Advanced Architectures and Critical Technologies for Exascale Computing | Apply Online

Frequently Asked Questions (FAQs): Advanced Architectures and Critical Technologies for Exascale Computing (DE-FOA-0000255)

1) What is this funding opportunity?

This is a U.S. Department of Energy (DOE) Office of Science funding opportunity, run through the Office of Advanced Scientific Computing Research (ASCR), titled "Advanced Architectures and Critical Technologies for Exascale Computing" (Funding Opportunity Number DE-FOA-0000255). It supports basic and applied research intended to make practical exascale computing achievable.

2) What is the main goal of the program?

The program aims to push the research community toward practical exascale computing by removing fundamental barriers that would otherwise make exascale systems too costly, too power-hungry, or too unreliable to be useful. The focus is on breakthroughs in computer architectures and critical technologies that enable exascale performance with acceptable cost, power use, and resilience.

3) Why is DOE emphasizing new approaches instead of traditional performance scaling?

The announcement states that the historical "free" performance gains from each new processor generation (largely driven by increasing processor clock speeds) have largely run out. Because of this, achieving exascale performance requires new advances across architecture and system technologies rather than relying on conventional scaling.

4) What problem does the FOA highlight with building exascale systems using conventional technology?

The FOA explains that building an exascale machine using the technologies of the time would be impractical, with estimates of about $1 billion in cost, power consumption above a gigawatt, and a projected mean time between failures of roughly ten minutes. The opportunity is designed to fund research that avoids those outcomes.

5) What kinds of research topics does this opportunity prioritize?

Based on the description provided, the opportunity targets research that removes fundamental exascale barriers, particularly work that leads to:

• Energy-efficient hardware and software architectures for high-performance computing at exascale
• Resilient (fault-tolerant) hardware and software architectures suitable for extreme-scale computing
• Advances spanning devices, systems, programming models, runtime systems, and applications

6) What larger mission needs is this FOA tied to?

The FOA ties exascale computing to major science and national security challenges that increasingly depend on computing at extreme scale, including examples such as:

• Climate modeling
• Combustion efficiency
• Fundamental physics questions related to dark energy and dark matter

7) Who is this opportunity intended for?

The opportunity targets interdisciplinary teams in computer science and computer engineering, especially teams organized around a cohesive and transformational research agenda.

8) Are multi-institution collaborations encouraged?

Yes. The FOA strongly favors multi-institution collaborations that are organized around a cohesive, transformational research plan. Partnerships are encouraged when they fill critical expertise gaps and support breakthroughs that a single institution could not likely deliver alone.

9) What types of organizations can be part of a collaboration?

The text encourages partnerships that can mix universities, DOE National Laboratories, and industry, and also notes collaborations may include a wide range of partners such as nonprofits, federal laboratories, and FFRDCs (including DOE National Laboratories), provided their participation is modest and aligned to critical expertise needs.

10) Can a Federally Funded Research and Development Center (FFRDC) apply as the prime applicant?

No. The FOA excludes FFRDCs as applicants. However, the FOA indicates that collaborations may include FFRDCs (including DOE National Laboratories) as part of a broader team, as long as their participation is a modest portion of the overall effort and is justified by critical expertise needs.

11) Are other federal agencies eligible to apply?

No. The FOA states that other federal agencies are not eligible to apply or participate as funded recipients.

12) Are there any nonprofit eligibility exclusions?

Yes. The FOA excludes certain 501(c)(4) nonprofit organizations that engaged in lobbying activities after December 31, 1995.

13) Is eligibility limited to domestic entities?

Yes. The FOA states eligibility is broadly open to domestic entities, subject to the specific exclusions listed (such as other federal agencies, FFRDCs as applicants, and certain lobbying 501(c)(4) organizations).

14) What is the award type for this opportunity?

This was described as a discretionary research and development opportunity using cooperative agreements. That means DOE anticipated active involvement during the project compared with a typical grant.

15) Does DOE expect to be actively involved if selected?

Yes. Because the mechanism is a cooperative agreement, DOE expected a more active involvement role than is typical for standard grant awards.

16) How much total funding was estimated for this FOA?

The FOA listed an estimated total funding amount of $5,000,000.

17) About how many awards were expected?

The FOA indicated an expectation of approximately five (5) awards.

18) Is cost sharing or matching required?

No. The FOA explicitly stated there was no cost sharing or matching requirement.

19) What is the CFDA/assistance listing for this opportunity?

The assistance listing is CFDA 81.049, identified as the Office of Science Financial Assistance Program.

20) What were the key dates for this opportunity?

• Posted date: January 29, 2010
• Application closing date: March 26, 2010
• Archived date: July 1, 2010

21) Is this opportunity still open?

No. The record states the opportunity was later archived on July 1, 2010, and it listed an application closing date of March 26, 2010.

22) Which DOE offices are involved in managing or coordinating this program?

The opportunity was released by the DOE Office of Science through ASCR, and the announcement notes the program is managed in cooperation with the National Nuclear Security Administration (NNSA) and the Defense Advanced Research Projects Agency (DARPA).

23) What does it mean that NNSA and DARPA are involved?

The FOA notes this cooperation as a signal that outcomes are expected to matter for both open scientific computing and national-security-related computing needs.

24) Does the FOA encourage large consortia?

Not for their own sake. The description emphasizes that proposals should not be sprawling consortium efforts simply to assemble a large group. Teaming is encouraged when it strengthens the science and engineering plan, fills critical expertise gaps, and remains a modest portion of the overall effort for certain partner types (such as FFRDC involvement).

25) What is the listed agency name in the record?

The listed agency name in the record is the Chicago Service Center.

26) Where were applicants directed to find the full announcement materials?

The FOA pointed applicants to FedConnect for the full announcement materials.

27) Who were the listed contacts for questions or access issues?

The FOA listed contacts including:

• Marilyn M. Oyler, Grants Analyst (phone and email were provided in the FOA)
• Daniel A. Hitchcock, Program Manager (phone and email were provided in the FOA)

28) What kinds of exascale barriers is DOE trying to overcome through this program?

From the information provided, DOE is trying to overcome fundamental barriers that would otherwise make exascale systems impractical, particularly barriers related to:

• Cost (avoiding billion-dollar systems)
• Power draw (avoiding gigawatt-scale consumption)
• Reliability (avoiding extremely frequent failures such as a ten-minute mean time between failures)