Weekly Report for Week Ending November 6, 2003

The LIGO Executive Committee Agenda for Monday  November 10, 2003 will be:

CANCELLED DUE TO LSC MEETING

Special Announcements:

Weekly Report Highlights
 

LSC Issues (Saulson)

no report

Laboratory Administration (Lindquist)

LIGO Weekly Site Telecon (Lindquist)

There was no site teleconference held Thursday, November 6, 2003.
The list of current actions revised to reflect the status of open actions assigned through October 30, 2003 may be found at ACTION LIST.

From: Ed Chargois <chargois_e@ligo.caltech.edu>

• No report this week 

>From: Linda Turner - turner@ligo.caltech.edu>

ACCOMPLISHMENTS (This report reflects activity for the past month.)

• No Report this week.

ACTIVITY

• Packaged/shipped 2 laptops and 3 projectors to Hanford for LSC Meeting next week.
• Electronic document activity has been steady.

Press here to access the DOCUMENT CONTROL CENTER WEB PAGE.

From: Esther Cunningham <esther@ligo.caltech.edu>

Press here for ACCOUNTS PAYABLE HISTORY DATA.

From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>

• Sent out the RFQ for the accumulators for Ken Mailand.  Received the responses and have provided the information to Ken Mailand.
• Completed change order #16 to Crystal Systems.  Completed change order #11 o Praxair.  Working on extending Digikey blanket purchase order.  Modified several existing purchase orders.
• Placed one time purchase orders as requested for Pasadena and the sites.  Placed and reconciled pcard orders.

• Completed RFQ for the HVAC Control And Security Maintenance (electrical instrument), which I've held for Ed Jasnow’s approval.
• The invoice issue between Sun Microsystems has now been resolved and ready for payment.
• Working on resolving un-paid invoice from Storagetek, PO and Invoice don’t match.
• Re-located closed PO’s to sub-basement.
• Created new storage space for open, PO’s to make room for subcontracts files.

• Financial Reports for Operations and R&D as of October for both fiscal years 2003 and 2004 were completed and posted on the network.
• Working on a detailed comparison of expenditures for FY 2003 to budgeted amounts for FY 2004.


Financial reports can be found at: http://docuserv.ligo.caltech.edu/~fireport

From: irena@ligo.caltech.edu (Irena Petrac)
Status on LSC updates:

