Pasar al contenido principal

SOFTWARE DE SISTEMA PARA ARQUITECTURAS Y APLICACIONES DE NUEVA GENERACION

PID2021-126576NB-I00

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2021
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD COMPLUTENSE DE MADRID

Resultados relacionados

Resultados totales (Incluyendo duplicados): 12
Encontrada(s) 1 página(s)

Docta Complutense
Publicación Artículo científico (JournalArticle)

Flexible system software scheduling for asymmetric multicore systems with PMCSched: A case for Intel Alder Lake

  • Bilbao, Carlos
  • Sáez Alcaide, Juan Carlos
  • Prieto Matías, Manuel
Asymmetric multicore processors (AMPs) couple high-performance big cores and power-efficient small ones, all exposing a shared instruction set architecture to software, but with different microarchitectural features. The energy efficiency benefits of AMPs, together with the general-purpose nature of the various cores, have led hardware manufacturers to build commercial AMP-based products, first for the mobile and embedded domains, and more recently, for the desktop market segment, as with the Intel Alder Lake processor family. This trend indicates that AMPs may become a solid and more energy efficient replacement for symmetric multicores in a wide range of application domains. Previous research has demonstrated that the system software can substantially improve scheduling—critical to get the most out of heterogeneous cores—by leveraging hardware facilities that are directly managed by the OS, such as performance monitoring counters, or the recently introduced Intel Thread Director technology. Unfortunately, the OS-level support enabling access to these hardware facilities may often take a long time to be adopted in operating systems, or may come in forms that make its utilization challenging from specific levels of the system software stack, especially in production systems. To fill this gap, we propose PMCSched, an open-source framework enabling rapid development and evaluation of custom scheduling-related support in the Linux kernel. PMCSched greatly simplifies the design and implementation of a wide range of scheduling policies for multicore systems that operate at different system software layers without requiring to patch the kernel. To demonstrate the potential of our framework, we conduct a set of experimental case studies on asymmetry-aware scheduling for Intel Alder Lake processors.
Docta Complutense
Publicación Contribución de congreso (ConferenceOutput)

Exploiting Elasticity via OS-Runtime Cooperation to Improve CPU Utilization in Multicore Systems

  • Rubio, Javier
  • Bilbao, Carlos
  • Sáez Alcaide, Juan Carlos
  • Prieto Matías, Manuel
The chip multicore processor (CMP) architecture has become the predominant design choice for contemporary general-purpose systems across multiple sectors of commercial technology. Thanks to technological progress, CMP systems can now feature hundreds of cores. While multithreaded applications may potentially benefit from the increasing core counts, leveraging all available cores is not always feasible due to limited Thread-Level Parallelism (TLP), load imbalance among threads, and other scalability bottlenecks. Colocating multiple applications on the same node is becoming a popular practice to maximize processor utilization. In HPC, malleability -the ability to dynamically alter the number of active threads within the same application-, is also being exploited at the runtime-system level to better deal with scenarios exhibiting time-varying scalability. In the cloud, application colocation is leveraged along with different forms of coarse-grained elasticity to cater to the varying resource demands. This work introduces an operating system (OS) level elastic mechanism designed to efficiently leverage idle CPU periods in workloads consisting of unmodified applications, many of which do not rely on a runtime system to function. This mechanism constitutes a form of fine-grained vertical elasticity that leverages cooperation between the runtime sys-tem and the OS to maximize CPU utilization. To this end, it opportunistically increases the active thread count of mal-leable applications during idle periods. We implemented our proposed OS extensions in the Linux kernel, and augmented the GNU's OpenMP runtime to show a proof of concept of the required OS-runtime interaction. By using diverse multi- threaded programs, we demonstrate the ability of the proposed OS support to substantially improve the system throughput.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Anatomy of the BLIS family of algorithms for matrix multiplication

  • Castelló, Adrián|||0000-0002-8576-8451
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
  • Igual, Francisco D.
[EN] The efforts of the scientific community and hardware vendors to develop and optimize linear algebra codes have historically led to highly-tuned libraries, carefully adapted to the underlying processor architecture, with excellent (near-peak) performance. These optimization efforts, however, are commonly focused on obtaining the best performance possible when the involved operands are large and "squarish" matrices. New computationally-intensive applications (e.g., in deep learning) are increasingly demanding high-performance BLAS (Basic Linear Algebra Subprograms) also for small operands in any of their dimensions. In this paper, we tackle this problem by refactoring the general matrix-matrix multiplication (GEMM) algorithm within a specific high-performance implementation of BLAS, named BLIS, proposing a complete family of algorithmic variants to implement GEMM with different strategies to exploit the target cache hierarchy, together with the changes to be applied to architecture-specific codes to instantiate a complete GEMM implementation. Experimental results on an ARM processor (NVIDIA Carmel) reveal significant performance differences between the members of the GEMM family, depending on the shape and dimension of the matrix operands., This work was supported by the research project PID2020-113656RB-C22 of MCIN/AEI/10.13039/501100011033, EU (FEDER), the Spanish MINECO, under grants RTI2018-093684-B-I00, PID2021-126576NB-I00, and CM via Multiannual Agreement with Complutense University in the line Program to Stimulate Research for Young Doctors in the context of the V PRICIT under projects PR65/19-22445 and CM S2018/TCS-4423. A. Castello is a FJC2019-039222-I fellow supported by MCIN/AEI/10.13039/501100011033.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Algorithm 1039: Automatic Generators for a Family of Matrix Multiplication Routines with Apache TVM

  • Alaejos-López, Guillermo
  • Igual, Francisco D.
  • Martínez, Héctor
  • Castelló, Adrián|||0000-0002-8576-8451
  • Alonso-Jordá, Pedro|||0000-0002-6882-6592
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
This work was supported by the research projects PID2020-113656RB-C22 (MCIN/AEI/10.13039/ 501100011033) and PID2021-126576NB-I00; and CM via Multiannual Agreement with Complutense University in the line Program to Stimulate Research for Young Doctors in the context of the V PRICIT under projects PR65/19-22445 and CM S2018/TCS-4423. A. Castello is a FJC2019-039222-I fellow supported by MCIN/AEI/10.13039/ 501100011033. H. Martinez is a postdoctoral fellow supported by the Consejeria de Transformacion Economica, Industria, Conocimiento y Universidades de la Junta de Andalucia. This project has received funding from the European High-Performance Computing Joint Undertaking (JU) under grant agreement No 955558 (eFlows4HPC project). The JU receives support from the European Union's Horizon 2020 research and innovation programme, and Spain, Germany, France, Italy, Poland, Switzerland, and Norway.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Micro-kernels for portable and efficient matrix multiplication in deep learning

  • Alaejos-López, Guillermo
  • Martínez, Héctor
  • Igual, Francisco D.
  • Castelló, Adrián|||0000-0002-8576-8451
  • Alonso-Jordá, Pedro|||0000-0002-6882-6592
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
[EN] We provide a practical demonstration that it is possible to systematically generate a variety of high-performance micro-kernels for the general matrix multiplication (gemm) via generic templates which can be easily customized to different processor architectures and micro-kernel dimensions. These generic templates employ vector intrinsics to exploit the SIMD (single instruction, multiple data) units in current general-purpose processors and, for the particular type of gemm problems encountered in deep learning, deliver a floating-point throughput rate on par with or even higher than that obtained with conventional, carefully tuned implementations of gemm in current linear algebra libraries (e.g., BLIS, AMD AOCL, ARMPL). Our work exposes the structure of the template-based micro-kernels for ARM Neon (128-bit SIMD), ARM SVE (variable-length SIMD) and Intel AVX512 (512-bit SIMD), showing considerable performance for an NVIDIA Carmel processor (ARM Neon), a Fujitsu A64FX processor (ARM SVE) and on an AMD EPYC 7282 processor (256-bit SIMD)., Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. This work was supported by the research projects PID2020-113656RB-C22, RTI2018-093684-B-I00, and PID2021-126576NB-I00, of MCIN/AEI/10.13039/501100011033 and by "ERDF A way of making Europe", and CM via Multiannual Agreement with Complutense University in the line Program to Stimulate Research for Young Doctors in the context of the V PRICIT under projects PR65/19-22445 and CM S2018/TCS-4423. A. Castello is a FJC2019-039222-I fellow supported by MCIN/AEI/10.13039/501100011033. H. Martinez is a postdoctoral fellow supported by the Consejeria de Transformacion Economica, Industria, Conocimiento y Universidades de la Junta de Andalucia.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Automatic generation of ARM NEON micro-kernels for matrix multiplication

  • Alaejos-López, Guillermo
  • Martínez, Héctor
  • Dolz Zaragozá, Manuel
  • Igual Peña, Francisco D.
  • Castelló, Adrián|||0000-0002-8576-8451
  • Alonso-Jordá, Pedro|||0000-0002-6882-6592
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
[EN] General matrix multiplication (gemm) is a fundamental kernel in scientifc computing and current frameworks for deep learning. Modern realisations of gemm are mostly written in C, on top of a small, highly tuned micro-kernel that is usually encoded in assembly. The high performance realisation of gemm in linear algebra libraries in general include a single micro-kernel per architecture, usually implemented by an expert. In this paper, we explore a couple of paths to automatically generate gemm micro-kernels, either using C++ templates with vector intrinsics or high-level Python scripts that directly produce assembly code. Both solutions can integrate high performance software techniques, such as loop unrolling and software pipelining, accommodate any data type, and easily generate micro-kernels of any requested dimension. The performance of this solution is tested on three ARM-based cores and compared with state-of-the-art libraries for these processors: BLIS, OpenBLAS and ArmPL. The experimental results show that the auto-generation approach is highly competitive, mainly due to the possibility of adapting the micro-kernel to the problem dimensions., Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. European Commission, European Union, 95555. Junta de Andalucía, POSTDOC_21_00025 Agencia Estatal de Investigación, FJC2019-039222, PID2020-113656R, PID2021-12657NB-I00. Generalitat Valenciana, CIDEXG/2022/013, PROMETEO 2023-CIPROM/2022/20
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Programming parallel dense matrix factorizations and inversion for new-generation NUMA architectures

  • Catalán, Sandra
  • Igual, Francisco D.
  • Herrero, José R.
  • Rodríguez-Sánchez, Rafael
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
[EN] We propose a methodology to address the programmability issues derived from the emergence of newgeneration shared-memory NUMA architectures. For this purpose, we employ dense matrix factorizations and matrix inversion (DMFI) as a use case, and we target two modern architectures (AMD Rome and Huawei Kunpeng 920) that exhibit configurable NUMA topologies. Our methodology pursues performance portability across different NUMA configurations by proposing multi-domain implementations for DMFI plus a hybrid task- and loop-level parallelization that configures multi-threaded executions to fix core-todata binding, exploiting locality at the expense of minor code modifications. In addition, we introduce a generalization of the multi-domain implementations for DMFI that offers support for virtually any NUMA topology in present and future architectures. Our experimentation on the two target architectures for three representative dense linear algebra operations validates the proposal, reveals insights on the necessity of adapting both the codes and their execution to improve data access locality, and reports performance across architectures and inter- and intra-socket NUMA configurations competitive with state-of-the-art message-passing implementations, maintaining the ease of development usually associated with shared-memory programming., This research was sponsored by project PID2019-107255GB of Ministerio de Ciencia, Innovacion y Universidades; project S2018/TCS-4423 of Comunidad de Madrid; project 2017-SGR-1414 of the Generalitat de Catalunya and the Madrid Government under the Multiannual Agreement with UCM in the line Pro-gram to Stimulate Research for Young Doctors in the context of the V PRICIT, project PR65/19-22445. This project has also re-ceived funding from the European High-Performance Computing Joint Undertaking (JU) under grant agreement No 955558. The JU receives support from the European Union's Horizon 2020 research and innovation programme, and Spain, Germany, France, Italy, Poland, Switzerland, Norway. The work is also supported by grants PID2020-113656RB-C22 and PID2021-126576NB-I00 of MCIN/AEI/10.13039/501100011033 and by ERDF A way of making Europe.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Cache-aware optimization of matrix multiplication and matrix factorizations on multicore processors

