Weekly Report for Week Ending December 21, 2000

The LIGO Executive Committee Agenda for Monday  December 25, 2000 will be:

CANCELED DUE TO XMAS

NO WEEKLY REPORT BETWEEN XMAS AND NEW YEAR'S DAY

WEEKLY REPORTS WILL RESUME DUE TO PHIL LINDQUIST ON JANUARY 4  , 2001.

EXECUTIVE COMMITTEE MEETINGS RESUME ON JANUARY 8. BARRY WILL CHAIR THAT MEETING.

 (Meeting time: 10:30 am Pacific Time)

Open meeting 10:30 - 11:30

1. Announcements
2. LSC Issues (Weiss)
3. Comments on Weekly Report
4. WBS 1 LIGO I Construction (Lindquist)
• Field Change Orders/Contingency Liens/Change Requests
5. WBS 2 LIGO Lab Operations
• Administration (Lindquist)
• Sites (Raab, Coles, Shoemaker, Sanders)
• Detector (Whitcomb, Coyne)
• Campus Research Facilities (Weinstein (40 Meter), Libbrecht (TNI), Zucker(LASTI))
• Data Analysis and Computing (Lazzarini)
6. WBS 3 and 4  Advanced R&D and LIGO II (Sanders)

Special Items:

Special Announcements:

Weekly Report Highlights
 

LSC Issues (Weiss)

No report

LIGO I Construction/LIGO Laboratory Administration (Lindquist)

WBS 1.2 LIGO Operations--Administration

LIGO Weekly Site Telecon (Lindquist)

The next site telecon is scheduled for Thursday, January 4, 2001.

The list of current actions revised to reflect open actions assigned through December 21, 2000 may be found at ACTION LIST.

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

• Assisted the Seismic Attenuation Group (R. DeSalvo) with shipping and customs clearance of the TAMA SAS from Gallie & Morelli (Italy) to Caltech's Receiving; expect delivery 12/20/00.
• Received the Insurance reimbursement of $3747.00 to be credited to LIGO.00007 2.2.5 NSFLIGO.000003.

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

Web pages for the DCC give simple how-to's for document numbering, easy access to the latest on-line documents, and search capabilities for the DCC database. Take a look. . .

ACCOMPLISHMENTS

• The audit of the Hanford construction drawings continues.
• I finished a review of the 40-M science bulletin board  to check that all documents are numbered and available in the DCC electronically. I met with Gary to review those documents to ensure which ones should be a part of the DCC.  I then met  with Alan and let him know the numbers and the links that need to be updated.  A review of the Advanced Interferometer board was also completed.  The ASIS page is in process.  Due to the complexity of the ASIS web page, this will be a labor-intensive effort.  Not only will first level links be reviewed, but many documents contain links which in turn need to be considered and carried over with any distilling of the documents.
• We had a small respite due to a lighter load of electronic documents submitted for processing.
• Received 18 PAC meeting presentations today.  Some were on disks, many were email attachments, and a few in hard copy only.  Once these are all renamed and moved to dcc/OUT, processing will begin to transfer to docspublic and posting by Barbara on the web.  With the holidays, presentations made not be available on-line until the end of next week.

• progress continues on auditing of Hanford Construction drawings.  Many duplicate copies were found and we were able to discard them.
• electronic document submittals were light for this reporting period.

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>

• Have been transferring the available funds on contracts to the new Operations poetas.  Have made transfers on 18 contracts, and am currently working on transfers to two contracts.  There are still contracts which have balances to be moved.  As soon the final invoices for FY2000 are paid, the transfers can be made.
• Praxair nationwide will be having a rate increase in their LN2 prices.  LIGO is exempt from the price increase due to the fixed contract Ed Jasnow has in place.
• Awaiting for the new poeta to be set up for the $2 million Brunt construction award.
• Resolved the VWR order where the vendor had not given LLO the preferred pricing and terms for Caltech.  Adjustments were made by VWR to reflect Caltech's pricing agreement for the LLO facility.  Evidently, a new account had been created by the vendor for LLO which was not linked to the Caltech account.  The vendor has since linked the file and adjusted order #5434666 from $3,837.78 to $2,525.90.
• Pending:  UC Berkeley Lab still has not submitted the invoice for FY2000 and FY2001 payments.

• Worked with Sharon Patterson, in Al Horvath's office, and Dick Seligman, in Sponsored Research, to determine the proper classification of the anticipated costs for the new building at Livingston.  The decision is that this qualifies as a Capital Project even though it will be Government-Owned, which means that all costs incurred in construction of the building will be free of Indirect charges.  We have now submitted a "Fabricated Equipment Request Form," signed by Barry, to have a new account number assigned for these costs.
• Continued working with Dot and Ruth to determine the encumbrance balances to transfer to the new Operations accounts, once all invoices for the period prior to December 1 have been paid.
• Assisted Cindy in the preparation of the Integrated Report for the Data Analysis Group.  Completed the Integrated Report for MIT.