• GEO600: Have been updating the GEO600 coordinates on the LSC Roster all day Monday and Tuesday. Tom helped with formatting the automatic Attachment Z macro to sort scientists by individual GEO600 groups rather than by alpha order for the entire consortium as set up before. There were a number of kinks to be sorted out and it took time. The draft-Attachment Z update for 08-15-03 to 02-15-04 is being processed for web posting today. Am also forwarding the finalized Attach. Z for Barry's sign-off today.
• Attach. A for 08-15-03 to 02-15-04 and Prog. Report through 08-15-03 are outstanding. A reminder e-mailed to A. Papa.
• Attach. B update for 08-15-03 to 08-15-04 and Prog. Report to 08-15-03 were received. Draft of Attach. C and Prog. Report were submitted to Linda for scanning and web posting.
• Attach. C update for 08-15-03 to 08-15-04 and Prog. Report to 08-15-03 were received. Draft of Attach. C and Prog. Report were submitted to Linda for scanning and web posting.
• Attach. D update for 08-15-03 to 08-15-04 and Prog. Report to 08-15-03 were received. Draft of Attach. D and Prog. Report are being submitted to Linda for scanning and web posting.
• NAOJ-TAMA: Attachment D updates for 02-15-03 to 08-15-03, and for 08-15-03 to 02-15-04 were signed-off by Barry and are being Fed-Exed to Japan for Prof. Kozai's sign-off. Attach. Z updates for the same LSC period went over to Japan Oct. 31st. Prog. Reports through 02-15-03 and 08-15-03 were submitted to Linda for scanning and web processing.
• MOU with Univ. of Southern California Information Sciences Institute (USCISI) was signed-off by Barry and went over to USCISI for their signature. USCISI PI's office was advised accordingly.
• LIGO sign-off on Attachments No. A and No.1 is pending Albert's approval. Albert is still concerned about the intellectual property issue. In summary, the MOU has no reference to intellectual property. Attach. No. A includes a standard NSF clause on Intellectual Property Rights. Per Albert, all we need to proceed is Caltech's clarification of intellectual property with respect to software involved in the data and results obtained. This issue has been forwarded to Ed Jasnow for disposition. Albert is aware of this but asked me to see this issue through somehow anyway.
• IAP/Russians: All outstanding inputs are in, Attach. C and Z updates for 08-15-03 to 02-15-04, and Prog. Report for 08-15-03. Drafts of Attach. A and Z, and Prog. Report are being submitted to Linda for scanning and web posting.
• Rochester: Attach. A and Z updates for 08-15-03 to 02-15-04 are in process. Attach. Z is in Barry's sign-off and Attach. A is pending Albert's approval. Prog. Report through 08-15-03 was submitted to Linda for scanning and web posting.
• Southeastern Louisiana: All outstanding inputs are in, Attach. A and Z updates for 02-15-03 to 08-15-03, and 08-15-03 to 02-15-04, and Prog. Reports for 02-15-03 and 08-15-03. Drafts of Attach. A and Z, and Prog. Reports are being submitted to Linda for scanning and web posting.
• IUCAA: All outstanding inputs are in, Attach. A and Z for 02-15-03 to 08-15-03, and 08-15-03 to 02-15-04, and Prog. Reports to 02-15-03 and 08-15-03. Drafts of Attach. A and Z and Prog. Reports are being submitted to Linda for scanning and web posting.
• Balearic: All outstanding inputs are in, Attach. A and Z for 08-15-03 to 02-15-04, and Prog. Report to 08-15-03. Drafts of Attach. A and Z and Prog. Report are being submitted to Linda for scanning and web posting.
• Florida: All outstanding inputs are in, Attach. A, C, D and Z for 08-15-03 to 02-15-04, and Prog. Reports to 08-15-03. Drafts of Attach. A, C, D and Z and Prog. Reports are being submitted to Linda for scanning and web posting.
• Loyola: All outstanding inputs are in, Attach. A and Z for 08-15-03 to 02-15-04, and Prog. Report to 08-15-03. Drafts of Attach. A and Z and Prog. Report are being submitted to Linda for scanning and web posting. Attach. A for 02-15-03 to 08-15-03 is pending Albert's approval. Attach. Z for the same period is being signed-off by the PI.
• MSURG/Russians: All outstanding inputs are in, Attach. B, D and Z for 08-15-03 to 02-15-04, and Prog. Report to 08-15-03. Drafts of Attach. B, D and Z are being submitted to Linda for scanning and web posting. Prog. report was submitted for web posting several months ago.
• Stanford: All outstanding inputs are in, Attach. A, B, C, D and Z for 08-15-03 to 02-15-04, and Prog. Reports to 08-15-03. Drafts of Attach. A and Z and Prog. Reports are being submitted to Linda for scanning and web posting.

ACIGA
CACR (have given up on this group)
CEGG (Prof. Drever promised)
Dominguez Hills (PI promised)

Goddard (sporadic)
LaTech
LSU (group very busy at LLO)
Michigan
Oregon (promised)
Southern University A&M (promised)
Wisconsin-Milwaukee
Washington State

From: Ed Jasnow <jasnow@ligo.caltech.edu>

• No report, on vacation.

SUPPORT (Baldon, Lloyd, Tischler)

>Irene Baldon

• No report.

• No report.

• Outreach Review follow-up
• NSF Review, PAC15 details and arrangements Vacation Monday and Tuesday of this week.

ADVANCED LIGO (Cost Schedule Control Systems) T. Frey
From: Thomas Frey <tfrey@ligo.caltech.edu>
 

Out of the office Friday 10.31
 

Accomplishments:

For list of documents that are being used to develop Adv. LIGO Cost and Schedule, see http://www.ligo.caltech.edu/~tfrey/Cost_MTG_082002/
 

Continued development of a proposal for implementing Primavera Project Planner Enterprise.

Continue working on draft / text of implementation.
Clink on links to view the Draft Work-plan and Estimate.
Continue testing of Primavera Enterprise version 3.5.1.  See the links below for information / technology demonstrations: (From Local Primavera Dealer)
Primavision - http://primavision.drmcnatty.com:8080/phoenixwl/login_p3ec  Login using Username:  admin and Password:  admin.

Progress Reporter - http://www.projwebsite.com/GroupServerP3e/en/  Login using Username:  admin and Password:  admin.

Static HTML Example - http://ligo.caltech.edu/~tfrey/SUS_WebProj/ (AdL Suspensions Only)  NOTE:  This is after importing and before any data corrections.

Static HTML Example - http://ligo.caltech.edu/~tfrey/DAQ_WebProj/ (AdL DAQ Only)  NOTE:  This is after making adjustments to the WBS so that it displays as expected.

