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
      1. 02C.06.02.05 Catalog Services
      2. · no change
    6. NCSA / University of Illinois
    7. Current accomplishments:
    8. NOAO
    9. Current accomplishments:

Data Management Monthly Report

June 2015
 
Project Status

The Summer 2015 release continued, with all teams working on the next round of features and tests against the LDM-240 Milestones and Key Performance Metrics. The rate of progress is similar to that reported in the May report, and continues to be consistent with the reduced level of staffing relative to the initial plan.   Significant results include:

·  Porting and merging HSC code continued, including the multi-band processing scheme. This provides a preliminary capability to detect and measure objects consistently across multiple bands. Other ports include code for better mapping between reference photometric systems and instrumental photometric systems (for photometric calibration) and changes to aperture corrections.

·  A flexible command line interface was developed for the "measurement transformation” framework, which provides a generic way of converting measurements made on images (such as pixel positions or raw fluxes) to astronomically meaningful quantities (such as celestial positions or magnitudes).

·  A prototype version of the Firefly user interface (server side extension), allowing developers to write their own Python executables to do a customized task. (This is mainly to support the Camera group development need at UIUC.)

·  A number of improvements to Qserv robustness at scale, query management and multi-node integration tests.
·  A prototype standard tool for assessing file system loading. The tool chain extracts file IO patterns from a running application and represents the data in a file. It includes Python and C codes to read and write files like the application, as well as a driver program to run the codes in parallel.


 
The Data Management Development Roadmap (LDM-240), has undergone an extensive update covering FY16 - FY18. We added 330 Level 3 Milestones to the plan in JIRA. These milestones will be imported into PMCS in July and linked to the Level 2 Milestones already in PMCS, providing significantly more detail in our ability to track and report on accomplishments in more user-relevant terms. Between now and the end of the LSST 2015 meeting in Bremerton, we will be analyzing inter-milestone dependencies and identifying additional Key Performance Metrics to add to the plan. An interim output of this activity was provided to the AMCL on July 17.
 
We are continuing the process of developing the FY16 budget so that we can start preparing FY16 sub award contract amendments. We will allocate some of the current underspend (due to slower ramp up than plan) to allow the institutions to hire additional resources to catch up deferred work.
 
The amendment to the NCSA agreement for the procurement of hardware was reviewed and updated again and we anticipate execution prior to the LSST 2015 meeting.
 
The Brazil MOA (Networks) has been reviewed and deemed ready for execution by LSST Corporation and the Brazilian parties. Three of the four Brazilian parties have signed, and the fourth is expected prior to the end of July. We anticipate execution of the MOA prior to the LSST 2015 meeting.
 
Recruiting and hiring activities continued across all DM institutions: 4 new positions were opened, 1 new offer was made, and 1 position were filled. Nineteen positions have been filled to date since the MREFC award, while 8 positions are currently open. Three new hires will start in Tucson prior to the LSST 2015 meeting, including :

·  David Nidever, DM SQuaRE Scientist
·  Jonathan Sick, DM SQuaRE Software Developer/Documentation Specialist
·  Angelo Fausti, DM SQuaRE Software Developer (1 year assignment)

 

 

Back to top



Current Photos
 

 
New Continuous Integration System based on Jenkins developed by SQuaRE
 

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). The Roadmap has undergone an extensive update covering FY16 - FY18. We added 330 Level 3 Milestones to the plan in JIRA. These milestones will be imported into PMCS in July and linked to the Level 2 Milestones already in PMCS, providing significantly more detail in our ability to track and report on accomplishments in more user-relevant terms. Between now and the end of the LSST 2015 meeting in Bremerton, we will be analyzing inter-milestone dependencies and identifying additional Key Performance Metrics to add to the plan. An interim output of this activity was provided to the AMCL on July 17.
·  Continued the process of developing the FY16 budget so that we can start preparing FY16 sub award contract amendments. We will allocate some of the current underspend (due to slower ramp up than plan) to allow the institutions to hire additional resources to catch up deferred work.
·  Continued development of the amendment to the NCSA agreement for the procurement of hardware. The amendment was reviewed and updated again and we anticipate execution prior to the LSST 2015 meeting.
·  Continued development of the Brazil MOA (Networks). The MOA has been reviewed and deemed ready for execution by LSST Corporation and the Brazilian parties. Three of the four Brazilian parties have signed, and the fourth is expected prior to the end of July. We anticipate execution of the MOA prior to the LSST 2015 meeting.
·  Prepared, submitted to the CCB, and received approval of an LCR to LSE-78 LSST Observatory Networks Design and LDM-142 LSST Network Sizing to implement changes from the International and Chilean Networks Acquisition Reviews and incorporate updates to the summit network.
·  Continued recruiting and hiring activities. Across all DM institutions: ?? new offer was made, and ?? position filled. ?? positions have been filled to date since the MREFC award, while ?? positions are currently open. Three new hires will start in Tucson prior to the LSST 2015 meeting, including :
o   David Nidever, DM SQuaRE Scientist
o   Jonathan Sick, DM SQuaRE Software Developer/Documentation Specialist
o   Angelo Fausti, DM SQuaRE Software Developer (1 year assignment)
 

