1. Project Status
  2. Current Photos
  3. Risk Management
  4. Detailed Project Progress and Status
    1. LSST Program Office
    2. University of Washington
    3. Princeton University and University of California, Davis
    4. IPAC / California Institute of Technology
    5. SLAC / Stanford University
    6. NCSA / University of Illinois
    7. Current accomplishments:
    8. NOAO
    9. Current accomplishments:

Data Management Monthly Report

August 2015
 
Project Status


·  The Summer 2015 release was frozen for features, and is complete except for final metrics analysis runs, release notes, and other documentation, which should be complete by September 15. At LSST2015 we were able to process HSC engineering frames through single-frame processing, coaddition, and multiband coadd measurement; load the resulting catalogs into a database; and make the catalogs available via data access Web services and Firefly.  We were also able to run difference imaging.  We determined several scientific and computational Key Performance Metrics.
·  We demonstrated Qserv stability and performance well beyond planned KPMs. Highlights demonstrated:

o   0.9 sec average low volume query response (planned KPM: no longer than 10 sec)
o   <1h for high volume queries (planned KPM: no longer than 12h)
o   79 low volume queries in 24h (planned KPM: 10 high volume queries in 24h)
o   query concurrency: 100 low volume and 10 high volume queries (planned KPM: 50+5)
o   continuous running under heavy load for over 48 hours (planned KPM: 24h)
·   The FY15 infrastructure procurement is being prepared and will implement a near-term existing hardware refresh. We are revisiting the overall infrastructure needs specified by the LSST developers (e.g., SUI and Qserv teams), for the FY16 Infrastructure Acquisition Strategy document.
·  We continued work on the Data Management Development Roadmap (LDM-240), improving Milestone and Key Performance Metric descriptions and analyzing resource loading.
·  The MOA (Networks/PIs) has been signed by all Brazilian parties and is being routed for signature by the U.S. parties.
·  The DM Project Scientist continued the verification and understanding of NEO detection efficiency and effectiveness.
·  The DM Project Engineer and other DM team members continued supporting the Technical Operations Working Group (TOWG), and LSST System Engineering activities, including reviews of ICDs and simulations requirements and plans.
·  The DM SQuaRE team gathered input for verification datasets and quality metrics and presented the plan at Bremerton.
·  Recruiting and hiring activities continued across all DM institutions. Twenty-eight positions have been filled to date since the MREFC award, while 9 positions are currently open.   Two new hires started in Tucson and one more is scheduled to start in September:
o  David Nidever, DM SQuaRE Scientist
o   Angelo Fausti, DM SQuaRE Software Developer (1 year assignment)
o  JMatt Peterson, DM SQuaRE Software Developer
·  Work continued on standing up an OpenStack-based development environment at NCSA.
·  The LSST Information Security Officer (ISO) presented a talk on LSST Cyber Security for the NSF Cyber Security Summit in Washington D.C.

 

Back to top



Current Photos
 

 
HSC riy-multiband image processed by Lauren MacArthur with LSST stack

 
High Cadence Transient Survey DECam images Subtracted with LSST Stack by Francisco Forster
 
 

Back to top



Risk Management
 
The DM Risk Register was reviewed in the monthly process. No new risks were added and no significant changes to existing risk exposure were made.

 

Back to top



Detailed Project Progress and Status


LSST Program Office

DM Project Management and Control

Current accomplishments:

The DM Project Manager:


·  Continued work on the Data Management Development Roadmap (LDM-240), improving Milestone and Key Performance Metric descriptions and analyzing resource loading.
·  Continued the process of developing the FY16 budget so that we can prepare FY16 sub award contract amendments.
·  Continued development of the Brazil MOA (Networks/PIs).  All Brazilian parties have signed and the MOA is being routed for U.S. signatures.
·  Attended and led sessions at the LSST 2015 Meeting in Bremerton.
·  Recruiting and hiring activities continued across all DM institutions. Twenty-eight positions have been filled to date since the MREFC award, while 9 positions are currently open. Two new hires started in Tucson and one more is scheduled to start in September:
o   David Nidever, DM SQuaRE Scientist
o   Angelo Fausti, DM SQuaRE Software Developer (1 year assignment)
o  JMatt Peterson, DM SQuaRE Software Developer
 

Planned activities:

The DM Project Manager will:


·   Brief the AMCL on DM Status, including accomplishments, plans and earned value, and important issues.
·   Continue recruiting and hiring, prepare for arrival of new hires.
·   Continue work with NCSA on FY15 infrastructure procurements and FY16 Infrastructure Acquisition Strategy.
·   Complete the execution Memorandum of Agreement with Brazil.
·   Create ENews article regarding Networks.
·   Participate in Export Controls training and program development.
·   Participate in Intra-Project Communications activities.
·   Complete Sub award FY16 budgets and contract amendments.