Will attend the Users conference in November at Orlando, Florida.  See http://www.primavera.com/events/conference.html

Phone conferences with Carol regarding preparation for the COC update, PSL update, IO update, and SEI update.

Finalized travel arrangements for attendance at meetings at Livingston as requested by Carol Wilkinson.

The following is a summary of status by sub system:
COC - Still need completed WBS dictionary and BOE.  Prepared revised update schedule for COC with progress through 09.26.02.  Posted changes for review.
http://ligo.caltech.edu/~tfrey/COC_AdvLIGO/COC_index.html
PSL -  No action items pending.  Followed up with Peter and Benno has provided data.  Peter has yet to review.  Will be attending meeting at Livingston pending schedule of meeting.

AOS - Still need completed WBS dictionary and BOE.

IO -  Still need completed WBS dictionary and BOE.

SUS - No action items pending.  Worked with Janeen and Calum preparing revised progress data / ETCs.  Janeen provided the ETCs and I will review with Carol.

SEI - No action items pending. Posted updated SEI schedule with progress through 09.26.02.

http://ligo.caltech.edu/~tfrey/SEI_AdvLIGO/SEI_index.html
ISC - Still need completed WBS dictionary and BOE.

LDAS - No action required at this point in time.

INSTALL - No action items pending.

DAQ - No action required at this point in time.

PM - No action required at this point in time.

FAC - No action required at this point in time.

SUP - No action required at this point in time.

• Updated the plan with progress as of 10.31.03 and  posted to site.
• Sent out update data requests for progress as of 11.14.03.

• Updated the plan with progress as of 10.31.03 and  posted to site.
• Sent out update data requests for progress as of 11.14.03.

• Updated the plan with progress as of 10.31.03 and  posted to site.
• Sent out update data requests for progress as of 11.14.03.

ROSTER DATABASE:
Continued input of changes / corrections regarding supervisors and departments.

COST BOOK DATABASE:
Reconciling cost account listing from cost book and schedule loading.

Project Web Site for posting schedule and progress related data continues to be updated with the latest and greatest.

See http://www.ligo.caltech.edu/~tfrey/index.html

A pointer for a draft Final Report for Cooperative Agreement NSF PHY-9210038, the cooperative agreement that covered construction, installation and commissioning, and initial operations of the sites, has been distributed for internal review.  The draft has also been placed on the web page for the NSF Review committee.

The following change requests are open.

CR-030015	FY 2003 Livingston Observatory Detector Maintenance Expenses (Increment)	R. Wooley	July 14, 2003
CR-030016	Hanford Facilities 2.2--Divide the Large Equipment Access to Facilitate Movement of Large Items (currently assigned to FY 2004 liens list)	J. Worden	July 31, 2003

CR-030017 proposes to upgrade the LIGO Laboratory Wide Area Network (WAN) at Livingston from 2xT1 (3.088 Mbps) to Gigbit Ethernet (1000  Mbps) at a cost of approximately $217K through 2006.  This change request was discussed and approved during the Executive Committee Meeting on Monday, October 27, 2003.  Minutes have distributed (LIGO-M030197-00-P). 

From: Cindy Akutagawa <cindy@ligo.caltech.edu>

• Various personnel/payroll /HR related work.

Quality/Safety (Tyler)

Nothing significant to report this week.

LIGO Hanford Observatory (LHO) and Interferometer Operations (Raab)

Summary of S3 Activities at LIGO Hanford Observatory  (compiled by M. Landry)

The opening week of S3 saw some excellent stretches of locking on both interferometers, and a few mysteries, as well.  Duty cycles to date, longest lock stretches, and mean inspiral ranges are 73%, 20.5 hours and 2.4Mpc (H1), and 63%, 13 hours and 0.9Mpc (H2).  Note the longest lock stretches completely overlapped, for nearly 13h of dual science data.  As in previous runs, operators and scimons have been manning shifts around the clock, however, for this run there are no (well, few) training scimons.

The atomic and gps clocks show stability to better than 100ns.

Fake pulsar signals (five public and five hidden ones for blind analysis) are being injected near-continuously into both IFOs (later, on dedicated h1 and h2 machines).  PEM injections have also occured.

Up-to-the-minute figures of merit, uncalibrated noise, excitation waveforms and site overview screens are available as JPGs at the CDS site http://blue.ligo-wa.caltech.edu .  Data
dropouts have been observed on DMT figures of merit, some of which were traced to overloading of DMT machines, others due to disk and RMF problems (datapusher fix here).  We would like to see the LLO state vector and range plots, but technical difficulties have prevented this data transfer to date.

One of the persistent mysteries of the first week of the run has been, why is the 4k inspiral range unstable?  Range estimates are about a factor of five noisier on the 4k than the 2k, despite both machines having full (dc) WFS applied.  Along with the extra hash, the range wanders on hour timescales between 1.5-3.0Mpc.  Saturation of the SPOB photodiode was suspected to be occuring, however, reducing the light via ND filtering on the appropriate photodiode had no effect on the range noise.

Wednesday proved to be difficult locking, particularly on the 2k, which had some input pointing issues.  A 2k mystery is why the high frequency noise (normally the quietest in displacement of all three IFOs at hf ) has risen.

Causes of lockloss include seismic transients traced to our nitrogen dewar.

LIGO Livingston Observatory (LLO) and Interferometer Operations (Zucker)

no report

40 Meter Interferometer (Weinstein)

• Commissioning (Miyakawa, Kawamura):
• Michelson, X-arm, Y-arm all locked simultaneously.
• Using three LSC RFPD (POX, POY, and AP), Osamu and Seiji have succeeded in locking these 3 degrees of freedom at the same time, robustly, last Monday. We celebrated with a lunch party, today (Thursday).
• The Y arm gain has a sign flip with respect to the X arm, probably due to flipped magnets on ETMX.
• The Michelson signal was very noisy at first. This was traced to ADC digitization noise from the error signal readout with low gain on the whitening boards. They adjusted the gains and software filtering for the LSC channels, Now, all 3 loops have software gain around 1, and exhibit robust, low-noise locking and relatively high UGF's.
• Osamu and Seiji continue to tune and improve the lock acquisition and in-lock servos optimization, including better tuning of the Michelson demod phase, to reduce the influence of the arm signals on the Michelson.
• The Michelson contrast is very good; to be measured carefully, soon. We still need to adjust the mode matching into the arms.
• We need to re-measure the arm reflectivity; but it sure looks like the mirror surfaces have lots of dust. We plan to vent soon, to clean them (and fix a misplaced OSEM on ETMX).
• Next step is to move to the common-mode and diff-mode control topology, and implement a common mode servo (based on the LLO incarnation).
• Then, characterize, measure response, calibrate, noise spectrum, etc.
• The longer-term plan proposed by Seiji and Osamu is to vent, clean mirrors, fix osems, but don't install the recycling mirrors yet. Then, re-establish operation, with clean mirrors. Then, commission 166 MHz sidebands, and see how robust the Michelson signal is in the (166-33) MHz demodulation, where the carrier light is not used. This is a key element of the AdvLIGO control scheme, allowing us to control the short degrees of freedom (Michelson, PRC, SRC) without being sensitive to the locking and control of the long arms. We need to build 133 MHz frequency distribution system and demod boards.
• The mode cleaner and its ASC WFS system are working well. Remaining low-frequency intensity fluctuations in the MC transmitted beam are from upstream of the MC, so Flavio is now commissioning the Intensity Stabilization servo for the PSL, to address this.
• Osamu and Seiji are optimizing all the control loops with the help of 6 scopes, including three fast (100 MHz or higher, with high sampling rate) 4-channel digital scopes. We need more scopes, more shutter controllers, more SR560's. (Basically, we need more money!).
• We also need to learn how to use the EZCA scripts for autolocking, calibration, etc.
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
• Jay found some problems with signals from one of the LSC PD interface boards. He and Ben trouble shot the problem, and found two independant causes. One was a low voltage problem on the +/- 5V supplies, and another was an ungrounded LEMO connector. Both have been corrected, and the signals are fine now. Ben will survey all the cables.
• Ben and Sander are getting parts for two more EO shutter controllers. All the cables for three EO shutters are in place.
• The PCB for the LSC PD tester board arrived Wednesday, and Ben will stuff and test it today.
• Mohana is completing the code, databases, and screens for the ASC auxiliary crate, and we hope it will go in today, so that we can read out all the QPDs in the vertex and the ends.
• We have a long list of new signals (ISS, ASC QPDs, etc) that need to go into master.config.
• Digital suspension controllers (Sakata, Goggin, B. Abbott, Heefner)
• Shihori applied her iterative procedure to better diagonalize the input matrix, to ITMY and BS (ITMX is already much better diagonalized).
• She is now working with Osamu to diagonalize the output (coil) matrix of the BS to help with michelson lock stability.
• Her work is severely hampered by the inability to run DTT with bandwidths smaller than 0.05 Hz; we hope that this is due to the slowness of the Sun Blade 100, and will be remedied when we get the new Sun E450's online.
• Ben is modifying six of our Universal Dewhite Boards to allow the nulling, or at least equalization, of offsets when switching in and out hardware dewhitening.
• Optical levers (Kawazoe, Heefner, Abbott):
• Fumiko has analyzed her oplev response calibration data; all 6 oplev sensors have similar slopes (response in volts versus pitch and yaw) near their operating point.
• She will use this information to calibrate the pitch/yaw bias controls, measure the rms noise (in radians) with the mirrors (a) undamped; (b) damped; (c) under IFO lock.
• Next step is to import the oplev servo filters from the sites, make any necessary mods, and commission the servos.
• Optical sensing (Smith, Kawazoe):
• Fumiko aligned the optical, photodetector, and electronics chains for the POX and POY beams. Osamu and Steve did the SP beamline. Still to be done: TRX, TRY, POB.
• Fumiko is installing all the QPDs on all the beamlines, and awaits the installation of the ASC auxiliary readout software that Mohana will install soon.
• Fumiko will commision the EO shutters on the SP and AP.
• She and Lisa will put in all the ccd cameras.
• Computing:
• We desperately need to bring our new Sun E450 online so that we can (hope to) run dataviewer and dtt reliably (we've had no end of problems with these programs, running on an under-powered Sun Blade 100). We will need help from the computing group, and Alex Ivanov.
• Modeling (Miyakawa, Kawamura):
• Seiji is continuing to study error signals with Finesse while sweeping through (l+,l-,ls) space.
• Per his request, Andreas Freise quickly implemented the ability to sweep 3 DOF simultaneously, into Finesse. Seiji now writing scripts to explore the 3D space.
• Facilities and vacuum envelope (Vass):
• Flavio and Ben are beginning to look at the STACIS controllers at the 40m. They are having a colorful array of problems, and we will characterize them as we find them. Ben has ordered another copy of the control software that we can use to better understand the fault conditions.
• Steve has been ordering commissioning equipment (scopes, sensor cards, camera cables, etc). He sent back the metone particle counter for repair. It has been reading zero on the South Annex flow bench, but it was a lie. He checked with our other (working) particle counter; it's 10-15, still quite low (but not 0).
• The turbopump and scroll pump that failed during the last pumpdown has been repaired and returned.
• South Annex Bake Ovens (Taylor):
• Bob has been on vacation for the last 2 weeks.