From: irena@ligo.caltech.edu (Irena Petrac)

Irena indicates that she has nothing significant to report.  However, we did assemble and send the package of materials that Vic Cook requested prior to approval of our FY 2001 funds.  This package included signature pages for the Annual Report, the MIT proposal, and a letter summarizing the status of the MIT faculty search.

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

As a result of a safety teleconference held  Thursday, December 14, security measures for laser use will be increased.  Studies will be performed to determine what security systems are available, while others look at what scientific research needs to be performed while the laser is on.  A decision on what system to use, and the procedures to accompany them, will be made shortly.

Advanced LIGO (Frey, Petrac)

Progress Period from 12.15 to 12.21

Accomplishments:

• Advanced LIGO Project Controls meeting on Monday the 18th.  Focus of meeting was status of Advanced LIGO MRE schedule and cost.
• Proposal plan development for Advanced LIGO MRE continues.
• Issued data requests for progress data as of 01.05.01 that are due Friday the 5th.
• Progress as of 12.15.00 was processed and posted.
• Review of cost and schedule data is now ongoing.
• Met with Mike to review cost and schedule data.  Data input into the cost book and schedule for AOS continues.
• Teleconferenced with David to review data and provide comments.  David will provide next iteration in January.
• Project Plan for the 40-Meter Lab Upgrade continues.
• Issued data requests for progress data as of 01.05.01 that are due Friday the 5th.
• Updated the plan with progress as of 12.15.00.  Posted to site.
• Continuing to push for a CDS procurement schedule.
• Waiting for comments from Alan and Mike regarding their requested changes that have been input.
• Planning process for the LASTI project at MIT is in progress.
• Issued data requests for progress data as of 01.05.01 that are due Friday the 5th.
• Progress as of 12.15.00 was processed and posted.
• Changes are pending from David per his discussions with GEO regarding the acceleration of prototype noise testing.
• Continue to test the Cost Book Tool.
• Development of the Advanced LIGO Project Controls Guidebook continues.
• 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

• Next weekly meeting will be Monday afternoon January 15th from 1 to 2pm regarding the planning and control of the Advanced LIGO MRE proposal development.
• Continue to pursue the planning and scheduling of the staging buildings currently being bid and designed for HLO and LLO.
• Advanced LIGO Proposal
• Will begin the effort of contacting each system leader to collect cost and schedule data.  Hope to begin importing schedules and cost ASAP. (AOS and IO ready for NSF review in January.)
• Will continue input of AOS and IO to the cost book and schedule.
• Next systems to be pursued are COC and PSL.
• Will being input of Operations Cost into the Cost Book Tool.
• Will Continue to update the LASTI Schedule with progress.
• Will post progress report by end of day Tuesday the 9th with progress as of 01.05.01.
• Will issue 23rd week of status data to teams by the 9th.
• Will continue to update the Advanced LIGO / Ops Proposal Plans and incorporate any changes.
• Will post progress report by end of day Tuesday the 9th with progress as of 01.05.01.
• Will issue 23rd week of status data to teams by the 9th.
• Will continue updating the 40 meter schedule and incorporate any changes.
• Will post progress report by end of day Tuesday the 9th with progress as of 01.05.01.
• Will issue 23rd week of status data to teams by the 9th.
• Will meet with Alan to review development of 40-Meter experiment tasks.
• Continue to test the Cost Book Tool.
• Cost Book Tool development continues.
• Will continue the development of the Advanced LIGO Project Web Site.
• Will continue the development of the Advanced LIGO Project Controls Guide Book.
• The 22nd, 25th, and 1st are observed holidays.
• Will be in the office on the 26th and out of the office on the 29th.

WBS 1.4.1.2   Project Controls (LIGO Construction)

Annual Report:  The Annual Report as of the end of November is due.  I hve requested inputs from the usual bunch and received material or promises from most.

The following Change Requests have been submitted:
 

Press for the latest Contingency Needs Projection.

From: Kris Duncan <kris@ligo.caltech.edu>

• No report this week.

• I completed the details for the monthly reports on the Construction and Advanced R&D accounts for November 2000 and emailed them to the Task managers.
• I received the MIT invoice for the month of November and as soon as I receive the TWONK (MIT allocation report) I will compile the data for payment.
• I continue to review the open encumbrances on the Construction and Advanced R&D accounts to see if the requested removals have been made. I will continue to review these accounts weekly until all encumbrances have been removed (Remember, I can only get these errors corrected if the Task Managers point them out to me).

 

 

The financial reports on the web provide supporting detail.

http://docuserv.ligo.caltech.edu/~fireport
http://docuserv.ligo.caltech.edu/~finance

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

Gerry Stapfer, Fred Asiri, and Ed Jasnow presented the proposed contract for the Livingston Staging Building to the NSF at Arlington, Virginia.  The presentation was well received, with lots of questions regarding the various building functions.   These questions were all answered satisfactorily, and the NSF representatives promised to get back to us quickly, since it was made clear that we did not want the construction of this building to interfere with detector commissioning activities.