  • Martínez-Pérez, Héctor
  • Catalán, Sandra
  • Igual, Francisco D.
  • Herrero, José R.
  • Rodríguez-Sánchez, Rafael
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
[EN] This paper advocates for a careful customization of the special general matrix multiplication (GEMM) kernels that are invoked from blocked routines for several relevant matrix factorizations in LAPACK, in order to improve their performance on modern multicore processors with hierarchical cache memories. To achieve this, we leverage a refined analytical model to dynamically tune the cache configuration parameters of GEMM for these kernels, taking into account the matrix operands' dimensions, in order to improve cache occupation. In addition, toward the same goal, we accommodate a flexible development of architecture-specific micro-kernels for GEMM that allows us to select the option that, depending on the operands' dimensions, ameliorates cache utilization. Our experiments for the LU and QR factorizations on two platforms, equipped with ARM (NVIDIA Carmel) and x86 (AMD EPYC) multi-core processors, demonstrate the benefits of this approach in terms of a better cache utilization and, in general, higher performance. Moreover, they also reveal the delicate balance between optimizing for multi-threaded parallelism versus cache usage as well as the positive effects of software prefetching., This work was supported by grants PID2020-113656RB-C22, PID2019-107255GB, PID2021-126576NB-I00 and PID2021-123627OB-C52 of MCIN/AEI/10.13039/501100011033, by "ERDF A way of making Europe", and 2021-SGR-01007 of the Generalitat de Catalunya. Hector Martinez is a postdoctoral fellow supported by the Consejeria de Transformacion Economica, Industria, Conocimiento y Universidades de la Junta de Andalucia. Sandra Catalan was supported by the grant RYC2021-033973- I, funded by MCIN/AEI/10.13039/501100011033 and the European Union "NextGenerationEU"/PRTR.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

The Cambrian Explosion of Mixed-Precision Matrix Multiplication for Quantized Deep Learning Inference

  • Martínez, Héctor
  • Igual, Francisco D.
  • Castelló, Adrián|||0000-0002-8576-8451
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
[EN] Recent advances in deep learning (DL) have promoted to a shift from traditional 64-bit floating point (FP64) arithmetic for scientific computing toward reduced-precision formats-such as FP16, BF16, or even 8-bit integers-combined with mixed-precision arithmetic. This transition enhances computational throughput, reduces memory and bandwidth usage, and improves energy efficiency, offering significant advantages for resource-constrained edge devices. To support this shift, hardware architectures have evolved accordingly, now including adapted ISAs (Instruction Set Architectures) that expose mixed-precision vector units and matrix engines tailored for DL workloads. At the heart of many DL and scientific computing tasks is the general matrix-matrix multiplication (GEMM), a fundamental kernel historically optimized using fused multiply-add (FMA) vector instructions on SIMD (single instruction, multiple data) units. However, as hardware moves toward mixed-precision dot (or inner)-product-centric operations optimized for quantized inference, these legacy approaches are being phased out. In response to this, our paper revisits the conventional, high-performance implementation of GEMM and describes strategies for adapting it to mixed integer precision (MIP) arithmetic across modern ISAs, including x86_64, Arm, and RISC-V. Concretely, we illustrate novel micro-kernel designs and data layouts that better exploit today's specialized hardware and demonstrate significant performance gains from MIP arithmetic over floatingpoint implementations across three representative CPUs. These contributions highlight a new era of GEMM optimization-driven by the demands of DL inference on heterogeneous architectures, marking what we term as the "Cambrian period" for matrix multiplication., This work is supported by grants PID2021-126576NB-I00, PID2023-146569NB-C2 and PID2024-158311NB-I00, funded by MCIN/AEI/10.13039/5011000 11033, by "ERDF A way of making Europe", Programa PROMETEO2023 (CIPROM/2022/20) by Conselleria de Innovacio, Universidades, Ciencia y Sociedad Digital from the Generalitat Valenciana, and by Ayudaa Primeros Proyectosde Investigacion (PAID-06-24) , Vicerrectorado de Investigacion de la Universitat Politecnica de Valencia. A. Castello is supported by the CIAPOS/2023/431 grant, and Fondo Social Europeo Plus 2021-2027 from the Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Experience-guided, mixed-precision matrix multiplication with apache TVM for ARM processors

