Benjamin Berg, Assistant Professor, University of North Carolina
A New Methodology for Parallel Job Scheduling
Huaicheng Li, Assistant Professor, Virginia Tech
Towards Predictable and Efficient Datacenter Storage
Lin Ma, Assistant Professor, University of Michigan
Putting Your Database on Autopilot:
Self-driving Database Management Systems
Validating Large Language Models with ReLM
OUTSTANDING PAPER AT MLSYS '23!
Michael Kuchnik, Virginia Smith, George Amvrosiadis
6th MLSys Conference, Miami Beach, FL, USA, June 4-8, 2023.
Although large language models (LLMs) have been touted for their ability to generate natural-sounding text, there are growing concerns around possible negative effects of LLMs such as data memorization, bias, and inappropriate language. Unfortunately, the complexity and generation capacities of LLMs make validating (and correcting) such concerns difficult. In this work, we introduce ReLM, a system for validating and querying LLMs using standard regular expressions. [...more]
Contiguitas: The Pursuit of Physical Memory Contiguity in Datacenters
BEST PAPER AT ISCA '23!
Kaiyang Zhao, Kaiwen Xue, Ziqi Wang, Dan Schatzberg, Leon Yang, Antonis Manousis, Johannes Weiner, Rik van Riel, Bikash Sharma, Chunqiang Tang, Dimitrios Skarlatos
ISCA ’23, June 17–21, 2023, Orlando, FL.
The unabating growth of the memory needs of emerging datacenter applications has exacerbated the scalability bottleneck of virtual memory. However, reducing the excessive overhead of address translation will remain onerous until the physical memory contiguity predicament gets resolved. To address this problem, this paper presents Contiguitas [...more]
GL-Cache: Group-level Learning for Efficient and High-performance Caching
Juncheng Yang, Ziming Mao, Yao Yue, K. V. Rashmi
21st USENIX Conference on File and Storage Technologies (FAST '23). Feb. 21–23, 2023,
Santa Clara, CA.
To adapt to changing workloads, three types of learned caches (learned evictions) have been designed in recent years: object-level learning, learning-from-distribution, and learning-from-simple-experts. However, we argue that the learning granularity in existing approaches is either too fine (object-level), incurring significant computation and storage overheads, or too coarse (workload or expert-level) to capture the differences between objects and leaves a considerable efficiency gap. [...more]
Congratulations to Juncheng and Mohammad, who have been listed among a stellar group of 35 current and recently graduated PhD students, globallyin the inaugural ML Commons Rising Stars cohort!...
Congratulations to Kaiyang Zhao, Dimitrios Skarlatos, and PDL alum Ziqi Wang on winning the Best Paper Award at ISCA '23 this June in Orlando, FL for their paper "Contiguitas: The Pursuit of Physical Memory Contiguity in Datacenters" ...
Outstanding Paper at MLSys 2023!
Congratulations to Micheal Kuchnik, Virginia Smith and George Amvrosiadis on receiving the award for Outstanding Paper at MLSys 2023, held this year in Miami Beach, FL, for their paper "Validating Large Language Models with ReLM." ...