DM Science

Current accomplishments:
The DM Project Scientist August activities continued to be focused on the study of the efficiency of LSST as a detector of NEOs, as well as participation in a series of meetings including LSST 2015. Specific work includes:
 
A 10-day participation at the International Astronomical Union (IAU) meeting in Honolulu, HI. While there, the DM Project Scientist worked with a number of members of the PanSTARRS PS1 project to understand the remaining uncertainties about the difficulties that PS1 encountered while attempting to employ an LSST-like asteroid discovery strategy. They came away reassured that PS1 difficulties were largely PS1 specific (experimental camera, difficult optical system, and under-resourced software). They identified only one potentially serious issue: the case where 2+2+2 observation strategy is not sufficient to filter false from real tracks in cases of extreme false positive densities. LSST simulations show that even in that case, the mitigation of switching to 2+2+2+2 would have only a minor impact to the overall discovery efficiency (few percent).
 
The DM Project Scientist presented a talk at the IAU Asteroids Symposium on LSST strategy for asteroids and participated in a discussion panel, gave a talk at the NSF booth about the LSST data management system, participated in LSST 2015, including giving a talk for the science collaborations on the DM system, and attended the Boutique surveys workshop at Caltech.
 
With the DM Project Engineer, the DM Project Scientist prepared the overall plan for finalization of Summer 2015 development cycle and planning of Winter 2016. They stressed the prioritization of design documents and upwards/downwards traceability as a prerequisite for a well understood development plan and meaningful resourcing.
 
The DM Project Scientist continues to lead the asteroid discovery efficiency study (image differencing, MOPS simulation, completeness simulations). He requested and received  preliminary approval for an additional resource for the MOPS simulation aspect of the study.
 
John Lurie continued to work on updating our understanding of the structure of the Milky Way disk based on precursor survey data. The density model is close to being finished and the focus right now is on understanding the asymmetries. This work will lead to a better estimate for the number of stars for the sizing model; initial galfast runs will be performed on the moya GPU server.
 
Colin Slater, the DM Data Systems Fellow, has arrived and already participated in LSST2015. Colin will begin by assisting in the image differencing aspects of the asteroid study.
 
Planned activities:
 
The DM Project Scientist expects to remain focused on the NEO studies.
 
DM System Engineering

Current accomplishments:

Accomplishments by the DM System Architect and Deputy System Architect include:

 
· Agreed on necessity for design document updates
· Compared current plan with Data Products Definition
· Started to define DM requirements for simulations
· Completed Data Butler notes and turned over to Nate Pease
· Defined end-to-end DM system components
· Drew Data Release sequence diagram for Operations
· Suggested potential contributions for IN2P3
· Presented at and ran sessions at DM All Hands meeting
· Estimated storage and compute needs for FY16
· Began organizing DM Boot Camp
· Assisted with Summer2015 release process
· Attended LSST 2015 Meeting at Bremerton
· Prepared for End-to-End testing at Bremerton
· Prepared two presentations on Python 3 and a pythonic stack API for Bremerton.
· Wrote script to convert JIRA Key Performance Metric information into a table or CSV file.
· Investigated Log4CXX issues on Mac OS X (DM-1662) and start looking at log package.
· Fixed race condition in “base” package (DM-3517)
The DM System Interfaces Scientist accomplished:
· Organized and led AHM session on refinement of LSE-68 interface for camera image data.
· Meetings with Telescope Scientist on LSE-75 DM-TCS interface; responded to memo with questions about PSF/WCS reporting and calibration.
· Began detailed review of LSE-70 and LSE-209 (OCS interface documents to be sent to CCB in the near future).
· Began detailed review of LSE-78 network design document to ensure it consistently reflects recent system design updates.
· Supported NCSA group in familiarization with implications of system interfaces for its work, especially for EFD replication.
· Continued working with Matias Carrasco Kind at NCSA to support his joining the project and working on Task system enhancements.
Planned activities:
The DM System Architect and Deputy System Architect will:
 
       
 
· Complete planning for DM Boot Camp
· Begin updating requirements and design documents
· Redefine release process and versioning
· Assist with 2016 hardware procurement planning
· Meet with DESC spokesperson to synchronize roadmaps
· Prepare migration documents for Python3 support (RFC-60).
· Continue to support the log package.
The DM System Interfaces Scientist will:
· Complete LSE-78 review so that Ron Lambert can produce an updated version in October.
· Complete DM review of LSE-70 and LSE-209.
· Rewrite LSE-72 to follow new LSE-70 and LSE-209 changes.  Prepare for October CCB meeting, assuming LSE-70 and LSE-209 are ready from the OCS group by then.
· Continue work with Telescope Scientist on LSE-75.  Evaluate whether Apps-related questions can be resolved in time for October CCB.
· Visit Tucson with Matias Carrasco Kind to advance Task project and align with QA group needs.
DM Science Quality and Reliability Engineering (SQuaRE)
Current Accomplishments:
02C.01.02 SDQA 
Planning activities for Winter 2016 started in August as did co-ordination for the Summer 2015 software release and cycle close-out. Time was also spent on a recruitment.
SQuaRE prepared for and attended the LSST 2015 meeting, in the midst of onboarding three new members of staff including two who started in August: David Nidever joined SQuaRE in August as its Lead Scientist; Angelo Fausti joined SQuaRE half-time as part of a collaboration with LINEA of Brazil.
Jonathan Sick gave presentations on the Discourse web forum platform and SQuaRE’s thoughts on updating the code documentation infrastructure.
David Nidever gave presentations and gathered input for verification datasets and quality metrics.
Frossie Economou gathered input for developer support and gave two public talks.
Planned Activities:
02C.01.02 Science Quality and Reliability Engineering
· v11_0 release engineering
· W16 planning
· Verification dataset roadmap
02C.01.02.03
· Discourse production deployment
DM Applications, Middleware, and Infrastructure
Current accomplishments:
· Refer to by institution reports below
Planned activities:
· Refer to by institution reports below


University of Washington

Current accomplishments:

02C.03.00 Alert Production Management Engineering and Integration

The UW DM team attended the Bremerton all-hands meeting.  Yusra AlSayyad (YA) worked with Ian Sullivan (IS; a new team member) to help him come up to speed with LSST.  This included installing the stack as well as introducing him to the LSST standard reading materials.

02C.03.05 Application Framework for Exposures

Russell Owen (RO) did a lot of work to clean up at the end of the cycle (DM-3491, DM-3454, DM-3347).  He also finished up the porting of aperture correction code to the LSST side (DM-3460, DM-2429, DM-833)

02C.03.08 Astrometric Calibration Pipeline

RO also worked to clean up some technical debt accumulated during the astrometry task refactor (DM-3453, DM-3455, DM-2939).

Planned activities:
02C.03.00 Alert production Management, Engineering and Integration
 
The UW DM team will welcome another new member, David Reiss (DR).  This will involve tutorials by YA and RO in code reviews, coding workflow, etc.  There is a bootcamp planned for the end of September.  YA, SK, and RO will act as tutors and help by giving talks on various aspects of the stack and development environment.
 
02C.03.04 Image Differencing Pipeline
 
IS and DR will work together to bring the team up to speed on the issue of differential chromatic refraction in the context of difference imaging.  YA will modernize the diffimTask so that it is easier to run.  This will also make it easier to develop.
 
02C.03.05 Application Framework for Exposures
 
RO will largely be working to remove technical debt accrued when the aperture correction code was brought over from HSC.

 


Princeton University and University of California, Davis
This report covers work carried out in FM11 of FY15 in the Data Release Production group (staff at Princeton plus Price and Gee working remotely).
 
Current accomplishments:
 
02C.04.00 Data Release Production Management Engineering and Integration
 
Work this month centered around the LSST2015 meeting in Bremerton.  A primary goal for the Data Management project as a whole is to use this meeting to construct as complete as possible an end-to-end data release pipeline, and then use it both to demonstrate the current state of the integrated DM system and to determine the key performance metrics (KPMs) to accompany the summer 2015 stack release. This drove management considerations through August at a number of levels:


·  Planning and coordination effort (primarily Bosch, some Swinbank) was devoted to organizing the work which needed to be accomplished both in advance of and at LSST2015 in order to facilitate achieving these goals; 

·  Hyper Suprime-Cam (HSC) engineering data was used for the LSST2015 effort; effort was required to identify, transfer and make this data available to the community; MacArthur focused on stack integration rather than new development or HSC mergers in advance of the meeting; 

·  That development which was performed through August was targeted to have the greatest possible impact in terms of the functionality available in Bremerton. Swinbank prepared a draft budget for FY16 which was discussed at the Bremerton meeting. 
In addition, Bosch and Swinbank continued to iterate on the long range planning in terms of milestones and meta-epics defined in the JIRA-DLP project and to plan the activities to be performed in W16. A first version of the W16 plan was provided to the Project Scientist and Project Engineer by the end of the month for comments. 