A schedule has  been set for review of the proposed architect/engineers for the Hanford Support Building.  The two Washington architects will make presentations at Hanford on Wednesday, January 4, 2001, while the California architect will be reviewed at his offices on January 24, 2001.  A decision will be reached shortly after these reviews.

From: irena@ligo.caltech.edu (Irena Petrac)

• No report this week; see Operations Section above.

Quality/Safety (Tyler)

• No report this week.

LIGO Hanford Observatory (LHO) Operations (Raab)

No report received.

LIGO Livingston Observatory (LLO) Operations (Coles)

Almost all of this week's activity at LLO was focused on commissioning work reported by others. Peter Saulson departed this week for Syracuse as his sabbatical stay here ended and Rai Weiss has taken over the helm as commissioning leader at LLO. Dave Reitze also departed this week at the end of his 6 month stay.
 

LIGO was the destination for two educational outreach visits by elementary schools this week, each with about 75 children. Thanks to Bonnie Wascom, Jonathan Kern, and Sany Yoshida for making presentations and leading tours.
 

We will have about half of the resident staff of people on site next week on Wednesday-Friday and will have regular visits by staff to keep an eye on the vacuum system and cryogenics throughout the holidays. The Burns Security staff will provide 24 hour security coverage over the official Caltech holidays and they will receive Fedex and UPS shipments that are expected to arrive on Friday and Tuesday, so there should be no delays in delivery due to the extended Caltech holidays.
 

Detector/Technical Support (Whitcomb, Coyne)

See also the daily electronic logs for the installation and commissioning activities:
Hanford Detector Log
Livingston Detector Log

1.1 LHO INSTALLATION & COMMISSIONING

2km Commissioning

We implemented optical lever damping on the recycling mirror, so that now all 6 optics (RM, BS, 2 FMs, 2 ITMs) comprising the power recycling Michelson (PRM) are being damped via their optical levers. This last step further improved the stability of the power buildup in the PRM; residual power fluctuations are now 1-2%.

Full interferometer locking attempts continued to show 1/10's of second lock stretches, with a characteristic behavior that the error signal for the power recycling cavity length (lp) would become unphysically large part way through the stretch, and the power in the recycling cavity would go quickly to zero. This behavior seems to be caused by an offset in the recycling cavity length servo: even with just the PRM, we see that it does not lock at the point of maximum power buildup, but about 4 nm away from
this point. Tests showed that the offset is not electronic in origin, but is some optical effect that happens when the PRM resonates. The most likely theory, it seems, is that the large fraction of the input light that is not mode matched into the PRM (the input beam is matched for a hot' PRM, with the thermal lens effect in the ITMs) is causing the offset through incomplete detection of the beam. This should be easy to test.

Bill Kells
Have been doing analysis (modal models, FFT, etc) to get further understanding of the latest data of the lock states at LHO 2k. In particular the PRM is now locked much more stably, so that some accurate modeling can be revisited (begun last JUly). I have been running the FFT code on this,and found flaws in the previous runs, with then some new insights.

E2 Run

COS

• there is no measurable shift of the projected reticle pattern from the true optical centerline (within an experimental error of <100 microradians) at projection distances between 35 and 70 feet (which encompasses the range encountered in the 4K alignment)
• there is no measurable shift of the projected reticle pattern due to the focus adjustment
• the autocollimated reticle pattern, as observed in the eyepiece, shifts by >200 microradians, when the autocollimator is de-focussed, due to the paralax of the eyepiece

• the projected reticle pattern represents the true optical centerline and can be used to accurately align optical elements along the projection path.
• the autocollimator is calibrated only when the focus knob is set at infinity (which is checked by placing a retroreflector in front of the autocollimator and adjusting the focus until the reflected reticle pattern is as sharp as possible); therefore, any change in the apparent autocollimator alignment during de-focus should be disregarded since the instrument is not calibrated under those conditions.

4 km PSL

4 km SUS

• SOS Dewhitening board layout is complete and will be sent to the board house next week.
• System wiring diagrams have been started and should be complete next week.
• 8 more PD interface boards are complete and being tested.

4 km Initial Alignment

4 km IO Controls

1.2 LLO INSTALLATION & COMMISSIONING

Commissioning

A set of problems in the DAQ and control system have been fixed to give a system reliability of at least several days before rebooting is needed in some of the control computers. The  control systems for the large optics are damping well enough to cause no more than 10 microradian rms wander of the beam. The angular jitter of the beam emerging from the mode cleaner is 10 to 20 microradians rms, both values are smaller than the distant mirror size of 60 microradians. The problems remaining before opening to the beam tubes are the proper functioning of optical lever on ITMX and the careful alignment of the beam internal to the LVEA. A test we are asking for prior to opening the gate valves is the ability to autocollimate on ITMX and seeing the beam at the symmetric (reflected) port of the interferometer. The system has not yet passed this test.

