Phobos RCF Facilities (June 11 2003)

For a general description of RCF facilities goto here

Phobos Machines:

Gateway machines:

General Use

rssh.rhic.bnl.gov - ssh-only gateway (round-robins between different machines).

NOTE: Once you are on a gateway, you can ssh to the machine name without
specifying the net:
E.g. to get to rcas4002, just type
> ssh rcas4002 [ENTER]

Large Data Transfers Use + CVS

rftpexp.rcf.bnl.gov - scp/kftp/ssh-ftp gateway. Linux

Analysis and Reconstruction:

Analysis Machines:

Accessible by interactive direct login or by using the LSF batch queue system. from an interactive machine. Available for use by phobos collaborators.

Interactive machines:
Used to login, edit files, run programs etc.

rcas4001->04 :4 dual 800 MHz, running RedHat Linux.
rcas4005->08 :4 dual 1400 mHz, running RedHat Linux.
rcas4009, rcas4010: 2 dual 2400 MHz, running RedHat Linux 7.2.
rcas4011->15: 5 dual 1000 MHz, running RedHat Linux 7.2.

LSF only (No interactive login unless specified above)
LSF5 queues for MC (Monte Carlo) and analysis. Non-general Phobos use queues are specified.

LSF 5 queue(s)	machines(dual CPU)	speed	#machines	purpose
phslow	rcrs4001-4029	450 MHz	29	CS, analysis, MC
phcas	rcrs4030-4074, rcrs4075-4238	800 MHz, 1 GHz	88	analysis, repass, MC
phcasfast	rcrs4139-4165	1.4 GHz	48	analysis, MC
phcrs	rcrs4166-4260	2.4 GHz	99	new data production, analysis, MC

Specific queue details
(User job limit, time limit (in CPU-hours, not actual hours), and job priority)

phslow_hi: job limit 10, 1 CPU-hour time limit, high priority
phslow_med: job limit 45, 24 CPU-hour time limit, medium priority
phslow_lo: job limit none, no time limit, low priority
phslow_mc: job limit none, no time limit, lowest priority, MC only
phcas_hi: job limit 10, 1 CPU-hour time limit, high priority
phcas_med: job limit 70, 24 CPU-hour time limit, medium priority
phcas_lo: job limit none, no time limit, low priority
phcas_mc: job limit none, no time limit, lowest priority, MC only
phcasfast_hi: job limit 10, 1 CPU-hour time limit, high priority
phcasfast_med: job limit 45, 24 CPU-hour time limit, medium priority
phcasfast_lo: job limit none, no time limit, low priority
phcasfast_mc: job limit none, no time limit, lowest priority
phcrs_hi: job limit 10, 1 CPU-hour time limit, high priority, MC only
phcrs_med: job limit 70, 24 CPU-hour time limit, medium priority
phcrs_lo: job limit none, no time limit, low priority
phcrs_mc: job limit none, no time limit, lowest priority. MC only
phobos_int (rcas4001-4015): job limit 3, 1.5 CPU-hr time limit, low priority
phobos_all (all machines): job limit 30, NOT FOR general use

Queue Notes:
For summer 2003, production repasses will be done on phcas (and phcasfast if required - announced ahead of time), so jobs running on those queues may be superceded by production. However, the new, fastest phcrs queues will be available for analysis and Monte Carlo.
When we start taking data (in the winter 2003/2004), the phcrs queues will be used for new data production and the phcas queues left alone for analysis (depending on the DAQ rate).

For how to setup to use LSF 5 queues, goto (still mostly valid!) the LSF in Phobos web page

For more information on LSF queues availibility and use (somewhat old document) goto here.

Central Reconstruction Server (CRS):
           Accessible in batch through Tom Throwe's Reconstruction Control Software scripts.
           PHOBOS-specific instructions should eventually appear here.
          Use of this resource should be coordinated through
CRS Queues:

CRS Queue No.= 4
Use = Runs all phobos reconstruction/Pass0 jobs +MC when necessary. (By permission only, priority over CAS jobs). Will not run repass jobs for summer 2003
Machine IP's = rcrs4166->rcrs4264
Machine Specs = 99 dual Pentium-IIIs (2400 MHz). Running RedHat Linux 7.2

CRS Queue No.= 3
Use = Runs MC jobs & repass jobs primarily. Overflow on production.
Machine IP's = rcrs4139->rcrs4165
Machine Specs = 48 dual Pentium-IIIs (1400 MHz). Running RedHat Linux 7.2

CRS Queue No. = 2
Use = Runs MC jobs & repass jobs primarily. Not normally production.
Machine IP's = rcrs4075-rcrs4138
Machine Specs = 64 dual Pentium-IIIs (1000 MHz). Running RedHat Linux 7.2

CRS Queues     = 1
Use                 = Runs MC jobs primarily, repass/production jobs only if necessary
Machine IP's    = rcrs4030->rcrs4074
Machine Specs = 45 dual Pentium-IIIs (800 MHz). Running RedHat Linux 7.2

CRS Queue No.= 5
Use = Runs Checksum jobs for data taking & MC jobs. (By permission only, priority over CAS jobs)
Machine IP's = rcrs4001->rcrs4029
Machine Specs = 29 dual Pentium-IIIs (450 MHz). Running RedHat Linux 7.2

For more information on CRS queues availibility and use goto here.

Official RCF list of CAS machines & CRS machines

Disk Servers and Tape Servers/Data Movers:

NFS/Centralized Disk Servers

rmine401 (phobmds)- You should not login here. This node has access to HPSS and also the big disks are served/mounted from this machine (/phobos/u,/phobos/data/common, /scratch,/scratch2,/data02. i.e. home/user directories,phobos common file directories,$SCR, $DV_OUT).

rmine402 - You should not login here. This node has access to HPSS and also the big disks are nfs served from this machine (/phobos/data2000 (1TB),/phobos/data2001 (1TB), /phobos/userdata(0.66TB),/phobos/mc(0.78 TB)).

Tape Servers/Data Movers

rmds01 - You don't log into this node, but it's where you pftp to in order to see our HPSS data.
rmds02 - You never touch this node, but it's the one that actually "moves" our HPSS data.

RCF Utility Machines:

Email Server

rcf.rhic.bnl.gov- email server, can not directly login to. For details goto here.

Print Queue Server+Solaris machine for X-term connections

rcf2.rhic.bnl.gov - See's NFS mounted disks, has print queues,not gateway. Netscape etc. Operating system Solaris. May have some software individual rcas nodes don't.
Note: To login to interactive nodes from rcf2 need to connect to the linux gateway. This is done by ssh rssh.rhic.bnl.gov

This page is maintained by

Last Edited: Wednesday, June 11, 2003 14:40