Hiring
 
Lupton, Bosch and Swinbank reviewed the applications for the Scientific Software Developer job which was advertised through June and July. We drew up a shortlist and scheduled interviews for the week beginning 31 August.
 
Fisher-Levine spent one week at Princeton and started getting up to speed with the project. He will not officially start working (at the 50% level) until October.
 
Our advertisement for one or more "postdoctoral positions in software" is live on the Princeton University website and will be on the AAS Job Register in September and October. We expect to begin reviewing candidates in November.
 
02C.04.01 Application Framework for Catalogs
When initializing an algorithm that refers to a particular slot, we resolve the target of the slot and refer to that instead. That means that if the slot definition is changed after measurement has been performed, we are still pointing to the correct information. However, if the algorithm is initialized before the slot it depends on, this resolution could not take place and "circular" aliases could result. We now explicitly check for and throw an error in this case. [DM-3400]
 
The obs_subaru package was extensively renovated. This included incorporating distortion estimation code which was previously packaged separately and which needed extensive modernization, as well as a number of bug fixes to the obs_subaru package it- self. This results in HSC now being fully supported by the LSST stack, a major milestone in terms of stack testing. An HSC-based integration test was produced and provided to the SQuaRE team for use in continuous integration. [DM-1794, DM-3403, DM-3493, DM-3663].
 
Testing at LSST2015 with HSC data (made possible by the above) unearthed some problems with the astrometric fitting routines which were corrected. [DM-3492, DM- 3442, DM-3209].
 
02C.04.02 Calibration Products Pipeline
 
No work was carried out in this WBS element in this month.
 
02C.04.03 PSF Estimation
 
No work was carried out in this WBS element in this month.
 
02C.04.04 Image Coaddition Pipeline
 
No work was carried out in this WBS element in this month.
 
02C.04.05 Object Detection and Deblending
 
No work was carried out in this WBS element in this month.
 
02C.04.06 Object Characterization Pipeline
 
The order of arguments to the forced measurement task was reversed, so that it takes a source catalog followed by an Exposure. This brings it into line with the single frame measurement interface. [DM-3459]
 
It is now possible to use the PSFEx PSF determination package as a measurement extension within the LSST stack. This is a significant step forward over the earlier PSF determination algorithms used in the stack. [DM-2961, DM-3461, DM-3463].
 
The measurement transformation framework was adjusted so that rather than throwing an exception when attempting to convert a negative flux into a magnitude, it simply returns NaN. This enables it to be used reliably as part of the database ingest system. [DM-3483]
 
A large number of optimizations and improvements to the way that large blended sources are handled were ported from HSC. This work included both code tuning to improvement performance and better logic as to which sources can be usefully measured. These provide significant enhancements to both processing time and memory usage, and have proved essential for processing large volumes of HSC data. [DM-2914]
 
A number of minor changes and documentation improvements were made to the Coad- dBase, AssembleCoadd, CoaddInputRecorder and MakeCoaddTempExp tasks. These brought the structure of the code better into line with the state-of-the-art development on Hyper Suprime-Cam, fixing a number of minor bugs in the process. This code is now ready to serve as the base for the the "safe coadd clipping" routines to be ported from HSC. [DM-2980]
 
Work on estimating the computational requirements for galaxy shear fitting continued with a focus on the infrastructure required to perform the tests. Test driver scripts were developed and the total amount of compute time needed was estimated. [DM-2663, DM-2664, DM-3367]
 
A number of bug fixes and minor enhancements to the CModel measurement code were ported from Hyper Suprime-Cam. [DM-2977]
 
A number of bugs were corrected in the stack, notably including:

·  Incorrect weighting was being used in photometric calibration, as the data was weighted by its error rather than the inverse. [DM-2423] 

·  An unmasked NaN value in the could cause the overscan region of the detector to be incorrectly fitted. [DM-2923] 

·  An incorrect exception name was being used in the deblender error handling routines. [DM-3402] 


 
Planned activities:
 
02C.04.00 Data Release Production Management Engineering and Integration 

 
The W16 plan will be finalized, resources allocated, loaded into JIRA, and made available for import to PMCS.
 
We expect the ongoing long-term planning project, and in particular updates to key project documentation, to continue to require most of the time available from Bosch and Swinbank through September and October.
 
02C.04.01 Application Framework for Catalogs
 
Continued focus on merging work from HSC with particular emphasis on the parallelization middleware.
 
02C.04.02 Calibration Products Pipeline
 
To date, detailed planning for the Calibration Products Pipeline has been on hold pending the availability of manpower. Fisher-Levine is expected to be available at the 50% level from October onward. Through September, Fisher-Levine, Lupton and Swinbank will start developing a more detailed plan for the work to be performed.
 
02C.04.03 PSF Estimation
 
No work is planned for this WBS element in this month.
 
02C.04.04 Image Coaddition Pipeline
 
No work is planned for this WBS element in this month.
 
02C.04.05 Object Detection and Deblending
 
No work is planned for this WBS element in this month.
 
02C.04.06 Object Characterization Pipeline
 
Continued focus on merging work from HSC with a particular focus on comparing the HSC and LSST pipelines for consistency, identifying and fixing points where they diverge. In addition, work remains to be done to calculate all the key performance metrics for the Summer 2015 release.
 

 


IPAC / California Institute of Technology

Current accomplishments:

02C.05.00  Science User Interface Management, Engineering, and Integration

·   Summer vacation continued.  Trey: 8 days;  Tatiana: 2 days; Loi Ly: 5 days; Lijun 5 days; David Ciaridi: 5 daysk; Xiuqin 2 days; Gregory: 10 days

·   Worked with IPAC IRSA group on collaboration in Firefly development, plan and schedule coordination, common system for issue tracking.

·   Finished the first round LDM-240 road map to Jira project DLP, add KPM and meta-epics.

·   Finished FY16 budget, and work planning for winter 16

·   Everyone attended the LSST 2015 all hands meeting

·   David Ciardi submitted a proposal to the LSST Corporation Science Enabling Committee: "Towards a Better Understanding of the Requirements of a Complete Transient Event Alert Notification System: Supporting Transient Science in the Era of LSST". Abstract: We propose to work with the astronomical community to address the scientific and technical challenges associated with the formidable LSST alert event rate and the ability of the community to effectively characterize, classify, and follow-up the alerts. We will engage the community by holding an open workshop dedicated to developing requirements and standardizing the interfaces for a service which is capable of handling the LSST event rate and which facilitates discussion of, and collaborations about, the detected transients. In particular, we will focus on supporting the annotation and tracking of the follow-up observations which are required to maximize the scientific value of the event stream.

02C.05.01   Basic Archive Access Tools

·   Continued the discussion of data access APIs with SLAC group if needed

·   Setup the access to NCSA hosts to test out the access to APIs and database

·   Finished the DM end-to-end exercise.

·   Developed a simple package and deployment of Firefly to make developers’s life much easier to start using Firefly. Please see   https://github.com/lsst/firefly/releases/

·   Start preparing for Firefly code conversion: from GWT based Java code to React based JavaScript code

 02C.05.02   Data Analysis and Visualization Tools

·   Continue the development on the JavaScript APIs and Python APIs for Firefly visualization components.

·   Worked on the design and implementation of overlay masks on the primary image.

·   Continued to work with Camera group to support their development.

·   Reafactor the Firefly code (ZScale.java) to make it easier to read and maintain

·   New functions in Firefly:  FITS cube reader at the server side, binary FITS table upload

 02C.05.05 User workspace

·   Finished  the user workspace concept document. 

·   Started discussing the authorization and authentication system with other DM groups.

Planned activities:

02C.05.00  Science User Interface Management Engineering and Integration


·  Summer vacation continues.  Xiuqin 5 days
·  Work with IPAC IRSA group on collaboration in Firefly development, plan and schedule coordination, common system for issue tracking.
·  Second round update in Jira project DLP, refine milestones,  KPM and meta-epics.
·  Finish Winter 16 resource loading

 
02C.05.01   Basic Archive Access Tools

·  Continue the discussion of data access APIs with SLAC group if needed
·  Start the bi-weekly discussion with SLAC, NCSA, SQUARE on authorization and authentication
·  Resolve issues found in the DM end-to-end exercise.
·  Start design of the new Firefly core system using React and FLUX framework
·  Start working on the SUI/T design and design document

 
 
 02C.05.02   Data Analysis and Visualization Tools

·  Continue the development on the JavaScript APIs and Python APIs for Firefly visualization components.
·  Finish the  implementation of overlay masks on the primary image.
·  Continue to work with Camera group to support their development.
·  Bug fix in Firefly for coordinate grid overlay on images

 
 

 


SLAC / Stanford University

Current accomplishments:

Highlights:

  
  
· Demonstrated Qserv stability and performance well beyond planned KPIs. Few highlights: demonstrated:
· 0.9 sec average low volume query response (planned KPI: no longer than 10 sec)
· <1h for high volume queries (planned KPI: no longer than 12h)
· 79 low volume queries in 24h (planned KPI: 10 high volume queries in 24h)
· query concurrency: 100 low volume and 10 high volume queries (planned KPI: 50+5)
· continuous running under heavy load for over 48 hours (planned KPI: 24h)
· Finished all planned S15 work and delivered stable Qserv release
02C.06.00 Science Data Archive and Application Services Management Engineering and Integration 
· Coordinated August Sprint for the Data Access Team
· Coordinated S15-cycle Qserv release
· Coordinated Data Access portion of the end-to-end testing
· Organized weekly Qserv and Data Access meetings
· Finalized W16 plan, came up with S16 plan
· Bremerton meeting – whole team attended
· Interviews:
o one on-site interview: Gregory Pogosiants
02C.06.01.01 Catalogs, Alerts and Metadata
· Completed ingest code (DM-210)
02C.06.02.01 Data Access Client Framework
· Completed Butler documentation
02C.06.02.02 Web Services
· Fixed problems related to default_engine
· Set up Webserv service for SUI tests
02C.06.02.03 Query Services
· Continued running large scale tests, measured and documented performance / KPIs: https://confluence.lsstcorp.org/display/DM/S15+Large+Scale+Tests
· Finished data distribution and replica management prototype
· Improved multi-node integration tests: added mysql-based test and result comparing
· Release highlights:
· Further improvements to reliability
· Implemented MySQL-based KVInterface for CSS
· Improved compatibility with webserv
· Code improvements:
· DM-3450: Fix cluster install procedure and improve docker support
· DM-3456: Fix problems with talking from webserv to qserv
· DM-3555: Ignore "SELECT tx_isolation" queries
· DM-3358: Add mysql-based test to multi-node integration test
· DM-3360: Minor cleanup related to max_hostname
· DM-1662: Make qserv dependencies build on OS X with clang
· DM-3450: Tweaks to configurations discovered during S15 tests
· New features:
· DM-3163: CSS/QMeta interaction in czar
· DM-3161: Implement MySQL-based KVInterface
· Bug fixes:
· DM-3522: Releasing un-acquired resources bug
· DM-2930: Fix problem with Qserv related to restarting mysql
· DM-3618: Fix bug related to restarting xrootd in wmgr
Planned activities:
02C.06.00 Science Data Archive and Application Services Management Engineering and Integration
· Organize weekly Qserv and Data Access meetings
· Search for candidate for the remaining open position
· Start updating Data Access design documents and connect them with long term planning epics
· Define and model detailed Qserv KPIs
02C.06.01.01 Catalogs, Alerts and Metadata
· N/A
02C.06.01.02 Image and File Archive
· N/A
02C.06.02.01 Data Access Client Framework
· Rework db module (switch to SqlAlchemy
02C.06.02.02 Web Services
· Add unit test to Webserv
· Migrate Webserv to the new db module
02C.06.02.03 Query Services
· Begin work on query cancellation (rework executor)
· Begin work on building data distribution system
· Add CSS structure for table and database deletion
· Migrate Qserv to the reworked db module
· Unify KVInterface between python and C++
· Finish integrating pipelines with mysql and qserv
· Improvements to butler (type aliases, registry-free repositories)
· Improve query coverage (fix broken IN, handle “where objectId between”)
· Refactoring for OS X
02C.06.02.04 Image Services
· Make location of images more flexible
02C.06.02.05 Catalog Services
· N/A


NCSA / University of Illinois


Current accomplishments:
02C.07.00 Processing Control and Site Infrastructure Management
 
Management activities in August included assessing the status of S15 and closing out all our activities, reviewing invoices from the past six months, developing a plan for W16 activities, preparing an estimate for NCSA’s FY16 labor budget, reviewing the proposal for FY15 hardware purchasing, interviewing candidates for NCSA’s two open positions (research programmer and research scientist) and reviewing our hiring strategy, hiring a candidate for the research programmer position, and attending the LSST2015 Conference in Bremerton.
 
ISO work (01C.01.01)
(DM-3378, 3482) – In August the ISO presented his talk on LSST Cyber Security for the NSF Cyber Security Summit in Washington D.C. The presentation reviewed covered challenges and successes of LSST’s security program, the LSST security plan, data security, user access, and the SCADA enclave/observatory site security.
(DM-3382) – He also created a template document for handling and responding to security incidents.
 
02C.07.01 Processing Control
 
TOWG
(DM-3525, 3526) – We continued our work with the Technical Operations Working Group in August. Following the ITIL model, we reorganized and incorporated uses cases for common IT operations into the Enterprise Architect tool. We prepared for presenting the ITIL model and IT use cases at the LSST2015 Operations Concepts breakout session.
 
Data Management Control System: Alert Production
(DM-3418) – As an action item from the July 2015 CCS-DAQ-OCS-DM Workshop, Steve reviewed LSE-70 (Observatory Control System Communication Architecture and Protocol) and LSE-209 (Software Component to OCS Interface Control Document) and provided feedback on needed changes on the Workshop Confluence page here: https://confluence.lsstcorp.org/display/SYSENG/2015+July+08-10+CCS-DAQ-OCS-DM+Workshop+IV . One suggestion was to expand the text in LSE-209 to give context to the diagrams.
 
(DM-2263) – Investigation of reducing HTCondor job startup times continued in August. The request by Miron Livny after his visit in July was to generate timing information about job submission to job start latency. To get job execution timings and statistics, Steve worked on adding support in ctrl_stats for parsing user logs written by HTCondor for each individual job (DM-3444). Various tests were run in groups of complete “visits” (189 workers each), comparing timings with claimed vs. unclaimed slots and generating timing statistics. The results were sent to the HTCondor team to get their input.
 
Data Management Control System: Design specification of L1 System
(DM-3215) – A major focus in August was continuing the work begun in July on refining the design specifications of the Level 1 System. A series of pages were started in Confluence ( https://confluence.lsstcorp.org/display/~petravick/Breakdown+of+the+to-be+facility ) that break down the system functionally and physically, including hierarchical diagrams of the L1 system and descriptions of the Engineering and Facility Database data ingestion and use. System and design specifications were collected from LDM-230, LSE-61, and LDM-148.
 
Data Management Control System: Processing DECam data with the LSST stack
Hsin-Fang continued to work on learning and running the LSST stack and working with DECam data. She started working with obs_decam and the development workflow, including working through build issues (DM-3429) and updating the bit masks in DecamMapper from the early Community Pipeline version to DESDM’s Year 1 version (DM-3538). She also began working with DECam data, collecting a testbed of DES public data from the NOAO archive (available at http://uofi.box.com/testdata-decam ) (DM-3421). Finally, she spent one week attending the LSST2015 meeting and participated in some of the DM developer breakout sessions that dealt with the LSST stack.
 
Data Management Control System: Extending the Process Execution Framework
(DM-3120) – Matias continued working on the Processing Execution Framework prototype, modifying/refactoring “Super-Task” in pipe_base, the base package for pipeline tasks, and generating test tasks to demonstrate the process.
 
02C.07.02 Infrastructure Services
 
Sizing Model
(DM-2217) – The hardware procurement contract amendment fully executed on July 31st. Per the contract agreement, NCSA completed the costing refresh of the sizing model. LDM-143 and LDM-144 were updated with current market prices on specified or comparable technologies. Updated versions were checked into DocuShare and an RFC was filed ( https://jira.lsstcorp.org/browse/RFC-90 ) to approve these updates. It is worth noting that an activity like this has not been done in at least two years, and we anticipate the annual technology refresh due in February will take a considerable amount of time.
 
System Administration Services: Configuration Management (Puppet)
(DM-2237) – As of August we have tested all base configuration Puppet modules, including otp, crashplan, nfs_client, selinux, timezone, and firewall modules. At the end of the month all but the firewall module was released to the developer servers; this final base configuration module will be pushed out during regularly scheduled maintenance in September.
 
02C.07.03 Environment and Tools
 
Development environment: OpenStack
(DM-3377) – Following feedback/issue reports from the nebula early users, the “evaluation” nebula OpenStack system was decommissioned. Work proceeded towards set up a new “production” system; the nebula admin team internally stood up a series of follow-up evaluation systems and Greg worked with them to test progress on resolving issues that LSST users presented.
 