Planned activities:

The DM Project Manager will:


·   Continue recruiting and hiring, prepare for arrival of new hires.
·   Complete work with AURA and NCSA on the amendment to the NCSA contract covering LSST equipment procurements.
·   Complete the execution Memorandum of Agreement with Brazil to signature.
·   Complete planning and coordination for the LSST 2015 Meeting at Bremerton August 17 - 21.

DM Science

Current accomplishments:
The DM Project Scientist May activities continued to be focused on the study of the efficiency of LSST as a detector of NEOs.
 
With Lynne Jones, the DM Project Scientist made progress on the analysis of NEO detection efficiency, specifically on the impact of the selection of DM detection algorithms on the ability to detect moving objects. The preliminary analysis shows that: the SNR of trailed asteroids is reduced due to the increased area of the effective PST (i.e., more background), but also the detection efficiency is lower because the default detection algorithms is optimized for stationary point sources. While both those effects were already recognized, the impact of the latter wasn’t appreciated and quantified until now. Even so, we show that the current (still not fully optimized) cadence is capable of reaching ~70% NEO discovery efficiency.
 
He has also begun participation on an ad-hoc panel studying the feasibility of joint processing of LSST, WFIRST and Euclid data.
 
Planned activities:
 
The DM Project Scientist expects to remain focused on the NEO studies in July (and, more broadly, at least through August).
 
DM System Engineering

Current accomplishments:

Activities completed by the DM System Architect include:


·   Defined overall cross-DM milestones
·   Technical deep dives and planning with UW, Princeton, SQuaRE, and IPAC
·   Worked out way of moving forward with gcc 4.8 and C++11
·   Defined RFD process to schedule tech talks and discussions
·   Conducted demos with Data Access and Science Pipelines teams
o   Iterated end-of-sprint demonstration procedure
·   Clarified position of "Product Owner"
·   Defined development processes including third-party contributions
·   TOWG: rewrote non-science use cases
·   Reviewed LDM-142 network spreadsheet
·   Kicked off Process Execution Middleware redesign
·   Started defining LSST 2015 agenda
·   Discussed cooperation with Euclid and WFIRST
·   Interviewed for SLAC position

 

The DM System Interfaces Scientist accomplished:

·   Several communications with Telescope Scientist Sandrine Thomas about revisions to LSE-75 ICD, including increased integration of wavefront and guider pipelines with DM and with calibration data to be produced by DM. Basic architecture outlined in meeting with Sandrine Thomas and K-T Lim at the end of the month

·   Initiated effort with new NCSA team member Matias Carrasco-Kind to refactor and modify the CmdLineTask interface to accommodate Camera data quality monitoring, DM data quality metric analysis, and Level 3 users, and facilitate interface with Firefly. Includes interactions with Applications team, notably Robert Lupton and Jim Bosch, to align this work with their needs as well

·   Supported Firefly team at IPAC in designing a new Python-process-starting capability

·   Attended Camera CD-3 Director's Review, with special attention to camera control system and DAQ issues


Planned activities:

The DM System Architect will:


