The Stanford Software Research Lunch is a weekly event on Thursday where students and researchers present their latest work to peers. Talks are open to anybody, but regular attendees are expected to give a presentation on their work.

Mailing list: software-research-lunch@lists.stanford.edu (subscribe via mailman)

Calendar: ical

Format: The lunch is held every week during fall, winter and spring quarter. The first week of every quarter is an organizational lunch where people can sign up to give a talk. If you'd like to give a talk, please contact Matthew Sotoudeh or Anjiang Wei.

Past quarters: Fall 2023, Spring 2023, Winter 2023, Fall 2022, Winter 2021, Fall 2020, Winter 2020, Fall 2019, Spring 2019, Winter 2019, Fall 2018, Spring 2018, Winter 2018, Fall 2017, Spring 2017, Winter 2017, Fall 2016.

Upcoming quarters: Spring 2024.

Ordering Food: For suggestions for those ordering food for the lunch, see here.

7/9: Making Databases Robust and Reliable: from SQL Generation to Test-Oracle Construction

Time: Tuesday, July 9, 2024, 12 noon - 1pm
Location: Gates 415

Speaker: Zu-Ming Jiang

Abstract: Database management systems (DBMSs) provide fundamental functionalities, such as storing, manipulating, and querying data, for modern data-driven software. Their reliability critically impacts any downstream applications; bugs in DBMSs can lead to serious consequences such as data loss and leakage. In this talk, I will present our recent work on improving the reliability of DBMSs via automated testing. We have introduced a series of techniques spanning from effective test-case generation (i.e., how to generate complex SQL queries for testing deep DBMS logic) to general test-oracle construction (i.e., how to build referencing results for validating the correctness of general SQL queries). Our work has helped improve the reliability of widely used production DBMSs like MySQL, PostgreSQL, and SQLite by exposing hundreds of bugs in them that cause crashes, incorrect query results, and non-compliant transaction executions. We have open-sourced our tools to facilitate further research and benefit the large DBMS community. Bio: Zu-Ming Jiang is a PhD student at ETH Zurich advised by Zhendong Su. His research interests include databases, operating systems, and computer security. His work has appeared in top systems and security venues such as OSDI, USENIX Security, and NDSS. His dissertation research focuses on developing novel, practical techniques for improving the reliability of database management systems (DBMSs).

Food:

7/25: DeCl: Deterministic and Metered Native Sandboxes

Time: Thursday, July 25, 2024, 12 noon - 1pm
Location: Gates 415

Speaker: Zachary Yedidia

Abstract: This talk will present Deterministic Client (DeCl), a software sandboxing system for ARM64 that runs untrusted machine code at near-native speeds while guaranteeing deterministic execution and termination before a specified instruction count. DeCl derives its security from a trusted machine code verifier, resulting in a small and simple trusted computing base (TCB). One key application is a smart contract engine in a digital currency system, which requires both deterministic execution and metered runtime. Today's blockchains, for example, run bytecode in a virtual machine, such as WebAssembly or the Ethereum Virtual Machine, which either incurs the extremely high overhead of a software interpreter or expands the TCB to include a just-in-time compiler, while still paying significant runtime overhead. Instead, with DeCl we can directly execute smart contracts as native ARM64 programs, after validating them with a simple and efficient verification pass that ensures memory isolation, determinism, and deterministic metering.

Food:

8/1: TBD

Time: Thursday, August 1, 2024, 12 noon - 1pm
Location: Gates 415

Speaker: Rohan Yadav

Abstract: TBD

Food: