venue: HPDC
year: 2024
papers:
- title: Efficient all-to-all Collective Communication Schedules for Direct-connect
    Topologies
  authors:
  - Prithwish Basu
  - Liangyu Zhao
  - Jason Fantl
  - Siddharth Pal
  - Arvind Krishnamurthy
  - Joud Khoury
  reason: "Derives near-optimal all-to-all collective communication schedules for direct-connect HPC topologies, directly improving bandwidth utilization in large-scale distributed systems."
- title: Reinforcement Learning-based Adaptive Mitigation of Uncorrected DRAM Errors
    in the Field
  authors:
  - Isaac Boixaderas
  - Sergi Moré
  - Javier Bartolome
  - David Vicente
  - Petar Radojkovic
  - Paul M. Carpenter
  - Eduard Ayguadé
  reason: "Applies reinforcement learning to dynamically mitigate uncorrected DRAM errors at production HPC scale, improving system reliability without sacrificing performance."
- title: 'IDT: Intelligent Data Placement for Multi-tiered Main Memory with Reinforcement
    Learning'
  authors:
  - Juneseo Chang
  - Wanju Doh
  - Yaebin Moon
  - Eojin Lee
  - Jung Ho Ahn
  reason: "Presents a reinforcement learning-driven runtime that automatically places data across heterogeneous memory tiers, reducing access latency in HPC nodes with complex memory hierarchies."
- title: 'FaaSKeeper: Learning from Building Serverless Services with ZooKeeper as
    an Example'
  authors:
  - Marcin Copik
  - Alexandru Calotoiu
  - Pengyu Zhou
  - Konstantin Taranov
  - Torsten Hoefler
  reason: "Reconstructs ZooKeeper as a fully serverless service and distills concrete design lessons for building stateful distributed coordination primitives on FaaS platforms."
- title: 'ESG: Pipeline-Conscious Efficient Scheduling of DNN Workflows on Serverless
    Platforms with Shareable GPUs'
  authors:
  - Xinning Hui
  - Yuanchao Xu 0001
  - Zhishan Guo
  - Xipeng Shen
  reason: "Introduces a pipeline-aware scheduler that shares GPUs across serverless DNN workflow stages, substantially cutting end-to-end latency and GPU idle time."
- title: 'FASOP: Fast yet Accurate Automated Search for Optimal Parallelization of
    Transformers on Heterogeneous GPU Clusters'
  authors:
  - Sunyeol Hwang
  - Eungyeong Lee
  - Hongseok Oh 0003
  - Youngmin Yi
  reason: "Provides a fast, model-driven search strategy that finds optimal tensor/pipeline/data parallelism configurations for transformer training on heterogeneous GPU clusters."
- title: Near-Optimal Wafer-Scale Reduce
  authors:
  - Piotr Luczynski
  - Lukas Gianinazzi
  - Patrick Iff
  - Leighton Wilson
  - Daniele De Sensi
  - Torsten Hoefler
  reason: "Designs and analyzes near-optimal Reduce collective algorithms tailored to wafer-scale interconnect topology, setting new performance bounds for next-generation HPC hardware."
- title: 'DataStates-LLM: Lazy Asynchronous Checkpointing for Large Language Models'
  authors:
  - Avinash Maurya
  - Robert Underwood
  - M. Mustafa Rafique
  - Franck Cappello
  - Bogdan Nicolae
  reason: "Introduces lazy asynchronous checkpointing that overlaps LLM training with I/O, dramatically reducing checkpoint overhead on large-scale HPC storage systems."
- title: 'ADTopk: All-Dimension Top-k Compression for High-Performance Data-Parallel
    DNN Training'
  authors:
  - Zhangqiang Ming
  - Yuchong Hu
  - Wenxiang Zhou
  - Xinjue Zheng
  - Chenxuan Yao
  - Dan Feng 0001
  reason: "Proposes an all-dimension top-k gradient sparsification scheme that reduces communication volume in data-parallel distributed training while preserving convergence quality."
- title: Accelerating Function-Centric Applications by Discovering, Distributing,
    and Retaining Reusable Context in Workflow Systems
  authors:
  - Thanh Son Phung
  - Colin Thomas
  - Logan T. Ward
  - Kyle Chard
  - Douglas Thain
  reason: "Introduces context reuse across scientific workflow tasks, allowing distributed workflow systems to cache and share intermediate computation artifacts and significantly reduce redundant work."
- title: 'CereSZ: Enabling and Scaling Error-bounded Lossy Compression on Cerebras
    CS-2'
  authors:
  - Shihui Song
  - Yafan Huang
  - Peng Jiang 0004
  - Xiaodong Yu 0001
  - Weijian Zheng
  - Sheng Di
  - Qinglei Cao
  - Yunhe Feng
  - Zhen Xie
  - Franck Cappello
  reason: "Ports and scales error-bounded lossy compression to the Cerebras CS-2 wafer-scale engine, enabling significant data reduction for HPC scientific workloads on novel accelerator hardware."
- title: 'EvoStore: Towards Scalable Storage of Evolving Learning Models'
  authors:
  - Robert Underwood
  - Meghana Madhyastha
  - Randal C. Burns
  - Bogdan Nicolae
  reason: "Designs a storage system that efficiently manages the versioned, incrementally evolving checkpoints produced during large-scale distributed model training, reducing storage overhead and retrieval time."