·   Review DM long-range planning
·   Technical deep dive and planning with NCSA
·   + Refine IT systems architecture
·   Understand production computing requirements for FY16
·   TOWG:
o   Present current state to DM for feedback
o   Rewrite science-based use cases
·   OCS workshop:
o   Define DM telemetry, events, configuration information
o   Review documents
o   Attend workshop
·   Team culture: Define DM community interaction policy
·   Further refine sprint demonstration processes
·   Complete LSST 2015 agenda
·   Finish write-up of Data Butler design and hand over
·   Inform Systems Engineering about DM developer tool plans
·   Define copyright and licensing policy
·   Discuss Wavefront Sensor processing with Telescope & Site group
 

The DM System Interfaces Scientist will:

·  Work on revision of LSE-75 ICD
·  Continue to work with SUI team to interface with Camera team, support the camera team in using Firefly.
·  Continue to work with other DM teams (NCSA and Princeton in particular) in revamp the CmdLineTask interface in support of Firefly server side extension.
·  Attend SciPy conference

 
DM Science Quality and Reliability Engineering (SQuaRE)
 

Current Accomplishments:

 02C.01.02 Science Quality and Reliability Engineering

IPAC worked with Frossie Economou to define 02C.01 tasks relevant to data quality analysis and Level 3 user support. There was a one-day visit by F.E. to IPAC as part of this.

02C.01.02.04 Release CI system to developers for branch builds

The new CI system, was deployed released, first to a number of beta-testers and after early feedback was addressed, to the entire DM developer list on July 1st. This has been SQuaRE’s first major deliverable in construction and is an important component to SQuaRE’s plans for deepening integration testing of the stack. At its current iteration, it offers to developers speed and usability improvements over the existing system but significant extensions will be made on the basis of other planned work. Response from the beta testers was very positive and led to a decision to open it up to further use in parallel with the old buildbot-based system.

The new CI platform was architected by Josh Hoblitt and is based on Jenkins and deployed via a standard cloud interface (currently AWS, but the packer/terraform/vagrant/puppet toolchain will be easily adapted to OpenStack when we SQuaRE has access to its cluster).

The code for the CI configuration and deployment is on Github under the SQuaRE organisation: https://github.com/lsst-sqre/sandbox-jenkins-demo

A small evaluation of Slack was done with a representative group from all sites. The response was positive and it is likely an RFC will result after some other communication tool changes occur in late FY15/early FY16.

02C.01.02 Science Quality and Reliability Engineering

Continued SQuaRE LDM-240 re-planning and a draft LDM-240 was developed.

Planned Activities:

02C.01.02 Continue LDM-240/DLT transition work, work with IPAC team to test out Firefly APIs in DQA and on L3 data.

02C.01.02.03 Support changes to stack build and test infrastructure (RFC-69)

02C.01.02.04 Further CI improvements

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

Simon Krughoff (SK) worked on integrating LDM-240 milestones into the Jira DLP project. This included adding the key performance metrics to the DLP project as well. Russell Owen (RO) and Yusra AlSayyad (YA) prepared for the sprint demonstration. RO, YA, and SK attended the sprint demonstration. Preparation for the sprint demonstration was fairly minimal in this go-around. Mario Juric has asked for more involved demonstrations for some of the more interesting stories. The intent is to generate a set of reference documents for impactful changes.

02C.03.01 -- Single Frame Processing

RO fixed an error in the calibration task that caused an improper assertion to be raised and resulted in buildbot failures on master branches (DM-2938). RO also worked extensively on porting over improvements in the HSC stack that never made it over to the LSST stack. In particular, he cleaned up and ported over code to allow better mapping between reference photometric systems and instrumental photometric systems for photometric calibration (DM 2798 and DM-2919). He also has worked on bringing over important changes to aperture corrections from the HSC side. DM-832 makes the aperture corrections persistable which is a fundamental feature for implementing other aspects of aperture corretion.

02C.03.05 -- Application Framework for Exposures

RO updated the obs_test package used for testing various parts of the stack that require mapper infrastructure. He also added significant functionality to the unit test utilities by adding the ability to test near equality of complex objects such as WCS, Box2[IFD], and pairs of floats (Coords, Extents, Points) (DM-2193).

In order to keep from making the HSC to LSST port a breaking change, some parts will have to live side by side with their LSST counterparts until the HSC port can be completed and the LSST stack changed fully to use the new classes and APIs. YA brought over the Bounded Field class object which is necessary for making progress on other aspects of the HSC port while she is working on improving the approximation/interpolation framework overall.