ISC

PSL

I spent some time checking out a satellite amplifier (#31) that was reported as being noisy.  Following the test procedure I found that the satellite amplifier performed to spec. on the easy measurements and was consistent with the spec. on the harder measurements.  My conclusion was that it was okay.

2.0 Other Engineering and Scientific Activities

2.1 Design/Analysis/Fab

Optical Metrology

PSL (4 km LHO, 40 m, LASTI builds)

• Alignmed and optimized the transmission power on the new polished PMC.
• Received and unloaded the new MOPA Laser from Lightwave (S/N 118).

• We are currently 80% thru bench test on the 80 MHz VCO -- so far so good
• The Frequency Stabilization Servo has been stuffed to the extent that it can begin the bench test cycle -- We are still challenged with some missing components -- Testing to begin tomorrow or Friday.
• Concurrent with our objective to produce the PSL suite of modules, we have been doing background work to refine existing questionable circuitry.  Worth noting is the correction work on the PSL RFPD.  This module is common to many parts of the interferometer and has been recently misbehaving (see in last week's report http://docuserv.ligo.caltech.edu/~sanders/001214.htm#design).  The origins of the instability are traceable. We are currently working with Mike Zucker and testing and evaluating three related solutions.  We hope to have final resolution in the next few days.

ISC

Rich Abbott
Still working on the revised modecleaner servo card.  It appears that the noise specification may have been set too conservatively; Peter F. is reviewing the specification.  In production mode on various timing boards for shipment to the sites.

Modulated OSEM Design

• 5 stabilized oscillator satellite amps are being stuffed. They should be ready for test next week.
• One suppressed carrier satellite amp is stuffed and is being tested.

Data Acquisition

New OSEM Heads

Mark Barton
This week I got the thermal test of the OSEMs set up and debugged. I get almost exactly the same temperature differences as Miyoki (T960148) for the old OSEMs, e.g., 24 C above ambient at the coil for 200 mA of coil current. The new OSEM prototype can be plugged in at any time.

GDS

2.2 Issues Concerns

40 Meter Interferometer (Weinstein)

• Report on building rehab, from Fred Asiri:
• The roofers are working on the roof, working rather quickly. They may be done by this Friday; but certainly done by the following Friday. Steve Vass is watching them carefully, and is happy with what he sees. The roofer (or contractor) will replace the stained ceiling tile, when they're done.
• The contractor applied sealant to the wall between the IFO hall and CES. Steve is happy with the work.
• The floor tiling work is (almost?) done.
• One of the two Topaz power conditioners has arrived, Fred will find out when the second is expected.
• Fred will make a punch list this week or next, for all the (minor) remaining things: windows on doors, doorbell, etc.
• Jay Heefner reports that the racks and crates are expected in the 1st half of January. Electronics and computers should also come in in January.
• Steve Vass will arrange to have the floors washed before the racks come in (and get bolted to the floor), and the PSL enclosure gets installed. Then a thorough cleaning of all surfaces, and we return to cleanroom conditions by the 2nd half of January.
• STACIS status:
• We are running the STACIS systems 24/7, and checking them by hand daily.
• Larry Jones reports that his attempt at levelling all the support beams with shims was not as perfect as he'd hoped: "I had hoped to end up with vertical eccentricities at 0.052" max. and slopes between beam ends at no more than 0.66 mrad. Shimming magnitudes were as planned, but measurements afterwards show that vertical eccentricities topped off at 0.073" max. and slopes are at 1.53 mrad max." Good enough!
• Dennis Ugolini has measured a full set of spectra for all 12 pedestals, all 3 dof.
• The spectra were both encouraging and puzzling. the transfer function (TF) is betweeen -20db and -30db, throughout the measurement range from 0.5 Hz -> 100 Hz; and below -10db all the way down to 0.1 Hz. BUT the coherence was only good between 2-20 Hz.
• This sounds good, but it is NOT what we expected! There is NO peak at the passive resonance at ~15 Hz! There is NO 1/f^2 falloff above that!
• I just talked with Pete Nelson of TMC (formerly of JILA and MPQ). He said that they made changes to the STACIS servo loops, and the spectra look exactly as they should. I asked him to send me some kind of semi-analytic model, but he declined.
• So, we're ready to declare victory on STACIS!
• Details: web page, postscript, pdf.
• Larry reports that hardware for the 12m mode cleaner is being shipped from Allied and Caltech to LLO, to be cleaned, baked, and rga scanned by Rich Reisen and company.
• Larry, Steve, and Mike Smith will go over all the remaining 12m mode cleaner parts to be ordered and made.
• Steve Vass will assemble a detailed schedule for remaining work required on the 40m vacuum envelope. We hope to have everything installed, including the 12m mode cleaner but not the output optic chamber, by the end of March.
• We shipped the old recycling mirror to Gregg Harry at Syracuse for coating thermal noise studies, and he has received it.
• Dennis Ugolini and Ben Abbott reports that all remaining cabling for the remote vacuum valve controls is essentially done, and Dennis will be ready to test the controls and pump things down as soon as the vacuum envelope is ready for him. We hope to do the first tests in early January.
• Rick Karwoski and Ben Abbott continue to work on the H4K PSL boards, and will soon be ready to take stock on what's built, and remains to be built, for the 40m and LASTI PSLs.
• Mike Smith is working full time on 40m optical layout.
• Autocad drawing has been updated with core optics placed according to cavity length specs by AJW.
• Mike has re-routed all oplev beams to all core suspended optics.
• He has re-routed the initial pointing beam, and the positions of steering mirrors from MMT to IFO.
• He has laid out the input 12m input mode cleaner vacuum pipes, and MC mirrors and beams. He needs "final" numbers from Steve Vass and Larry Jones on new (10" OD) pieces.
• He has laid out a fixed monolithic output mode cleaner, with associated MMT, and routed input beam, output beam, and MC control beam out to a nearby table (which also has the symmetric port beam; crowded table!).
• We agreed that the beam from PSL to IFO should enter the input optic chamber through a different port; and we are considering rearranging the PSL table to accomadate the PSL, input optics, AND MC control optics.
• When Mike is finished with this job, we'll need to take stock off all the equipment we need (much of which we have from previous incarnation), where we need optical-quality viewports, where electrical feedthroughs are required, etc.

