Theo X. Olausson
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Theo X. Olausson has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Armando Solar-Lezama
Massachusetts Institute of Technology
1 shared publication - Catherine Wong
Massachusetts Institute of Technology, USA
1 shared publication - Daniel J. Sorin
Duke University
1 shared publication - Gabriel Grand
Massachusetts Institute of Technology, USA
1 shared publication - Joshua B. Tenenbaum
Massachusetts Institute of Technology
1 shared publication - Kevin Ellis
Cornell University
1 shared publication
Latest Publications
- 10 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 43 (4) , 62-70
HeteroGen: Automatic Synthesis of Heterogeneous Cache Coherence Protocols
We address the two challenges architects face when designing heterogeneous processors with cache-coherent shared memory. First, we introduce HeteroGen, an automated tool for composing clusters of cores, each with its own coherence protocol. Second, we show that the output of HeteroGen conforms to a precisely defined memory consistency model that we call a compound consistency model. We also demonstrate that HeteroGen can correctly fuse a wide range of coherence protocols. Our experiments indicate that protocols generated by HeteroGen perform comparably to a publicly available manually generated heterogeneous protocol.
- 10 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 43 (4) , 62-70
HeteroGen: Automatic Synthesis of Heterogeneous Cache Coherence Protocols
We address the two challenges architects face when designing heterogeneous processors with cache-coherent shared memory. First, we introduce HeteroGen, an automated tool for composing clusters of cores, each with its own coherence protocol. Second, we show that the output of HeteroGen conforms to a precisely defined memory consistency model that we call a compound consistency model. We also demonstrate that HeteroGen can correctly fuse a wide range of coherence protocols. Our experiments indicate that protocols generated by HeteroGen perform comparably to a publicly available manually generated heterogeneous protocol.
- Open Access
- 9 January 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Programming Languages
- Vol. 7 (POPL) , 1182-1213
Top-Down Synthesis for Library Learning
This paper introduces corpus-guided top-down synthesis as a mechanism for synthesizing library functions that capture common functionality from a corpus of programs in a domain specific language (DSL). The algorithm builds abstractions directly from initial DSL primitives, using syntactic pattern matching of intermediate abstractions to intelligently prune the search space and guide the algorithm towards abstractions that maximally capture shared structures in the corpus. We present an implementation of the approach in a tool called Stitch and evaluate it against the state-of-the-art deductive library learning algorithm from DreamCoder. Our evaluation shows that Stitch is 3-4 orders of magnitude faster and uses 2 orders of magnitude less memory while maintaining comparable or better library quality (as measured by compressivity). We also demonstrate Stitch’s scalability on corpora containing hundreds of complex programs that are intractable with prior deductive approaches and show empirically that it is robust to terminating the search procedure early—further allowing it to scale to challenging datasets by means of early stopping.
- 9 January 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Programming Languages
- Vol. 7 (POPL) , 1182-1213
Open AccessTop-Down Synthesis for Library Learning
This paper introduces corpus-guided top-down synthesis as a mechanism for synthesizing library functions that capture common functionality from a corpus of programs in a domain specific language (DSL). The algorithm builds abstractions directly from initial DSL primitives, using syntactic pattern matching of intermediate abstractions to intelligently prune the search space and guide the algorithm towards abstractions that maximally capture shared structures in the corpus. We present an implementation of the approach in a tool called Stitch and evaluate it against the state-of-the-art deductive library learning algorithm from DreamCoder. Our evaluation shows that Stitch is 3-4 orders of magnitude faster and uses 2 orders of magnitude less memory while maintaining comparable or better library quality (as measured by compressivity). We also demonstrate Stitch’s scalability on corpora containing hundreds of complex programs that are intractable with prior deductive approaches and show empirically that it is robust to terminating the search procedure early—further allowing it to scale to challenging datasets by means of early stopping.