Planned activities:

02C.03.00 - Alert Production Management Engineering and Integration

YA will attend the Scipy conference in Austin. SK will prepare a talk on the current state of LSST difference imaging.

02C.03.01 -- Single Frame Processing

RO will continue to work on porting the HSC aperture corrections to the LSST stack.

02C.03.05 -- Application Framework for Exposures

YA will continue work on the approximation and interpolation framework.


Princeton University and University of California, Davis

This report covers work carried out in FM8 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

Reduced velocity this month due to travelling (Swinbank away 24 May — 7 June, 12 June — 21 June. Lupton away 31 May — 14 June. Bosch vacation 3 June — 7 June. MacArthur vacation 29 June — 10 July).

The complete LDM-240 “milestone roadmap” Excel sheet (v33) was programmed into the JIRA DLP project as a combination of “meta-epics” and additional milestones. A number of additional milestones and minor tweaks to the existing plan were incorporated.

Following a risk register review, Swinbank flagged up a possible new risk involving the impact of the heterogeneous focal plane on the storage, compute and architecture of the DM system. This discussion is ongoing.

Swinbank produced a series of notes on development practice within the Data Release Production sub-project of DM. The aim is to avoid any possible ambiguities or confusion over how developers are expected to interact with the planning and development process. These are available on Confluence (https://confluence.lsstcorp.org/display/DM/DRP+Development+Notes ).

We started maintaining “work in progress” release notes for the Summer 2015 version of the stack. All developers have been asked to ensure they are kept up-to-date as new functionality is added to the stack. These are also available on Confluence (https://confluence.lsstcorp.org/display/DM/Data+Release+Production+WIP+S15+release+notes ), and will form the basis of the notes which accompany the next stack release.

 

Hiring:

An advertisement for a scientific software developer was posted on the Princeton website, the Data Management site (http://dm.lsst.org/ ), the “LSST is hiring” site (http://www.lsst.org/hiring/ ) and GitHub Jobs (https://jobs.github.com ). The advertisement will run on the AAS Job Register in July.

02C.04.01 - Application Framework for Catalogs

A substantial modernization and simplification of the of the afw::table system was completed. This focused on removing the compound field types: that is, specialized types of table field devoted to storing particular types of object such as points and coordinates. The machinery surrounding the definition of these compound types was complex and inflexible, making it hard to add new types of data to tables. They have been replaced by a new scheme based on “FunctorKeys”, which is more scalable, flexible and lightweight. Old tables will be converted to the new format when then are read. This work was covered by DM-1766.

In previous months we have reported on the “measurement transformation” framework, which provides a generic way of converting measurements made on images (such as pixel positions or raw fluxes) to astronomically meaningful quantities (such as celestial positions or magnitudes). In this month this work was completed by the addition of a flexible command line interface. (DM-2191)

The code used to load complex data types (such as MaskedImages and Exposures) from multi-extension FITS files was re-engineered to be less error prone and to provide better feedback to the user when unexpected data was found in files. (DM-2599)

A number of minor fixes to the Footprint class were ported from HSC. (DM-2606)

A number of minor fixes were made to the code used to load and interpret WCS information from malformed and/or exotic FITS files. This was a result of testing the LSST stack on data from a number of sources (PanSTARRS, SkyMapper, etc) at the Dynamic Universe meeting in Aspen. (DM-2883, DM-2890, DM-2924, DM-2926, DM-2931)

02C.04.03 - PSF Estimation

No work was carried out under this WBS during this month.

02C.04.06 - Object Characterization Pipeline

The multi-band processing scheme was ported to LSST from HSC. This provides a preliminary capability to detect and measure objects consistently across multiple bands. It presents a clear improvement over our previous approach, and, while it does not yet meet our requirements for use in operations, it provides a better base on which to build. This work provides four new command line tasks for detecting sources on multiple coadds, merging the detections to form a consistent list of objects, performing measurements on each coadd, and then combining those measurements to form a consistent catalogue. This work was covered by DM-1945.

It is now possible to use Chebyshev polynomials for approximating image backgrounds. This provides greatly improved background subtraction around bright objects compared to the previous approach (which was based on linear or spline interpolation). (DM-2778)

Preparations for generating galaxy simulations from GalSim and PSFs from PhoSim continued, together with the creation of a test harness for galaxy shear measurements (DM-2657, DM-2658, DM-2659, DM-2662). The ability to execute these tasks in parallel across multiple CPUs was added (DM-2856).

Planned activities:
02C.04.00 Data Release Production Management Engineering and Integration
 
With the exception of continued effort on the galaxy shear measurements (DM-1108), work in July is expected to focus exclusively on merging changes from the HSC stack. We aim to port all of he major components by late July, then spend August focusing on remaining clean-up in the run up to the LSST All Hands Meeting in Bremerton.
 
02C.04.01 - Application Framework for Catalogs
 
The HSC parallelization middleware will be merged to LSST (DM-2983).
 
No other work is explicitly planned under this WBS for July, but we expect a series of bug-fixes and minor improvements will be necessary in support of the HSC porting effort.
 
02C.04.03 - PSF Estimation
 
No work is expected to be undertaken under this WBS.
 
02C.04.06 - Object Characterization Pipeline
 
The major components of the HSC stack will be ported to LSST. This includes:
 • The PSFEx PSF measurement code (DM-2961)
 • The meas_mosaic multi-exposure astro-and photo-metric fitting code (DM-2674)
 • Improved deblending code (DM-1954, DM-2914)
 • The HSC distortion estimation code and camera description (DM-2194, DM-1794)
 
In addition, work will continue on the estimation of galaxy shear measurement parameters (DM-1108).


IPAC / California Institute of Technology

Current accomplishments:

02C.05.00 Science User Interface and Analysis Tools Management Engineering and Integration

·  Summer vacation. Xiuqin: two weeks; Trey: few days; John: one week; Tatiana: one week.

·  Finished transfer the milestones in LDM-240 road map to Jira project DLP.

2C.05.01 Basic Archive Access Tools

·  Added Python API for accessing XY plot in FIrefly

·  Started working on the design and implementation of reading in FITS binary table, extending the current capabilities to cover more types of binary table.

·  Improved the table reading in Firefly to handle extra wide table.

·  Allow XY-plot to handle multiple tables with the same name.

02C.05.02 Data Analysis and Visualization Tools

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

·   Finished a prototype version of Firefly server side extension, allow developers to write their own Python executables to do a customized task. This is mainly to support the Camera group development need at UIUC.

·   Finished adding two new stretch algorithm (asinh, power law) in Firefly for image visualization.

Planned activities:
02C.05.00 Science User Interface and Analysis Tools Management Engineering and Integration

·  Summer vacation continues. Trey: one week; Tatiana: two weeks; Loi Ly: couple of days

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

·  Finish LDM-240 road map to Jira project DLP, add KPM and meta-epics.


 
02C.05.01 Basic Archive Access Tools
 

·  Continue 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
·  John Rector will be attending SciPy conference for one week
·  Start on the conversion of one or two GWT widget into pure JavaScript code, use the experience as a test case to draft a plan for the conversion.

 
 02C.05.02 Data Analysis and Visualization Tools
 

·  Continue the development on the JavaScript APIs and Python APIs for Firefly visualization components.
·  Work on the design and implementation of overlay masks on the primary image.
·  Continue to work with Camera group to support their development.

 
 


SLAC / Stanford University

Current accomplishments:

Highlights:

