitor, and control a batch request.
Appendix C, “Using FTP with NQS”, provides information on how to use the ARPAnet standard file transfer protocol (FTP) with NQS on UNICOS or UNICOS/mk systems.
Glossary defines terms used in this guide.
The following documents contain additional information that may be helpful:
NQE Administration, publication SG-2150, provides information on configuring, monitoring, and controlling NQE. This publication may also be accessed online by using the Cray DynaWeb server and through the Silicon Graphics Technical Publications Library World Wide Web page at the following URL:
http://techpubs.sgi.com/library/ |
Introducing NQE, publication IN-2153, provides an overview of NQE functionality and describes how to access documentation online. This publication also may be accessed online by using the Cray DynaWeb server and through the Silicon Graphics Technical Publications Library World Wide Web page at the following URL:
http://techpubs.sgi.com/library/ |
NQE Installation, publication SG-5236, describes how to install or upgrade the NQE software. This publication also may be accessed online by using the Cray DynaWeb server and through the Silicon Graphics Technical Publications Library World Wide Web page at the following URL:
http://techpubs.sgi.com/library/ |
NQE Release Overview, publication RO-5237, provides NQE release information. This publication also may be accessed online by using the Cray DynaWeb server and through the Silicon Graphics Technical Publications Library World Wide Web page at the following URL:
http://techpubs.sgi.com/library/ |
The User Publications Catalog, Cray Research publication CP-0099, describes the availability and content of all Cray Research hardware and software documents that are available to customers. Cray Research customers who subscribe to the Cray Inform (CRInform) program can access this information on the CRInform system.
To order a document, either call the Distribution Center in Mendota Heights, Minnesota, at +1-612-683-5907, or send a facsimile of your request to fax number +1-612-452-0141. Cray Research employees may send electronic mail to orderdsk (UNIX system users).
Customers who subscribe to the CRInform program can order software release packages electronically by using the Order Cray Software option.
Customers outside of the United States and Canada should contact their local service organization for ordering and documentation information.
The following conventions are used throughout this document:
The default shell in the UNICOS and UNICOS/mk operating systems, referred to in Cray Research documentation as the standard shell, is a version of the Korn shell that conforms to the following standards:
Institute of Electrical and Electronics Engineers (IEEE) Portable Operating System Interface (POSIX) Standard 1003.2-1992
X/Open Portability Guide, Issue 4 (XPG4)
The UNICOS and UNICOS/mk operating systems also support the optional use of the C shell.
If you have comments about the technical accuracy, content, or organization of this document, please tell us. You can contact us in any of the following ways:
Send us electronic mail at the following address:
[email protected] |
Contact your customer service representative and ask that an SPR or PV be filed. If filing an SPR, use PUBLICATIONS for the group name, PUBS for the command, and NO-LICENSE for the release name.
Call our Software Publications Group in Eagan, Minnesota, through the Customer Service Call Center, using either of the following numbers:
| 1-800-950-2729 (toll free from the United States and Canada) |
| +1-612-683-5600 |
Send a facsimile of your comments to the attention of “Software Publications Group” in Eagan, Minnesota, at fax number +1-612-683-5599.
We value your comments and will respond to them promptly.