Larry Jones needed some RGA scans for AdvLIGO SEI parts. Steve ran some scans on coils, which failed. Steve later discovered it was due to a blown fuse on the heater strips on the bake oven. He's now doing a magnet bake/scan for Larry.

Thermal Noise Interferometer (Libbrecht)

Last week we reported a measurement of coating thermal noise. This week
we have broken down several different noise source contributions and
shown that they do not account for the noise floor. (See the plot.below).

Also, we have performed two additional tests on the data to rule
out unknown noise sources before or after the cavities. First, we
varied the laser power and looked at the total noise in the instrument
at two separate modulation depths. The total voltage noise scaled
linearly with the power, ruling out shot noise or electronic noise in
the photodetector and subsequent measurement electronics. This, and the
fact that the Pound-Drever-Hall discriminant is proportional to the
laser power, indicates that the equivalent length noise is independent
of laser power.

Second, we varied the modulation voltage applied to the arm-cavities'
Pockel cell. Again, we found that the equivalent length noise was
independent of the modulation depth, as expected for fluctuations in
the arm cavities' lengths as opposed to, say, phase noise introduced by
the Pockels cell. (Thanks to David Shoemaker for pointing out this
possibility.)

These two tests, along with the measurement of the servo noise shown in
the figure, indicate that the noise originates inside the cavities, and
not before or after them in the servo or measurement loop. This,
combined with the fact that the noise floor between 500 Hz and 20 kHz
is incoherent between the two cavities, provides strong evidence that
our noise floor is dominated by thermal noise and not some other,
spurious noise source.
 

LASTI (Ottaway)

no report

Data Analysis and Computing (Lazzarini)

Simulation and Modeling (Yamamoto)
 

>> Commissioning support
Matt worked at LLO on ASC system (QPDs, WFS3 and WFS4), and scripting tools
(ezgain, ezlockin).
Biplab is on S3 shift at LHO.
 

>> Linear noise Model
Matt has been working on a linear noise model to analyze the sensitivity.
He has sent out first version of LinLIGO, a linear model of the initial LIGO detectors
to limited parties.
 

>> Mechanics modeling
There was a biweekly teleconference of the mechanical simulation on Wednesday.
Cella, Virginio, Mark Barton and Hiro attended. Virginio showed a measured
transfer function of a simple blade which will be used to validate the MSE blade model.
A few test programs of MSE have been discussed which showed apparent bugs to be fixed.
Cella and Virginio will address the problem this week working together over phone.
Mark has built a LIGO I like simple suspension model which can be used to validate the
MSE model.
 