 
· continuing debugging problems with Qserv at scale, narrowed down the problem, it took a lot more effort than expected. We were reserving lots of time for the unexpected so we are on track, though we decided to slow down with data distribution and ingest and put all the effort we could into solving the scalability problems.
· hired Nate Pease
· good progress with query management and multi-node integration tests
02C.06.00 Science Data Archive and Application Services Management Engineering and Integration 
· Coordinated June Sprint for the Data Access Team
· Revisited long term budget
· Interviews:
· Qiua Zhen (phone screening)
· Tony Wildish (phone screening and full interview)
· Finalized job offer for Nate Pease, he is starting July 7
· Organized weekly Qserv and Data Access meetings
· XLDB
· finished and released XLDB-2015 workshop report
· discussions about XLDB-gov with NSF
· SLAC related
· Data Access and Database talk for the SLAC lsst-local group
02C.06.01.01 Catalogs, Alerts and Metadata
· No change / no progress. Ingest code (DM-210) still to-do here. Debugging problems with large scale tests had higher priority.
02C.06.01.02 Image and File Archive
· No change, webform still to-do. Tony was busy with Camera CD-3 review.
02C.06.02.01 Data Access Client Framework
· Butler discussion, progress with cleaning up prototype implementation of Butler v2.
02C.06.02.02 Web Services
· Progress with error handling, will finish in July
02C.06.02.03 Query Services
· Continued debugging problems related to multi-node. Narrowed down the problem to the new code in xrootd.
· Built Qserv Release 2015_06. Release highlights:
· Multi-node integration tests
· Basic query metatdata framework
· Many valgrind/helgrind fixes
· More C++11 migration (std::scoped_ptr, unique_ptr)
· Code improvements:
· DM-2720: Migrate boost::scoped_ptr to std
· DM-2953: Migrate auto_ptr --> unique_ptr
· DM-2803: Adapt multi-node tests to latest version of qserv / loader
· DM-2643: Migrate Qserv to ssi v2
· DM-2892: Keep track of database of the director table
· DM-3091: Remove unused function populateState
· New features:
· DM-2627: Add support for configuring multi-node integration tests
· DM-2804: Implement query metadata skeleton
· DM-2858: Add support for "ORDER BY f1, f2" for has-chunks query
· Bug fixes:
· DM-2777: Fix races in BlendScheduler
· DM-2779: Fix race in Foreman
· DM-2854: Fix Qserv SsiSession worker race
· DM-2681: Fix race condition in userQueryProxy
· DM-2935: Fix qserv-admin CREATE NODE failure
· DM-2982: Fix qserv-admin node status INACTIVE failure
· DM-2864: Fix selection of rows by objectId from non-director table
· DM-2945: Fix wmgr service on remote interfaces
· DM-2947: Fix data load of secondary index in multi-node case
· Documentation improvements
· DM-2841: Add User Manual
· DM-3018: Update dev quick-start guide for new qserv_testdata repo
· DM-2846: Remove hardcoded username from doc/publish.sh
02C.06.02.04 Image Services
· DM-2467: Implement stitching multiple patches across tract boundaries in a coadd v2


02C.06.02.05 Catalog Services


·   no change


Planned activities:

02C.06.00 Science Data Archive and Application Services Management Engineering and Integration

·  Organize weekly Qserv and Data Access meetings

·  Search for candidates for remaining open position

·  FY16 budget planning

02C.06.01.01 Catalogs, Alerts and Metadata

·   ingest

02C.06.01.02 Image and File Archive

·   Finish work on improvements to the form

02C.06.02.01 Data Access Client Framework

·   Wrap up butler v2

02C.06.02.02 Web Se rvices

 
· Implement RESTful python client
· Research and Document API Versioning
· Contextual error handling
· Add unit tests for webserv
02C.0 6.02.03 Query Services
· Migrate Qserv to ssi v2
· Resolve problems with large scale tests
· Finish designing Data Distribution, start lightweight prototyping of data distribution
· Finish work on Query Management system
· Continue work on Qserv Refactoring (DM-1707)
02C.06.02.04 Image Services
· Research supporting cutout from images with overlaps
02C.06.02.05 Catalog Services
· Implement RESTful interfaces for database (GET)


NCSA / University of Illinois


Current accomplishments:
02C.07.00 Processing Control and Site Infrastructure Management
 
Management effort was limited in the month of June due to several weeks of vacation. Activities of note include reviewing applicants for NCSA’s open positions (system development lead, research scientist, and research programmer) and on-boarding Matias Carrasco-Kind as a part time developer for the Process Execution Framework.
 
02C.07.01 Processing Control
 
Data Management Control System: Alert Production
(DM-2830) – For the first half of June, middleware development focused on fixing orphaned threads in the archive DMCS and in the distributor. Threads created by connections from external processes were remaining alive even if the processes died, but now the system will check to ensure that the connections are still viable and kill the threads if they are not.
 
Data Management Control System: OCS Software
(DM-2932) – The second half of June was spent understanding the latest version of the OCS software in preparation for the July CCS-DAQ-OCS-DM Workshop IV. The software was installed on two CentOS VMs stood up by the NCSA system admins to simulate two subsystem sites. To test the Service Abstraction Layer (SAL) software that publishes and transmits telemetry data through the OCS, software was written to test sending and receiving messages from the SAL software.
 
