Skip to content

Publications

Empowering WebAssembly with Thin Kernel Interfaces

Arjun Ramesh, Tianshu Huang, Ben L. Titzer, Anthony Rowe
wali-image2

Webassembly (Wasm) adoption for new domains is often hindered by the lack of standard system interfaces. Kernel interface support in Wasm engines allow numerous languages and applications to target Wasm, including arbitrary high-level Wasm APIs (WASI), driving ISA portability and software safety down to even deeply embedded OS ecosystems

EuroSys 2025 Pre-Print WALI Rust Target


Unveiling Heisenbugs with Diversified Execution

Arjun Ramesh, Tianshu Huang, Jaspreet Riar, Ben L. Titzer, Anthony Rowe
wali-image2

The detection of quirky Heisenbugs are influenced by numerous factors including (but not limited to) hardware platforms and, strangely, the extent of instrumentation used by debuggers themselves. We propose a debugging methodology that harnesses diversity across platforms and the debugger's instrumentation to enable low-overhead debugging at scale. Tune into the paper for a characterization of over 100 numerous funky bug behaviors!

OOPSLA 2025 Data Artifact


Silverline: Lightweight Virtualization and Orchestration of Distributed Systems

Arjun Ramesh, Tianshu Huang, Emily Ruppel, Dakshina Dasari, Behnaz Pourmohseni, Fedor Smirnov, Marco Giani, Paolo Pazzaglia, Charles Shelton, Nuno Pereira, Arne Hamann, Dirk Ziegenbein, Anthony Rowe
wali-image2

Real-time cyber-physical systems naturally lend themselves to distributed compute, but are often programmed monolithically to curb complexity and safety exploits. Our framework, Silverline, explores programmability and orchestration of distributed applications transparently across wide range of hardware using WebAssembly for lightweight virtualization.

RTAS 2025 Silverline Org


Interference-aware Edge Runtime Prediction with Conformal Matrix Completion

Tianshu Huang, Arjun Ramesh, Emily Ruppel, Nuno Pereira, Anthony Rowe, Carlee Joe-Wong
wali-image2

Performance prediction in heterogeneous edge computing systems is critical for effective resource allocation and cost optimization. We formulate this problem as matrix completion, which can be extended to handle complex, edge-specific concerns such as interference and uncertainty quantification accurately.

MLSys 2025