multi-topic, publish from gh-pages branch

site/data/cloud-edge/papers/IPDPS-2025-digest.yaml

@@ -0,0 +1,65 @@
+venue: IPDPS
+year: 2025
+date: "2025-05-19"
+tags: [parallel-computing, hpc, distributed-systems, gpu]
+
+selected:
+  - dblp_key: "conf/ipps/CincaRSV0025"
+    title: "Enhancing OmpSs-2 Suspendable Tasks by Combining Operating System and User-Level Threads with C++ Coroutines"
+    tldr: "Extends the OmpSs-2 task-based runtime with C++ coroutines to implement suspendable tasks that can yield while blocked on I/O or communication without stalling the OS thread."
+    why_notable: "Suspendable tasks are a key missing primitive for overlapping computation and communication in task-graph runtimes; the hybrid OS/user-level thread design avoids the overhead of full context switches while remaining portable, with broad implications for OpenMP-style programming on modern heterogeneous nodes."
+
+  - dblp_key: "conf/ipps/ColemanK25"
+    title: "PISA: An Adversarial Approach to Comparing Task Graph Scheduling Algorithms"
+    tldr: "Introduces an adversarial instance-generation framework that automatically synthesizes task graphs that expose worst-case performance gaps between competing scheduling heuristics."
+    why_notable: "Benchmark-driven comparison of DAG schedulers is notoriously biased toward whoever designed the benchmark; PISA's adversarial synthesis provides a principled, algorithm-agnostic methodology that could become a standard evaluation tool for the task-scheduling community."
+
+  - dblp_key: "conf/ipps/FradetGH25"
+    title: "Parallel Scheduling of Task Graphs with Minimal Memory Requirements"
+    tldr: "Proves tight bounds and provides scheduling algorithms for task graphs that minimize peak memory usage while preserving parallelism, targeting memory-constrained accelerators."
+    why_notable: "Peak memory is increasingly the binding constraint on accelerators with fixed HBM capacity; deriving schedules that are jointly memory-optimal and parallel addresses a theoretically hard trade-off that directly impacts large-scale neural network and sparse-solver pipelines."
+
+  - dblp_key: "conf/ipps/BinderSSL25"
+    title: "FATHOM: Fast Attention Through Optimizing Memory"
+    tldr: "Redesigns the attention kernel memory access pattern to maximize reuse across the Q, K, and V tiles, achieving significant throughput gains over FlashAttention on modern GPUs."
+    why_notable: "Attention is the dominant compute bottleneck in transformer inference and training; FATHOM's memory-centric reformulation improves on the widely deployed FlashAttention baseline and demonstrates that analytical cache-tile reasoning still yields practical gains at scale."
+
+  - dblp_key: "conf/ipps/FallinADCB25"
+    title: "Fast and Effective Lossy Compression on GPUs and CPUs with Guaranteed Error Bounds"
+    tldr: "Presents a portable error-bounded lossy compressor that runs natively on both GPU and CPU, delivering competitive compression ratios with strict point-wise error guarantees and low latency."
+    why_notable: "Scientific simulations increasingly need in-situ compression on the same GPU that runs the simulation; providing tight error bounds alongside GPU portability bridges a critical gap between scientific fidelity requirements and storage bandwidth constraints at exascale."
+
+  - dblp_key: "conf/ipps/CarpentieriCBFC25"
+    title: "Phase-Based Frequency Scaling for Energy-Efficient Heterogeneous Computing"
+    tldr: "Dynamically scales CPU and GPU frequencies based on detected application phases to reduce energy consumption while preserving performance on heterogeneous nodes."
+    why_notable: "Energy efficiency is a first-class constraint at exascale; phase-aware DVFS that jointly controls both CPU and GPU frequencies avoids the over-provisioning inherent in static policies, with measured gains directly applicable to production HPC clusters."
+
+  - dblp_key: "conf/ipps/HidayetogluGSSH25"
+    title: "HiCCL: A Hierarchical Collective Communication Library"
+    tldr: "Implements a multi-level collective communication library that decomposes all-reduce and other collectives into hierarchical sub-collectives matched to node-local, intra-rack, and inter-rack bandwidth tiers."
+    why_notable: "Flat NCCL/MPI collectives ignore the hierarchical bandwidth structure of modern GPU clusters; HiCCL's topology-aware decomposition achieves substantial throughput improvements on large distributed training jobs and is designed to be backend-agnostic."
+
+  - dblp_key: "conf/ipps/ChenYXS025"
+    title: "Unified Designs of Multi-Rail-Aware MPI Allreduce and Alltoall Operations Across Diverse GPU and Interconnect Systems"
+    tldr: "Develops unified multi-rail-aware algorithms for MPI Allreduce and Alltoall that exploit all available NIC ports simultaneously across heterogeneous interconnect systems."
+    why_notable: "Multi-rail configurations are increasingly common in HPC clusters yet most MPI libraries treat them as a single logical link; this work systematically closes the performance gap and the unified design enables deployment without per-system hand-tuning."
+
+  - dblp_key: "conf/ipps/PlanetaBRH25"
+    title: "CoRD: Converged RDMA Dataplane"
+    tldr: "Proposes a converged RDMA dataplane that unifies one-sided and two-sided communication semantics over a single kernel-bypass path, reducing software overhead and improving scalability."
+    why_notable: "RDMA stacks remain split between one-sided verbs and two-sided message passing, forcing application writers to choose and preventing optimal use of NIC offload capabilities; CoRD's converged abstraction enables a new class of communication patterns at near-wire speed."
+
+  - dblp_key: "conf/ipps/XiaoZMWZJLZZ25"
+    title: "FlexRLHF: A Flexible Placement and Parallelism Framework for Efficient RLHF Training"
+    tldr: "Introduces a placement and parallelism co-optimization framework for Reinforcement Learning from Human Feedback (RLHF) training that jointly schedules the actor, critic, and reward models to maximize GPU utilization."
+    why_notable: "RLHF is the dominant fine-tuning paradigm for large language models but its multi-model, heterogeneous-workload structure makes naive data/model parallelism highly inefficient; FlexRLHF's co-placement approach delivers meaningful throughput gains and establishes a design template for future RLHF infrastructure."
+
+  - dblp_key: "conf/ipps/MuneraQR25"
+    title: "GuardianOMP: A Framework for Highly Productive Fault Tolerance Via OpenMP Task-Level Replication"
+    tldr: "Adds transparent task-level redundant execution to OpenMP applications, enabling automatic detection and recovery from silent data corruption without application source changes."
+    why_notable: "Silent data corruption is an escalating concern as DRAM and compute elements scale into billions of transistors; GuardianOMP's integration at the OpenMP runtime level makes resilience accessible to the broad HPC community that already uses OpenMP without requiring manual checkpoint/restart logic."
+
+  - dblp_key: "conf/ipps/SalwasserSG025"
+    title: "Tera-Scale Multilevel Graph Partitioning"
+    tldr: "Scales multilevel graph partitioning to trillion-edge graphs through a distributed coarsening and refinement pipeline that maintains partition quality competitive with state-of-the-art tools on billion-edge benchmarks."
+    why_notable: "Graph partitioning is a prerequisite for almost every distributed graph workload; reaching the tera-scale regime with near-optimal quality is a significant algorithmic and engineering milestone that directly enables graph-parallel simulation at the largest current HPC scales."