>> FFT
Hiro has replaced the Fast Fourier Transform code in the FFT model by FFTW v3.0.1 and
the speed was further improved by 40%. Now the speed is entirely dominated by the
speed of the actual FFT calculations for the field propagation.  The speed of FFT program
on linux box is almost 10 times faster than the same job on the SUN box using the original code.
 

Hiro tried to use multiple threads to further speed up the calculation, but, for some reason,
no speed gain was achieved. Multi threads can be used, but the gain of using multiple threads
is somehow waisted somewhere.
 

The comparison between the FFT and the e2e calculations for the same LIGO I configuration
showed some sign difference. Biplab is looking into this problem.
 

Aside from this sign issue, the code can calculate 3 pick offs, POB, POX and POY.
 

>> Modeler code
Melody and Hiro have been working on the implementation of the C++ based FUNC,
taking the comment of Matt into account.
 

The interface design is underway which allows to implement easily a simple case like
out0 = PI*in0;
as well as to implement a code which takes full advantage of using the C++ compiler.
 

The infrastructure testbed has been implemented to demonstrate the basic design.
The compilation is needed only for those which are modified since the last build of
the shared library.
 

>> SimLIGO run
Xiao is running SimLIGO to test the effect of the arm offset (wrt to the beam axis).
He is installing the latest version of JAVA on his linux box to use alfi.
 

Hiro is studying the mysterious excess noise at around 100 Hz predicted by SimLIGO
by building a simple box, TestSeis, which contains only the seismic motion and the
LOS box. Several people had hard time to use these modules, so more comments will
be added to boxes to clarify the usage.
 

LIGO Data Analysis System
 

Software Systems (Blackburn)
LDAS continues to run very well at the sites during the science
run. This is particularly true of the createRDS commands that are
generating the Reduced Data Sets that are being moved back to
Caltech and on to the Tier II centers. A few dataPipeline jobs
have been submitted at LLO and LHO for waveburst and knownpulsar
search codes. But the full onsite searches have not begun. It
looks like that may be delayed until next week. Also a new release
of LALwrapper will be needed before some of the planned onsite
searches can begin. With the LSC meeting next week, it looks
like this will be more likely to begin late next week at this
point.
 

A small patch to the mpiAPI was placed at the sites this week
which corrected a problem with wrapperAPI-master processes being
assigned to nodes on the cluster when they should not have been.
This didn't cause any issues in nominal jobs, but would have
caused debugging and tracing confusion without the patch in a
job that was behaving poorly.
 

The LDAS software is currently based on the TCL 8.3.5 release. The
current release is 8.4.4. Until recently, we have experienced SWIG
wrapper issues with post 8.3.5 TCL. This week we made significant
progress towards wrapping up C++ code with SWIG for the newest TCL
release 8.4.4. This new release has fixes for several problems we
have routinely seen in LDAS.
 

A few minor changes were made to the build-ldas script to allow
more flexibility in the selection of compiler options.
 

The FrDump6 utility that is used to look at the contents of frame
files was "polished" up a bit this week.
 

Changes were made to the mpiAPI and the wrapperAPI to decrease the
amount of diagnostic information logged at runtime.
 

The official release of STLPort (version 4.6) came out this week.
We began testing it within LDAS on Monday. There may be a memory
leak exposed by this version of STLPort. We will follow the memory
usage over the next week to determine if there is a problem or not.
We have also backed out our home growned STL::String class in LDAS
in favor of the true STL::String class provided by STLPort. The
STLPort version is just as stable in a threaded environment as the
one we had to develop just in time for the science run AND is
definitely far more stable than the GCC version of this class. The
STLPort library comes with the option to use different memory
allocators. We are experimenting with these to see if any are
superior in performance on Solaris and/or Linux.
 

We have begun a pilot effort to provide graphical presentations
of the results in the database using the controlMonitorAPI. This
will help with system testing of LDAS and may also have a role in
the monitoring of science results during the science run in the
control room.
 

Hardware Systems (Wilson)
LHO
Ben Johnson
1) Replaced T3 disk u1d9@t3-8. Disk reconstructed without mishap.
2) Storagetek came and replaced the 3rd 9940B tape drive in the L700. He
initially installed and powered on the drive, without unlocking the drive
head. Drive would not initialize properly until the head was unlocked.
3) Replaced motherboard on defiance/datagateway Linux IDE RAID box. The box
booted successfully and sees all of the original devices.
4) Developing and testing new features in the disk2disk copy script. The
script should now automatically attempt to recover missing files from the
backup framebuilder when it sees data gaps or if no new data has been
written for a long time. I have also added the capability to run FrCheck on
each file and report any errors.
 

