Searcharxiv⌕ Search

arXiv subjects

Xinmin Tian

Publications and source records attributed to Xinmin Tian.

3 recordsLinked to original sources

SyclKittens: A Tile Programming Model for Programmers and Coding Agents on Intel GPUs

New AI accelerators arrive before the kernels that make them fast, because peak kernel performance requires architecture-specific expertise in operand pipelines and data-movement techniques. Coding agents can now write, compile, and tune kernels on their own, so they could greatly accelerate kernel development and optimization. What agents produce depends on the interfaces they are given. These interfaces may expose a machine's raw capabilities or encode the known-good methods for exploiting its hardware features efficiently. We test the performance impact of these two interfaces on Intel GPUs through controlled experiments with five coding models on three kernels. With raw SYCL and execution feedback, Opus 4.8, the strongest of the three models tested with Claude Code, writes a single-GPU GEMM kernel reaching only 57.5% of Intel's tuned oneDNN library. We present SyclKittens, a hardware-aware tile programming model for Intel GPUs that encodes these known-good methods, so its operations place operands in matrix-engine layouts, prefetch through the L1 cache, and communicate over the fabric that links GPU stacks into a node. With SyclKittens under the same agent, task, and feedback budget, the agent reaches 82.1%, showing that encoded methods turn hardware capabilities into performance. Workload-specific policies stay programmable, and engineers and agents jointly refine these schedules in SyclKittens to build a kernel suite that reaches approximately 96% of oneDNN in geometric mean across GEMM shapes. The suite runs Llama-3.1-8B inference 1.59 times faster than torch.compile on one GPU and up to 2.91 times faster than a matched multi-GPU decode path on Intel's oneCCL. SyclKittens is open source and available at https://github.com/intel/SyclKittens.

cs.DC↗

When Calibration Data Matter in Sub-4-Bit Language Model Quantization

Low-bit post-training quantization reduces language-model memory but can sharply degrade downstream task accuracy. Calibration usually relies on generic text, but when task-informed data recover accuracy and which quantizers benefit remain poorly understood. We introduce FragCal-DC, which supplements generic text with training documents from task families with low retention of full-precision accuracy in a two-bit scan. We use a two- to four-bit sweep to locate functional regimes and four-seed calibration comparisons to evaluate six data-dependent methods across six dense models (1.7B-8B) and ten tasks. At their lowest tested functional bit widths, iterative SignRound V2 and OmniQuant improve at two bits, whereas single-pass GPTQ and AWQ show small mean responses at three bits. At two bits, SignRound V2 gains 1.83 percentage points in mean ten-task accuracy over generic-text calibration across six models. At 70 billion parameters, GPTQ and SignRound V2 both improve at two bits, with functional generic-calibration baselines. For two-bit SignRound V2, task-level controls show that recovery concentrates on families included in calibration, while incorrect answer designations retain about 90% of the mean accuracy gain. Changes in reconstruction error do not consistently track accuracy gains. These findings support task-informed calibration as a design principle for preserving downstream accuracy in low-bit quantization.

cs.LG↗

Aurora: Architecting Argonne's First Exascale Supercomputer for Accelerated Scientific Discovery

Aurora is Argonne National Laboratory's pioneering Exascale supercomputer, designed to accelerate scientific discovery with cutting-edge architectural innovations. Key new technologies include the Intel(TM) Xeon(TM) Data Center GPU Max Series (code-named Sapphire Rapids) with support for High Bandwidth Memory (HBM), alongside the Intel(TM) Data Center GPU Max Series (code-named Ponte Vecchio) on each compute node. Aurora also integrates the Distributed Asynchronous Object Storage (DAOS), a novel exascale storage solution, and leverages Intel's oneAPI programming environment. This paper presents an in-depth exploration of Aurora's node architecture, the HPE Slingshot interconnect, the supporting software ecosystem, and DAOS. We provide insights into standard benchmark performance and applications readiness efforts via Aurora's Early Science Program and the Exascale Computing Project.

cs.DC↗