Thermal Noise Interferometer (Libbrecht)

Our laser's fast input response was measured to be 0.65 MHz/Volt, far lower
than the spec of ~4.5 MHz/Volt, as measured against the PSL reference
cavity at 5 Hz.  This supports our earlier conclusion that this laser has
problems, and was probably responsible for the poor performance of the mode
cleaner lock earlier.
 

With our successfully locked configuration, there is a servo crossover at
several kHz, which is the cause of some servo oscillations.  This results
in a ringing in the error signal. These oscillations do not destabilize the
lock.  Shanti created a matlab model of the servo which suggests that, for
higher servo gain (and hence noise suppression) we need a lead compensator
to increase the phase margin at the unity gain point in the mirror path.
Attempts to implement this are so far unsuccessful, and the high-frequency
ringing is still present.
 

LASTI (Zucker)

Received reworked legs & casters for HAM21 softwall cleanroom.
Continued work on details of BSC cleanroom and prepared CAD
drawings for review with installation planning team. Staged
additional support equipment received from MIT Center for
Space Research.

Data Analysis and Computing (Lazzarini)

Simulation and Modeling (Yamamoto)

Data analysis & Simulation

Biplab atteneded the GWDAW last week, and had a good communication with
data analysis group. We discussed about the effort toward the data
analysis support in the development of the simulation program. We
concluded it is a good time to start toward that, although we defered that
effort until now because of the lack of resources. From now on, Biplab
will work in coordinating and working to that the simulated data can be
used efficienctly for the data analysis as soon as possible. Biplab sent
out a mail summarizing the direction and works needed toward that goal.
We call for contributions from all interested people.

Simulation code

The parallelization effort is going. The code optimization of the
summation cavity improved the overall speed by factor of 2 for multi mode
simulation. The overall parallelization in the global execution
arrangements of module needs automated arranegements of module
execution taking into account the overhead of the use of thread. Coarse
level partitioning is close to be completed. A built-in profiler has been
improved to help analyzing the performance.

Alfi

The basic bug fixes are mostly done. Implementatioin of copy and paste and
improved interface for setting parameters for modules are in
progress. A new version of wxWindow is being tested and will be adopted
very soon. This will ease the compilation on Solaris 8.

12/14 Report:

Simulation engine

Tavio and Hiro worked on the parallelization of the simulation using threads.

* Hiro modified the power recycled michelson code to parallelize the execution
by doing the calculation of each sideband in separate threads. There is a big
overhead (see below) when using thread, ~100 microseconds per thread.
The speed gain for different number of modes simulated are as follows.
( number of sidebands = 5 )

scalar field      0.93
TEM n+m <= 1    1.3
TEM n+m <= 2    1.9
TEM n+m <= 3    2.2

* Tavio is working to parallerize the execution in the module level.
He completed and tested first draft of multithreaded e2e simulator, using execution

nodes (interface class between the module and execution scheduling)  which contain
only one module apiece.

Tavio used this to obtain timing estimates of ~20msec per-execution node thread
overhead.  This result conflicts with Hiro's seperate testing of mutex locking
which
suggests ~40-150msec to acquire a single mutex lock or signal.  Unfortunately
we can provide no explanation for this at present.  Thread overhead, and its
minimization,
will be discussed over the following week with Hiro and Sun microsystems.
Tavio sent an e-mail to SUN engineer whose name was given by Larry about this
issue.