  • Castelló, Adrián|||0000-0002-8576-8451
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
  • Martínez, Héctor
  • Catalán, Sandra
  • Igual Peña, Francisco D.
[EN] Deep learning (DL) generates new computational tasks that are different from those encountered in classical scientific applications. In particular, DL training and inference require general matrix multiplications (gemm) with matrix operands that are far from large and square as in other scientific fields. In addition, DL models gain arithmetic/storage complexity, and as a result, reduced precision via quantization is now mainstream for inferring DL models in edge devices. Automatic code generation addresses these new types of gemm by (1) improving portability between different hardware with only one base code; (2) supporting mixed and reduced precision; and (3) enabling auto-tuning methods that, given a base operation, perform a (costly) optimization search for the best schedule. In this paper, we rely on Apache TVM to generate an experience-guided gemm that provides performance competitive with the TVM auto-scheduler, while reducing tuning time by a factor of 48x., This work received funding from projects PID2020-113656RB and PID2021-126576NB-I00 of MCIN/AEI/ 10.13039/501100011033; PROMETEO 2023-CI PROM/2022/20. H. Martinez is a POSTDOC_21_00025 postdoctoral fellow supported by Junta de Andalucia. S. Catalan is supported by the grant RYC2021-033973-I, funded by MCIN/AEI/10. 13039/501100011033 and the "NextGenerationEU"/ PRTR, and UJI-2023-04, funded by Universitat Jaume I.
Docta Complutense
Publicación Artículo científico (JournalArticle)

Divide&Content: A Fair OS-Level Resource Manager for Contention Balancing on NUMA Multicores

  • Bilbao, Carlos
  • Sáez Alcaide, Juan Carlos
  • Prieto Matías, Manuel
Chip multicore processors (CMPs) constitute the cherry-picked architecture for high-performance servers employed in supercomputers and cloud datacenters. In the last few years, Non-Uniform Memory Access (NUMA) multicore systems have become the dominant choice in these domains. Regardless of the technology advances enabling to pack an increasing number of cores and bigger caches on the same chip, contention for shared resources still represents an important challenge for the system software. Cores in CMPs typically share multiple resources, such as the last-level cache (LLC) or a DRAM controller. The competition for the usage of these resources leads to uneven performance degradation across co-running applications. Previous research has demonstrated that contention effects on CMPs can be mitigated via smart partitioning of the LLC or by distributing threads across groups of cores so as to even out the degree of competition on multiple LLCs or memory nodes. However, most existing resource-management strategies fail to effectively combine both contention-mitigating techniques, thus providing suboptimal results on NUMA multicores. In this paper, we analyze how to best combine these techniques to improve system-wide fairness, and, based on the conclusions of our analysis, propose a fair OS-level NUMA-aware resource manager that leverages dynamic contention-aware thread-to-socket mappings and cache-partitioning. We implemented our resource manager in the Linux kernel and assessed its effectiveness on a real dual-socket system featuring Intel Skylake processors. Our results show that it reduces unfairness by more than 17% on average compared to Linux and a state-of-the-art NUMA-aware resource manager.
Docta Complutense
Publicación Artículo científico (JournalArticle)

Solving the task scheduling and GPU reconfiguration problem on MIG devices via deep reinforcement learning

  • Villarrubia Elvira, Jorge
  • Costero Valero, Luis María
  • Igual Peña, Francisco Daniel
  • Olcoz Herrero, Katzalin
Recent advances in dynamic GPU partitioning, such as NVIDIA's Multi-Instance GPU (MIG) technology, have enhanced resource utilization by enabling task co-execution without contention. However, existing MIG schedulers remain limited to static or task-agnostic methods that sacrifice optimality for tractability. This paper presents a Deep Reinforcement Learning framework that seeks to minimize the completion time of a task queue by holistically addressing the dimensions of the problem: task molding, GPU reconfiguration and execution order. To manage the vast solution space, we apply optimizations such as discrete and canonical representation of states, unification of equivalent configurations, action masking, or promoting the exploration of reconfigurations; this offers insights for similar resource management scenarios. The proposed models are extensively evaluated with widely used benchmarks of the Rodinia and Altis suites, and synthetic workloads generated to emulate a wide range of plausible real situations. The final model improves to the state-of-the-art, especially in workloads that clearly contradict the assumptions of previous proposals, achieving a difference of less than 20% to the optimum. Additionally, two different approaches to the problem are faced (offline vs. online), discussing their theoretical advantages and disadvantages, and evaluating them experimentally for the final model.