Yang Wang
Ohio State University, OH·Computer Science
Modern internet services run on clusters of thousands of machines, and at that size software breaks in ways it never does on a laptop. A bug may lie dormant in every small test and only surface once a system is deployed across hundreds or thousands of nodes.
- Fields
Computer Science
Engineering
- Topics
distributed systems
scalability faults
large-scale systems
- Grants
- 1 known grant, through Sep 2027
What they work on
- Build a multi-dimensional scale-testing platform that can exercise and reproduce many dimensions of system scale efficiently, enabling large-scale unit tests for faults that only manifest in deployments of hundreds or thousands of nodes.
- Develop program analysis methods that detect scalability bugs, latent defects whose symptoms appear only at large scale, and investigate how to prevent and fix them.
- Create a unified database of software program properties and hardware performance profiles with clear programming interfaces, so that correctness and performance analyses can be automated.
- Learn and predict software performance across heterogeneous devices using detailed device profiling, aiming to predict fine-grained task performance accurately and project it to future architectures.
- Define criteria for certifying correctness checkability and performance predictability of cross-layer software and hardware interactions, toward reproducible- and verified-by-construction development methods.
More about the research
These "scalability faults" are latent, scale-dependent defects, and today there is no systematic way to find them before deployment or to predict how fast software will run on a large, mixed collection of hardware. This collaborative project asks how to make large systems checkable for correctness and predictable in performance.
Why it matters
Cloud services, databases and large computing clusters underpin everyday applications, and faults that appear only at full deployment scale are expensive and hard to diagnose. Tools that expose such bugs during testing and predict performance across heterogeneous and future hardware could make large-scale systems more reliable and more efficient for the companies and users that depend on them.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Milos Gligoric
University of Texas at Austin, TX·Computer Science
Develop a language and framework for writing fine-grained tests that check individual statements and code fragments inside functions, and evaluate it on open-source projects.
Tianyin Xu
University of Illinois Urbana-Champaign, IL·Computer Science·Early-career
Establish the foundations of configuration testing for cloud and datacenter software, a missing piece of how production configuration changes are checked today.
Heike Jagode
University of Tennessee-Knoxville, TN·Computer Science
Add performance and power monitoring for AI hardware to the MINCER infrastructure so researchers can make informed choices about optimizing and configuring AI systems.
The school
Ohio State University: 71 labs on PI Pages, $1.6B a year on research, 882 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.