Within the simulator a partitioning algorithm has been constructed which breaks
a given e2e module topology into the largest number of seperable pieces.  Hiro
briefed me on how delay and data-in modules are to be handled which allows the
partitioning to take place.  These partitioned pieces can be then be run
concurrently with a minimum of thread overhead.  Another routine is being
written which takes this partitioning and uses it to create pseudo ticklists
which contain large numbers of modules in them, within a small number of
execution nodes.

Alterations to dynamic queueing to decrease thread overhead were discussed.  In
addition to partitioning a topology, changes were discussed to eliminate
overhead within the 'first layer' (first set of execution nodes) of modules
within a topology by removing independent modules from the dynamic queue and
placing their dependents at the queue's beginning.

This next week, partitioning into execution nodes should be completed and way
to efficiently multithread these partitions ('node collapsing') will be sought
through graph-theory algorithms.  Discussion with SUN about thread overhead will
be ongoing.

**** Alfi

A new feature, Copy and Paste will be supported in the near future.
More stability and more compatibility are coming by the hardwork by Ed and Bruce.

**** e2e client and server

Soon, we will need to run e2e jobs on CACR multi CPU computers, because one
job time to get a meaningful result is getting longer and longer.  For this
situation,
we discussed about the client-server structure of the simulation engine.
Ed will summarize his thoughts about this structure and this will be discussed
in January.

LIGO Data Analysis System

Software Systems (Blackburn)

 

 

LDAS staff (KB & AL) met with the upper limit search group heads at the
LIGO Livingston Observatory during the GWDAW to discuss the expected LDAS
functionality needs anticipated by each group. Naturally, expectations
are very high for the given amount of manpower available within LDAS.
One of the first action items for the participants in this group will
be to identify a common set of functionality which overlaps most with
each groups needs. These areas of functionality will mostlikely be
centralized in the dataConditionAPI and will be higher/highest priority.
The group is planning to meet weekly to maintain tracking metrics and
respond to the requirements placed on LDAS by the upper limits searches.

LDAS staff (KB and AL) attended a one day GriPhyn Architecture Meeting
at Argonne National Laboratory. The focus of the meeting was on Computer
Science technologies and design. An example of management of virtual data
and its generation discussed by example. Much of the example was derived
from LIGO data and LDAS databases. The meeting was attended by roughly
12 members of the GriPhyn team from various high energy physics labs and
experiments, LIGO and Digital Sky. The meeting began to break up early
due to the familar weather patterns in the Chicago area this time of year.
Key members of the team will be drafting a schedule by mid January which
will cover GriPhyn activities for the next year.

KB went to ANL a day before the GriPhyn Architecture Meeting to meet with
the MPICH development team members Bill Gross and Rusty Lusk. The timing
of the visit was exceptional given that the MPICH group had just finalized
a MPI-2 development schedule. The MPICH team is expecting to have a formal
release of MPICH with full MPI-2 support in twoyears. The group expects that
no code from the current release of MPICH will survive the evolution to
MPI-2.
The current release will however be maintained for the interim between now
and this next major release. KB has asked to be included in development
releases
of MPICH/MPI-2 so that LDAS may better prepare for a version of the
wrapperAPI
based on MPI-2. Some of the unique area which the MPICH team expect to carry
out strong R&D programs in support of full MPI-2 support are in areas of MPI
thread safety. The user will be able to specify the degree of thread safety
needed by MPICH in balance with performance needs. Another area of active
R&D
is in the optimal balance of shared memory communications with TCP/IP
communications.
The MPICH team with also be working to develop a truely efficient and well
behaved collective communication set which exceeds the performance and the
reliability of point to point communications.

The new UDT class in the dataConditionAPI under went extensive testing
during
the past week and a half. Numerous bugs were identified and submitted to the
bug tracking system. These are being resolved in preparation for the next
release
of LDAS scheduled for the first week of January.

The wrapperAPI is in good standings know and has been tested with an
inspiral
search library provided by UWM. The mpiAPI still lacks support for multiple
instances of wrapperAPIs running on the BEOWULF. More effort will be
directed
towards getting this support in place by the next release.

The LDAS build process has experiences some minor problems with the
inclusion
of the mpiAPI and wrapperAPI which require MPI as part of the build process.
The problems are being sorted out and new documentation on the build process
added to the instructions. In addition the problems associated with the
auto-
mated building of documentation using the new PERCEPS2 are being fixed.

Charlton;

This week I have worked mostly on the XML "metaio" parser for Peter
Shawhan.

As of this writing the parser is close to being complete and I'm able to
extract rows of data from the sample XML file Peter gave me. The remaining
steps to perform are to:

* Extend the set of data types to the full set allowed in a table (this
just means adding a few more cases)

* Adding translation of special symbols such as &gt;

* Adding some user-interface utilities

* Rigorous testing

* Documentation

Bruce Sears:
* Ilog:
        - Continuing work associated with MIT installation, including
          needed upgrades at Hanford and Livingston.
* CDS Gnats:
        - Maintenance.  Fixed problem with set up of CDS Gnats tool.