02C.07.04 Site Infrastructure
 
With the finalization of the procurement contract at the end of July NCSA spent most of August working on the design specification of the production system (noted above), updating the sizing model, and beginning to plan FY15 procurements by reviewing existing infrastructure targeted for replacement and considering near-term needs to support LSST developers. Administration of the development cluster at NCSA included setting up three accounts for new users, fixing some issues with the SUI development server, debugging a BuildBot issue with the sqlite database, and scheduled monthly maintenance.
 
Planned activities:
 
02C.07.00 Processing Control and Site Infrastructure Management
 
Management plans for September include:


·   finalizing the FY16 labor budget for the contract amendment
·   finalizing the W16 planned activities with resource-loading
·   finalizing the FY15 hardware purchasing plan, a one-off proposal to spend down the remaining FY15 budget
·   drafting the FY16 Annual Acquisition Strategy Document (collecting requirements from developers and considering a prototype production system)
·   on-boarding our new research programmer at the very end of the month
·   reviewing candidates for the two open positions at NCSA and coordinating with other NCSA groups to understand staffing/institutional support needs for FY16 and beyond

We will work extensively on specifying the design of the production, functionally and physically breaking down the Level 1 and Level 2 systems.
We will continue to develop use cases for operations and participate in planning in the Technical Operations working group, in particular considering processes and tools for IT governance and service management.
We will continue our collaboration with CC-IN2P3 via our monthly meeting of the JCC, focusing on planning for operations.
We will also perform a small exercise using the Sizing Model to understand the cost of keeping two data releases available on readily accessible storage.
 
02C.07.01 Processing Control
 
DMCS: Alert Production Simulator
In September we will investigate task switching with overlapping execution threads, which will require configuring the limited hardware with two Condor pools. Steve will also start documenting the current workflow of the AP simulator, a detailed system specification/description that is aligned with our high-level design specification activities described above.
 
DMCS: Prototype DRP Sequence
Hsin-Fang will work on running the LSST stack on raw DECam data. She will also be coordinating her work with SQuaRE’s verification dataset processing plans, an outcome of the LSST2015 Conference.
 
DMCS: Orchestration work to support verification dataset processing
As described above, several developers will be involved in processing datasets for verification. There was interest in using the existing orchestration code Steve wrote and used to run early (pre-construction) data challenges. We anticipate work in September to update the code and documentation.
 
DMCS: Implementing Features in Process Execution Framework
Matias will continue refactoring Super-Task in pipe_tasks and create a demonstration to test and visualize the workflow. At the end of September he will travel to Tucson to meet with Gregory for a focused planning/work session.
 
02C.07.02 Infrastructure Services
 
Security and Access Control Services: Evaluation of AA and IdM service
We anticipate a collaboration with members of the NCSA Cybersecurity Directorate (CSD) to evaluate authentication and authorization services in the context of SUI, Qserv, and user workspaces. In September we will have initial discussions with the CSD team, introduce them to the SUI and Qserv teams and their design ideas and needs, and develop a SOW.
 
02C.07.03 Environment and Tools
 
No planned activities.
 
02C.07.04 Site Infrastructure
 
Infrastructure work in September involves management level activities described above, such as finalizing the FY15 hardware procurement plan and further developing the FY16 strategy document. We will provide administrative support of the development cluster, including monthly maintenance in September when we plan to roll out the final Puppet base configuration module. We also anticipate the “production” nebula system will be made available in September and we will work select users to commission it for broader developer use.

 


NOAO


Current accomplishments:
02C.08.00 International Communications and Base Site Management Engineering and Integration
 
This month approximately 2-3 days was spent on T-Cam preparations and JIRA.
 
02C.08.01 Base Site
 
Stared to develop Base and Summit design network topologies.
 
02C.08.03 Long-Haul Networks
 
We prepared the RFP with Reuna for the DWDM end equipment.
 
Prepared and delivered a presentation or the Bremerton LSST annual meeting.
 
Met with the camera team whilst at Bremerton to discuss their requirements.
 
Planned Activities.
 
02C.08.00 International Communications and Base Site Management, Engineering, and Integration
 
All except the first week of the month will be vacation.
 
02C.08.03 Long-Haul Networks
 
At the start of September we will hold the RFP Open day to present the project to the vendors.
 

Back to top