Greg Mendell
1) Started archiving full-frame S3 data, and generating level 1, 2, and
3 RDS frames.  RDS frames are being archived onto tape as well.  Each week
tapes with full-frame data are being sent to CIT, and tapes with level 1 RDS
data are being exchanged between LHO and LLO.  The level 3 RDS data is being
pulled over the internet to CIT, for distribution to tier II sites.
2) There have been some problems with tapes and tape drives at LHO.  The
system is redundant and this has not caused any significant problems with
archiving or RDS generation.  We are working to fix the problems.
3) Archived preS3 stochastic injection data.  Currently creating a L2 RDS of
this data for the stochastic group.
 

LLO
Igor Yakushin
1) Restored all the old databases at all the sites.
2) Restored DB2 installation at MIT after the disk crash.
3) Configuring sites for regular database backups.
4) Archiving raw S3 data, creating L1,L2,L3 RDS is going fine.
 

MIT
Keith Bayer
Completed work on pcraid series 2 motherboard replacement doing cabling and
setup for two more racks of cluster nodes (45-87) rebuilt metaserver after
failed drive LDR setup work
 

CIT
Hari Pulapaka
1. Published S3 L3 rds data from the sites.
2. Migrated the LDR installation to ldas-cit from ldas-cit-g
3. Developed a new script for publishing the data with some improvements on
performance.
4. Wrote a new storage module for ldas.mit.edu, setup LDR at MIT and now S3
data is flowing to mit using LDR.
5. Installed some parts of globus again on desktop, as there were problems
created during hte upgrade.
 

Dan Kozak
Imported tapes with S2 L1/3 RDS at observatories.
Set up S3 directory structures and archive sets at observatories.
Electronically labeled tapes at both sites.
Transferred pre-S3 full frames from observatories to CIT successfully.
Recovered bad file in archive from old tape.
Helped sites with tape problems.
Set up web monitoring of files remaining to be archived at sites.
Worked to figure out slow frameAPI performance...
 

Data Analysis Activities (Lazzarini)
Lazzarini:
Wrote a note on optimally combining the two Hanford interferometer data streams
with the Livingston stream when it is known that the H1, H2 machines have
intstrumental cross-correlations.
See http://docuserv.ligo.caltech.edu/docs/public/T/T030250-01.pdf
 

Mendell:
Mike Landry and I are working hard on the stackslide code.  We have done
a preliminary analysis of the pulsar injections in the E10 data for the
first 5 pulsars, stacking the data only, but not sliding.  The results
look good, so far.  That is, we see the injected pulsars that we
expected to see.  However we need to do a more in-depth study to make
sure we are getting the proper SNR. We are working to get results with
sliding by the PULG F2F this Sunday, Nov 09.
 

Yakushin:
1) Ran simulation to estimate waveburst sensetivity to BH-BH merger
depending on strain, direction and black hole mass..
2) Preparing waveburst presentation for the LSC meeting.
 

Sutton;
- Looking at the E10 hardware injections.
 

General Computing (Wallace)
MIT:
(Keith)
-Replaced emvogil-3 hardware - bad board or cpu
 (after trying just about everything else I could think of)
-Investigating cisco wireless hub to solve problems with
 mini-pci wireless card access failures (IBM & Mac laptops)
 

Livingston:
(Shannon)
-Still working with the new file server.  I have been trying to tweak the
performance issues for a Solaris NFS client and a Linux server.  I have
started by changing the rsize and wsize on the Solaris client.  I will
start tweaking settings on the Linux server tonight.  I will mail out
the performance numbers when I get them all compiled.  I have also set
up RAID 10 vs. RAID 5 on the file server.  Because of the instability
that I experienced in the first two weeks, this gives me a little extra
peace of mind due to the added redundancy.
-Shuffling some files off of our existing file server to the new file
server to free up some space.  Most of these are archived files that are
not being currently used.  I have also burned some of them to DVD for
archival purposes.
-Still working on recompiling all of the gnu and GPL applications on the
new file server.  This is a time consuming process since each one has to
be compiled individually and at last count there were over 100
applications compiled in the existing directories.  I am also employing
the gnu stow software to aid in future updates for software.
-Placed an order for a replacement machine for our web server.  The web
server currently is an Ultra 10 which is aging and is also short on disk
space.
-Received a quote from Sun for some additional software in their Sun ONE
line.  This includes a web mail server and a calendar server, among
other things.
The usual round of patches, ordering parts/computers for people, support
etc.
 