Pipeline Execution Services
(DM-3004) – Matias Carrasco-Kind joined the NCSA LSST team (.25 FTE) toward the end of June. He began working on the process execution framework, prototyping and documenting a framework design and corresponding closely with Gregory Dubois-Felsmann for input and feedback.
 
Technical Operations Working Group
In preparation for contribution to IT use cases for the TOWG, a review of ITIL 3.0 was conducted, as it provides a standard framework breakdown of processes used in an IT system. An ITIL-like spreadsheet was created (DM-2958). The ITIL spreadsheet was then used as a checklist to begin developing use cases (DM-2998).
 
 
02C.07.02 Infrastructure Services
 
Security and Access Control Services: ISO work
 
During June the ISO attended the ACM/CLHS 2015 conference and discussed the LSST SCADA security plan with CTSC professionals, who were crucial in developing the LSST cybersecurity program last January (DM-3016).
 
For the rest of the month the ISO focused on authentication and authorization services and Acceptable Use Policies. To investigate the use of API keys, he did a short case study on the OSG model of managing API keys and concluded that this model was not a viable method for LSST (DM-2611).   The ISO began outlining the AAA requirements and considering potential solutions for the system. He discussed the Central Authentication Service (CAS) system with Tony Johnson (DM-2852). He worked with the LSST system admin in Tucson to review the current system for AUP acknowledgment and to consider an automated system for new and existing LSST users to acknowledge receipt and acceptance of the AUP (DM-2963, DM-3000). Additionally, he began drafting a whitepaper on the LSST SCADA security challenges in preparation for the August NSF Cybersecurity Summit (DM-3001).
 
System Administration and Operations Services: Configuration Management (Puppet)
 
(DM-2237) – Puppet standardized configuration management services were expanded to include a user module. This module manages users, groups, and ssh public keys on different servers depending on the server role. The module was created, tested, and deployed during the scheduled monthly maintenance, and was then used in production to add a new user.
 
File System Services: File System Research & Prototyping
 
(DM-2898) – A major accomplishment in June towards researching file systems was the creation of a prototype standard tool for assessing file system loading. The tool chain extracts file IO patterns from a running application and represents the data in a file. It includes Python and C codes to read and write files like the application, as well as a driver program to run the codes in parallel. Additional work to profile file systems was done by gathering strace logs of software packages during a build of the 10.1 stack (DM-2851).
 
(DM-2695) – Work continued in deploying a Ceph prototype, in particular identifying and documenting administrative procedures to investigate, testing failure modes and solutions, adding Puppet-managed OSD servers to monitor the Ceph cluster.
 
File System Services: File Management Technology
 
In June, addressing file corruption within iRODS 3.x continued. iRODS relies on maintaining replica data as a key to preventing data loss. An iRODS 3.3 Resource Group was set up with the iCAT metadata server and the replica server, and a microservice was implemented that auto-replicates and checksums data every 12 hours (DM-2572, DM-2692). As a related activity, the iRODS servers were installed inside Docker containers, and images were made and stored in the Docker repository (DM-2336).
 
Additionally, investigation began on using iRODS to access tape archives, with initial testing done against the HPSS at NERSC (DM-2962).
02C.07.03 Environment and Tools
 
Environment and Tools: Deployment plan for version 1 of OpenStack
 
(DM-2153) – At the end of June the system engineering team at NCSA attended a 3-day seminar on OpenStack, where they gained valuable insight into integrating OpenStack with LDAP. This is well-timed with the expected deployment of the NCSA OpenStack in the next few weeks.
 
02C.07.04 Site Infrastructure
 
Development and Integration Infrastructure: Setup Qserv prototype for Qserv and SUI teams
 
(DM-2327) – To support the SUI developers, the two spare servers identified last month were deployed, with most of the OS configuration done using Puppet and provisioning support provided by the NCSA networking group.
 
Archive Site External Network: Wide-Area Network Work
 
(DM-3023) – The network engineer at NCSA reviewed LSE-78, the LSST Observatory Network Design document, to suggest updates.
 
(DM-3024) – Additionally, some work was done investigating US WAN options with a representative in the NCSA ESnet group.
 