 

Hardware Systems (Anderson)

 

 

Negotiated with Disk and Tape vendors for various 6-600TB storage systems.

Reactivated the LIGO HPSS web based table describing the LIGO data archive
contents, cf., http://www.srl.caltech.edu/personnel/sba/ligo/hpss,
and created E2 specific directories for copies of the trend frames.

CD-RW burner installed in the LDAS system at LHO.

Replacement hard drive installed in linuxbox1 in the LDAS system at LLO.

 

Data Analysis Activities

General Computing

MIT:
Nothing to report this week.

Livingston:
(Tom E.)
-We have been working on improving the SSH connection to the
CDS network.
-We have received and installed a new PC for running simulation
packages.
-We have re-installed our second scanner.
-We are working on upgrading our WWW server.

Hanford:
(Nothing to report at this time, check the Hanford report)
 

CIT:
(Suresh)
-Installed and tested latest version of teTeX (Latex) packages in sargas. It
will remain there for a few weeks for testing and will later be put on another
system after the compiled version for Solaris 2.6 has been tested out.
-Installed and tested Solaris Jumpstart installation over the network.
-Worked on some e-mail aliases.
-Finally moved home2 users directories to 36GB disk drive.
-Worked on Hiro's workstation to provide extra disk space accessible.

(Sam)
-Reworked a number of installation and support procedures. The documentation
will be put on the web after the Holiday.
-Doing another computer inventory and will be updating the s/w on a number of
the PC units.

(Lisa)
-I've got the new modem pool plugged in.  It's not properly configured yet, but
I can put in a user, dial-up and authenticate.
-I've been dealing with some fallout from last week's mail server problem.
Anyone on a unix workstation that is configured to use the local host will have
problems sending e-mail out until they specify acrux as their mail server.
Being
configured for localhost causes netscape to crash whenever an e-mail message is
sent out.  Mailer just gives an error message.
- Installed a new version of VXworks (Tornado 2.0) on luna.  This was just the
installation. The configuration will need some work by D. Barker. The software
automounts to /opt/CDS/wind86.
- Turned off the inet services on computers on the 113 subnet.
- Upgraded a couple more workstations on the 6th floor of Milliken to Solaris 8.
- Worked on a funky little emacs e-mail problem.  Emacs uses the local host in
it's From field instead of the mailhost.  This causes reply messages to bounce.
You can get around this by configuring a default "reply-to" in the .emacs file.
- Documented the scripts and procedure for doing manual monthly backups.

(Barbara)
-Working on a number of modifications for the WEB.
-Updating items for the DCC.

(Larry)
-Getting quotes for a number of purchases including new laptops.
-Setup a number of PC's to work with the new DCC printer. There will be limited
access to the printer and most jobs are to go through the DCC.
-Fixed a few calendar issues for both PC's and Unix boxes.
-Worked through a number of documentation issues of what is to be put on the
public web and what will be internal to LIGO. As things of this nature progress
we will be getting a few more people involved.

12/14 Report:

MIT:
Nothing to report.

Livingston:
Supporting the PAC meeting.

Hanford:
Nothing to report
 

CIT:
(Barbara)
-Have been tidying up after the DCC switch to the new server and Office
2000.  The overnight backup and update is running.  Investigated problems
tabbing through the subforms.  Cleveland and Linda are both linked directly
to the application on Antares.  Hopefully this will end the disk errors
messages Cleveland has been getting.
-Prepared sample outputs from the roster database for Gary and Liz to review.
-Caught up with web site changes -- LDAS, meetings, graduate students page.
Moved the 'LIGO listens to the universe' link back to the home page because
Marcia's book is generating renewed interest.

(Lisa)
- The base unit for the new modem pool has arrived.  I've started setting that
up.
- I spent a day going around to all of the unix workstations and turning off
sendmail.
- I spent some time trying to get the StorEdge on cdssol10 back up.  I am
waiting on a differential scsi card from sun before I can move forward on that.
- Thanks to some help from Sam, HP will be sending us a new scanner for the DCC.
- I contacted sendmail.com to get pricing for their commercial version of
sendmail.
- Worked with Rick Kossoris from across the street on figuring out why the
Citrix client will not install on Gina's pc.  We've determined that she needs a
fresh installation of Win98.  Sam is going to schedule getting that done.
- A hardware technician came out from sun and replaced the mother board in an
Ultra 10. It was a warranty replacement on one of the machines that came in this
summer.
- I'm still waiting to get a list of people from Rick Karwoski to finish up
creating a shared folder and mailing list for his lab.

(Sam)
-MMM, I actually came to work this week.  Wow!!!