Hanford:
(Christine)
 

- Installed a new wireless access point that does both b and g
standards, which is  11 Mbs and 54 Mbs at 2.4 GHz.  This will be
available in the auditorium during the LSC meeting.  I plan to upgrade
all the wireless access points to at least the b&g, possibly the a
standard, which is 54 Mbs at 5 GHz.
- Most of the week has been spent getting ready for the LSC meeting.
- The usual user support.
 

CIT:
(Mike)
-Worked on Calum Torries computer. He is having problems with all
applications and I cannot get the new video card to work with solid
works. This computer seems to be having a memory/motherboard problem.
-Worked on the printers up on the 3rd floor of W/B; mostly paper jams.
-Worked on VRVS. I setup a meeting with Veronica running the share
option within VRVS. This approached seemed to have worked as far as
being able to read the speakers presentation.
-I worked on loading a visitors workstation that is currently running
win98. I am loading a workstation that is a lot faster and will have
2000 pro loaded.
-I worked on Cindy's new laptop. The internal wireless card seems to have
taken a dump. I called Dells tech support on this and after a few hours
of running diagnostics with them they have decided to send me out
another wireless card.
I have received the wireless card and swapped it out.
-Worked on all NTSRVs getting all end of month ghost backups.
-Worked on Antares troubleshooting DCC database. I used the access tool
that was designed to troubleshoot the database/drive mappings. This tool
works.
-Setup two laptops and got them ready to ship out to Hanford.
(Angelo gave me a hand with this)
-Worked on writing up Angelo some installation notes for General
Computing Workstations.
-I worked on finishing up loading Janeen's new laptop with GC software
and multiple engineering packages.
-Searched for a compatible video card for solid works that will work
for Calum Torrie & Ken Mailands computers. This is an additional video
card to implement a dual monitor system per workstation. I spent a lot
of time with this, due to a good percentage of the video cards according
to solid works HCL are APG cards. This slot is already taken by the
original video card. This forces me to look for a pci card that works
with solid works. I found a dual head AGP video card that is compatible
with solid works. This should take care of the problem. I will be seeing
about ordering this card through Larry Wallace.
 

(Veronica)
 

- A good portion of the week was spent producing a video of beamspot
scatter for Bill Kells. Combined trimmed clips into a single movie. I
captured it in different formats to see which one gives the best playback
on his laptop. Bill needed to download the upgrades of his media players.
Working on making a compressed version that will play back on his slower
machine without compromising too much quality.
Captured and compressed a video of last LIGO seminar by Karl Gebhardt. The
video was posted online for streaming within 2 hrs of the talk. Set up and
monitored VRVS for the talk. Downloaded the latest VRVS/VNC package on
Karl's laptop. This required a trip to Carnegie observatory where he was
stationed. As Mike reported, the broadcast went well, we received a 'thank
you' from Pisa.
Helping Karen Willacy of JPL with setting up a document database for LISA,
similar to our DCC. Meeting with her to explain the 3-tier application
architecture and practical implementation.
Updated the personnel roster. Updated the LIGO website.
- LSC website: Posting updates to the LSC meeting website.
 

(Bruce)
* (BS) Systems Work:
       - Continuing work on Ilog upgrade to implement entry
         editing features.
 

(Lisa)
- Started monthly backups and verified the dailys.
- Continued evaluating a new anti-spam package.
- Cleaned up random tasks that accumulated while I was out of town.
 

(Larry)
-Spent time going over a number of procurement related issues. Paperwork for the
Pcard system has really been taking up some time. Along that same line, I
discovered that those who have real old versions of the P-card s/w will need a
couple of files changed. On Oct. 25th, the P-card server was changed and the
older s/w needs to be updated by hand. Most of the LIGO people will be OK but
there are a few of us that this does affect.
-Assisted the E2E group on a number of computer setups and moves. One of the
computers appears to have a h/w problem but that will not be resolved for a few
more weeks.
-Worked with Christine and others on a couple of logistical issues for the LSC.
-Assisted a number of users with different PC related issues.
-The regular setting up of accounts and helping end users work out their
configuration issues.
-Continued working on the server system in the 40M. The OS is loaded with the
patches but now there are a number of configuration issues as well as the
in-house s/w loads that still need to be addressed.
 

Advanced LIGO and supporting R&D (Shoemaker)

For additional information about this report, contact sanders@ligo.caltech.edu