Managing government funded scientific consortia Page: 4 of 5
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
the Lattice QCD Collaboration. Each of these line items has a
baseline value, a planned improvement value, and actual
MEASUREMENT INDICATORS - EXAMPLES
New Customers and
Number of distinct users of the facility (includes
DOE labs and academic communities)
% of generated necessary improved staggered -
LQCD science deliverables
Teraflop-Years delivered towards the completion of
the 2008 Scientific Program
% of trouble tickets closed within 2 business days
% of average machine uptime at the clusters and
C) Choose service type and components (using SRM): The
FEA SRM framework is designed to identify services
provided by the investment. Identification of service types
and components allow projects to determine services that
cover multiple lines of business. LQCD SRMs are
functionally independent of the individual sites. This section
may also be used to indicate if a service provided by the
LQCD project may be reused by another OMB IT project.
Service components for LQCD project include library and
storage; simulations; and information sharing.
D) Complete technical support information (using TRM):
LQCD project places high emphasis on "state-of-the-art"
scientific computing dedicated to Lattice QCD simulations.
The Table 3 shows a subset of the technical components and
their specifications for LQCD.
The culture of formalisms described above is somewhat
new to the scientific environment. In the university culture,
with floating student population and dependence on grants, it
is often difficult to quantify the performance. Establishing a
suitable project methodology using the FEA framework was
challenging at the beginning of the project. FEA also demands
various progress reporting. As time progresses, FEA guides
are becoming clearer and accepting the changing culture is
becoming easier. The process is beneficial to the centralized
project office, as everybody in the consortium can follow the
The most important management challenge to such projects
is the financial one. Annual appropriation of federal funding
has an element of uncertainty. Certain projects may be
impacted by changes in administrations. Financial challenges
must e\be matched with the technical challenge to meet the
performance goals. For LQCD project success, efficiency of
the labor force is critical because the labor cost must be
minimized to maximize the hardware purchase cost.
Consequently, the project management process must be most
efficient as well. Technical experts are constantly working on
increasing the satisfaction of scientific customers. A small and
dedicated group of world-class technical experts continue to
fine-tune the facility. They perform benchmarks on machines
to maximize the efficiency of the computer hardware
purchased and work on improving the robustness of
distributed computing facility operation.
EXAMPLES OF LQCD TECHNICAL SPECIFICATIONS
FEA SRM TRM Service Specification
Simulation Torque (OpenPBS) Batch System
Information Sharing Kerberos (MIT krb5)
Data Exchange International Lattice Data Grid LQCD Metadata
Library / Storage Jetstore RAID Disk Arrays (SCSI-Attached EIDE
and SATA Disk Arrays)
Mellanox Infiniband Switches and Host Channel
Simulation Custom QCDOC Supercomputer
Simulation SciDAC Lattice QCD Software Libraries
Considerations for alternate analysis for the investment,
including analysis of net present value and life cycle costing
are required for all OMB300 projects. The LQCD project uses
various computing machine benchmarking and active research
on vendor roadmaps to determine alternate analysis. Risk
identification and defining mitigation strategies are other
mandatory management components. The project, with its
stringent investment reporting requirements is, to a degree, a
first-of-a-kind project managed at Fermilab. Designing and
managing such projects is always challenging .
The diversity of people, institutions, and locations imposes
another challenge to consortia. Members of Lattice QCD
collaboration are located throughout US and operate under
different management principles and structures. Out-front
efforts to negotiate uniform project management methodology
and parameters for the project helped. Solutions to security
and privacy issues are becoming increasingly complex and
costly for all IT projects and LQCD project is no exception.
The project office tracks the compliance to appropriate
regulatory issues. The deployment of related tools and
procedures is becoming a recurring cost to such consortia.
The primary lesson is to be cognizant of the changing
culture of governance required by the funding agencies. It is
critical to take into account the cost of governance
appropriately from the onset. It helps in minimizing the
governance costs and improves performance of the project.
Here’s what’s next.
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Banerjee, Bakul. Managing government funded scientific consortia, article, June 1, 2007; Batavia, Illinois. (https://digital.library.unt.edu/ark:/67531/metadc877433/m1/4/: accessed April 20, 2019), University of North Texas Libraries, Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.