DARPA is seeking innovative proposals in the technical areas of compiler design, programming languages and optimization.
MOCHA seeks to build a new generation of compiler technology that can realize the full potential performance of a system comprising multiple heterogenous computational elements. It will accomplish this goal by
- using data driven methods, machine learning (ML), and advanced optimization techniques to rapidly adapt compilers to new hardware components with little human effort and
- developing new internal representations and programming languages that enable compilers to determine how to make optimal use of available hardware, rather than depending on humans to do so.
Without this capability, the world remains constrained by current compiler technologies and lacks the ability
to fully and rapidly capitalize on emerging specialized hardware.
MOCHA will be a 36 month effort, with annual performance assessments guided by two key metrics:
- Human effort reduction – specifically, how much human effort is required to accommodate a new heterogeneous computing ensemble. This will be measured by the number of lines of compiler annotations that are required.
- Performance of the compiled code – (throughput, power consumption, memory footprint) compared to results from conventional approaches.
Abstracts (strongly encouraged) are due 22 August 2024, with full proposals due 26 September 2024.
Document
Broad Agency Announcement – MOCHA: HR001124S0035