-Preparing for the annual computer inventory
-Fixed three computers that have been needing it
-Worked on the DCC scanner
-Burned some CD's
-Backed up a home account to CD
-Worked on documentation

(Suresh)
-Installed AnswerBook2 Solaris documentation in server sargas.
-Worked on resolving recent spam mail attacks on mail server acrux. Turned off
sendmail daemons on most of the Sun Systems in order to avoid open relays.
Monitoring of mail services taking place. Assisted users for their problems on
several workstations as a consequence of these attacks. Got help from Minerva
Calderan from SRL to resolve this issue.
-Moved home2 directories to 36GB drive. But still mounting to that drive yet
to be accomplished.
-Completed the printer/plotter installation documentation and ready to be put
in Admin web pages.
-Created a PPP dial-up account for a Grad student.

(Larry)
-Spent time working with everyone involved on the spam attack. Minerva from SRL
was able to work through things while I was returning from vacation. Aaah, the
fun things that happen while one is away.
-Trying to get a differential SCSI card ordered from SUN. I have spent a number
of hours on tracking this down and once again they have lost the order. The
order has been placed again and presently we are waiting for a confirmation
number.
-Went over the groups work load and reprioritized a number of things on their
todo lists.
-Modified a few scripts for calendar support and e-mail account checking.
Cleaned out a number of e-mail accounts. We will be revamping the entire e-mail
alias file again in the near future.
-Getting a number of quotes for various items that will be ordered in the near
future.
-Modified some of the cabinet hardware in the computer server room.
 

LIGO II/Advanced R&D (Sanders)

Bill Kells
We have gone the next round with comparing MELODY to the FFT.
   Ray B. has agreed to put in some necessary improvements to
   upgrade melody. OUR goal now is to wrap this up, since we seem
   to have gotten a good education in melody by now. So its on to
   full thermal lensing situations.
 
I am pursuing a new measurement of the Rayleigh scattering in some
   of our Sapphire substrates, since we (Jordan) have an apparent
   suprisingly small value from  our work with coating birefringence.

From: Riccardo DeSalvo <desalvo@ligo.caltech.edu>

Akiteru, Riccardo and Alessandro,
* Taken receipt of parts for the two TAMA towers and shipped from Lucca
to Pasadena. Arrived at LAX with CV 730 of 17/12/00, clearing customs
this morning eta in Pasadena this afternoon 12 20th.  All parts were
checked and were OK according to drawings.
* Visited  SMA in Lyon with Helena Armandula and had a meeting with Jean
Marie Makowsky on the technical feasibility of making a ultra low loss
(less than 0.1 ppm) cryogenic mirror development plan over three years
and the feasibility of making LIGO II mirrors as well.   That meeting
was followed by a second meeting of Gary, Mackowsky, Jordan, Helena,
Garylinn and Riccardo in Pasadena on 12 10th.    Further studies and
proposals will follow.
* Visited Onera in Paris to see their Sapphire cutting capabilities.
* Visited a company in Torino that makes sapphire cutting heads,
designing a Sapphire and Quartz NC cutting machine.  Fred will follow up
with visits to Onera and Torino to get started.
 

Frederick, Jose Edwin
Advancing on creep cabling to new r/o hardware. Connected 3 LVDTs to DAQ
chain, problems in Read-out of data, debugging. Almost ready for dry run
. perhaps next week before Fred leaves for France.  Edwin will carry
over in his absence.
 

Akiteru:
Preparing TAMA SAS validation group.
Manpower:
Akiteru, Edwin, Chenyang and Hareem already on site.
Kenji should be available in December, Tatsuo will not be available
before January.
Riccardo and Virginio only partially available over the Christmas
vacations.
Fred will be in Paris/Torino,  will be back mid January.
Other volunteers?
 

Akiteru, Ric
Prepared calibration setup for new TAMA towers, parts ready to be
installed Wednesday 20th.
Prepared clean new linoleum floor the assembly space in the south lab.
Prepared and started the inertial damping tests on the old TAMA test
tower in the north lab.
 

Akiteru, Ric, Chenyang.
Installed old TAMA SAS in the inertial damping testing location.
 

Fred
Simulating ribbons and shaped flex joints for mirror suspensions.
Search for stress accumulation points and modal cross talk.
Simulated photoelasticity for ribbon prototype diagnostics.
Gathering technical and commercial data for ultra sound machining.
 
 

Alessandro
Front end electronics, mostly populated.
All coils almost finished.
Preparing electrodes for Chenyang, some problems with gold adhesion.
Working on thesis.
 

Chenyang
preparing the setup of the actuator calibration. Some detailed
calculation on it will be available on next Tuesday.
 

Seiji, Akiteru, Riccardo, Irena
Working on technical/financial issues for the TAMA SAS towers.
The technical solution is that we are going to "sell" them to NAO. See
catalogue in
 

http://www.ligo.caltech.edu/~citsas/documents/tama/catalog.pdf
 

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