AGILE

Advanced Graphic Intelligence Logical Computing Environment
AGILE

Intelligence Value

The AGILE program seeks innovative, energy-efficient, and reliable computer architectures that can address the Intelligence Community’s large-scale data-analytic applications, as well as other classes of data intensive applications. AGILE offers the potential to enable a predictive analysis of massive data from diverse sources and methods, not just forensic analysis after the event has occurred.

Summary

The fundamental problem with current computer architectures is their inefficiency at operating on sparse, time-varying data that is randomly distributed across the system. The AGILE program seeks to solve this problem by developing new system-level intelligent mechanisms for accessing, moving, and storing complex data streams and structures that enable efficient data-analytic algorithms.

New architectures developed under the AGILE program will be driven by representative data-intensive applications through the co-design process. Co-design is a process for designing computer systems whereby the application requirements influence architecture decisions, and the architecture affects the design of the applications. AGILE system designs must emphasize optimization of the fully integrated system rather than independent optimization of individual functionalities (e.g., memory, computation, or communication), and must not be constrained by existing component interfaces and protocols, legacy architectures, or current practices.

A fundamental rethinking of computer architectures that can revitalize performance growth trends in computing capabilities is long overdue. Currently, there is a renewed interest in developing specialized hardware components. However, this approach will not resolve the fundamental data movement challenges that restrict the historical performance growth trends. The AGILE program will seed a new generation of computers with unprecedented pathways for continuing performance gains for the IC.

The AGILE BAA was released in November 2021 and the program is slated to run for three years.

BAA Information

Press Releases

AGILE BAA Press Release

Technical Details for Proposers

Description of the four workflows and their derived kernels are detailed in the AGILE Program Workflows Document
 
Performer will be responsible for providing models of their designs using design-tool-neutral modeling or hardware description artifacts that are integrated with A-SST (SystemC, C/C++, SST, Verilog, or System Verilog). The description of AGILE extensions to SST are detailed in the A-SST Documentation.
 

The AGILE BAA refers toFireSim,  a cycle-accurate, FPGA-accelerated scale-out computer system simulation platform that may be used by performers. Description of the FireSim environment is detailed in this FireSim Documentation.

A baseline system will be used by the T&E to establish the baseline performance of the AGILE application suite. The baseline system will also be used to evaluate the performers’ implementations of the AGILE applications suite. Description of the baseline system is detailed in this Baseline Computer System.

Reports and Publications

Coming soon.

Presentations, Speeches, and Interviews

ISC High Performance 2022 Presentation 

Frequently Asked Questions (FAQs)

AGILE FAQs

Questions and Answers (Q&A)

AGILE Q&A

Proposers' Day Briefings


Contact Information

Program Manager

Dr. William Harrod

Research Area(s)

Computer architecture, Data analytics, Co-Design, Data movement, Modeling and simulation

Broad Agency Announcement (BAA)

Link(s) to BAA

W911NF22S0001

Solicitation Status

CLOSED

Proposers' Day Date

December 22, 2020

BAA Release Date

November 29, 2021

Proposal Due Date

January 24, 2022

Program Summary

Testing and Evaluation Partners

  • Lawrence Berkeley National Laboratory
  • Sandia National Laboratory
  • Pacific Northwest National Laboratory

Additional Information

Questions

All questions received were considered and the ones that we are answering in the AGILE Q&A have been re-stated in order to generalize them for broad public consumption. 

Administrative, technical, or contractual questions:
Email This email address is being protected from spambots. You need JavaScript enabled to view it.

Registration process, system requirements, application form, or submittal process questions:
Use https://www.grants.gov/web/grants/support.html or call Grants.gov at 1-800-518-4726

Teaming Information

Teaming Form

Teaming Information