LOE (sys admin)
 
A few system admin LOE activities emerged in June, including adding three new users, providing feedback on rack specifications for the Chilean data center, and fixing issues with two qserv and one NFS servers. At the request of the developers, the research was done on gcc-4.8x compiler options for EL6 for potential deployment during next month’s scheduled maintenance.
 
Planned activities:
 
02C.07.00 Processing Control and Site Infrastructure Management
 
Management activities in July will be focused on finalizing the procurement contract and investigating regulations for purchasing cloud computing capabilities. We will welcome Hsin-Fang to our group as our new post doc; she will be focusing on using the LSST stack to process DECam data. We will also interview applicants for the three open positions. Additionally, we will begin to think about near-term plans and long-term milestones as impacted by our development of the system design document.
 
02C.07.01 Processing Control
 
In July NCSA will continue work on the OCS software and host the three-day CCS-DAQ-OCS-DM Workshop IV. Preparing for this workshop in June delayed planned work on creating pilot Condor jobs in the AP simulator, so this will be revisited in July.
 
Development of a processing execution framework prototype will continue, with documentation and graphical representation of the system.
 
02C.07.02 Infrastructure Services
 
The ISO will be attending the NSF Cybersecurity summit in August, and for that he will spend much of July preparing a whitepaper about the LSST cybersecurity program.
 
We plan to add into Puppet and deploy another module for standardizing the configuration of firewall rules, continuing our work to standardize configuration management.
 
Following our investigation of addressing file corruption in iRODS 3.x, we will consider the tools and configurations to handle this in iRODS 4.x.
 
02C.07.03 Environment and Tools
 
Our biggest goal in this area is to stand up the NCSA OpenStack by the end of July.
As the procurement contract is still under negotiation, we do not plan to purchase any new equipment for further development in this area.
 
02C.07.04 Site Infrastructure
 
The plan in this area is to begin drafting a fundamental specification document for the systems we will need to design and build for operations. This will include collecting together requirements scattered throughout LSST design documents and deep thinking about use cases. In the event that we hire a system management and development lead, he/she will focus in this area.
As the procurement contract is still under negotiation, we do not plan to purchase any new equipment for further development in this area.


NOAO


Current accomplishments:
 
02C.08.03 Long-Haul Networks
 
The DM Network Architect attended a CiscoLive conference in San Diego to learn and explore new Data Center network design architectures. Am in the process of producing a document of my findings and suggestions for the Summit and Base Network designs.
 
Somyl representatives came to La Serena for the day to inspect and examine the type of surface on the road from Cerro Tololo to Cerro Pachon with the intention of providing a quote for a fiber installation.
 
As of end of June Reuna and Telefonica had not signed the contract for the La Serena to Santiago dark fiber IRU. Apparently it is very close but until that is done Reuna’s suggestion is that we do not change anything from the verbal agreements made at the end of last year. The fact is though that Telefonica did not want to get involved with a buried fibre from Pachon to Tololo and at their suggestion we went out to get a quote on that link. That quote came in at the end of June for $12/meter which is the best we can expect. So unless we want to install a ring immediately we should wait until Telefonica are willing to discuss the aerial run from Pump1 in San Carlos to Cerro Pachon and deduct that portion from their original quote.
 
Discussions held with Reuna and Fernando Liello for the organization of the RFI from a selected number of DWDM manufacturers. To be held July 9 – 14 with seven companies.
 
Planned Activities
 
02C.08.03 Long-Haul Networks
 
We will hold the RFI meetings with seven companies who are interested in bidding for the DWDM nodes for the Summit to Base and La Serena to Santiago links. The dates will be July 8 – 14 with two companies per day. Namely: Ciena, Infinera, Alcatel, Padtec, Huawei, ECI Telecom, Coriant.
 
Designated LSST and DC representative on the panel for Solicitation of the Base Facility, which will be held on 7 July.
 
Working with Cisco for quotations for DC network equipment Somyl are expecting an answer to their proposal for installation of fiber. An off the wall idea now we have the $12/meter figure is to contract a buried fiber from Gate to Tololo/Pachon which would prove more reliable but with many ramifications. However that is one we can keep in our pocket for a rainy day!!
 
Finalize and issue the document on DC design and start discussions with the teams.
 

Back to top