اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدAccelerating BLAS on Custom Architecture through Algorithm-Architecture Co-design
99
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Basic Linear Algebra Subprograms (BLAS) play key role in high performance and scientific computing applications. Experimentally, yesteryear multicore and General Purpose Graphics Processing Units (GPGPUs) are capable of achieving up to 15 to 57% of the theoretical peak performance at 65W to 240W respectively for compute bound operations like Double/Single Precision General Matrix Multiplication (XGEMM). For bandwidth bound operations like Single/Double precision Matrix-vector Multiplication (XGEMV) the performance is merely 5 to 7% of the theoretical peak performance in multicores and GPGPUs respectively. Achieving performance in BLAS requires moving away from conventional wisdom and evolving towards customized accelerator tailored for BLAS through algorithm-architecture co-design. In this paper, we present acceleration of Level-1 (vector operations), Level-2 (matrix-vector operations), and Level-3 (matrix-matrix operations) BLAS through algorithm architecture co-design on a Coarse-grained Reconfigurable Architecture (CGRA). We choose REDEFINE CGRA as a platform for our experiments since REDEFINE can be adapted to support domain of interest through tailor-made Custom Function Units (CFUs). For efficient sequential realization of BLAS, we present design of a Processing Element (PE) and perform micro-architectural enhancements in the PE to achieve up-to 74% of the theoretical peak performance of PE in DGEMM, 40% in DGEMV and 20% in double precision inner product (DDOT). We attach this PE to REDEFINE CGRA as a CFU and show the scalability of our solution. Finally, we show performance improvement of 3-140x in PE over commercially available Intel micro-architectures, ClearSpeed CSX700, FPGA, and Nvidia GPGPUs.
قيم البحث
اقرأ أيضاً
Nanopore genome sequencing is the key to enabling personalized medicine, global food security, and virus surveillance. The state-of-the-art base-callers adopt deep neural networks (DNNs) to translate electrical signals generated by nanopore sequencer
s to digital DNA symbols. A DNN-based base-caller consumes $44.5%$ of total execution time of a nanopore sequencing pipeline. However, it is difficult to quantize a base-caller and build a power-efficient processing-in-memory (PIM) to run the quantized base-caller. In this paper, we propose a novel algorithm/architecture co-designed PIM, Helix, to power-efficiently and accurately accelerate nanopore base-calling. From algorithm perspective, we present systematic error aware training to minimize the number of systematic errors in a quantized base-caller. From architecture perspective, we propose a low-power SOT-MRAM-based ADC array to process analog-to-digital conversion operations and improve power efficiency of prior DNN PIMs. Moreover, we revised a traditional NVM-based dot-product engine to accelerate CTC decoding operations, and create a SOT-MRAM binary comparator array to process read voting. Compared to state-of-the-art PIMs, Helix improves base-calling throughput by $6times$, throughput per Watt by $11.9times$ and per $mm^2$ by $7.5times$ without degrading base-calling accuracy.
We present efficient realization of Householder Transform (HT) based QR factorization through algorithm-architecture co-design where we achieve performance improvement of 3-90x in-terms of Gflops/watt over state-of-the-art multicore, General Purpose
Graphics Processing Units (GPGPUs), Field Programmable Gate Arrays (FPGAs), and ClearSpeed CSX700. Theoretical and experimental analysis of classical HT is performed for opportunities to exhibit higher degree of parallelism where parallelism is quantified as a number of parallel operations per level in the Directed Acyclic Graph (DAG) of the transform. Based on theoretical analysis of classical HT, an opportunity re-arrange computations in the classical HT is identified that results in Modified HT (MHT) where it is shown that MHT exhibits 1.33x times higher parallelism than classical HT. Experiments in off-the-shelf multicore and General Purpose Graphics Processing Units (GPGPUs) for HT and MHT suggest that MHT is capable of achieving slightly better or equal performance compared to classical HT based QR factorization realizations in the optimized software packages for Dense Linear Algebra (DLA). We implement MHT on a customized platform for Dense Linear Algebra (DLA) and show that MHT achieves 1.3x better performance than native implementation of classical HT on the same accelerator. For custom realization of HT and MHT based QR factorization, we also identify macro operations in the DAGs of HT and MHT that are realized on a Reconfigurable Data-path (RDP). We also observe that due to re-arrangement in the computations in MHT, custom realization of MHT is capable of achieving 12% better performance improvement over multicore and GPGPUs than the performance improvement reported by General Matrix Multiplication (GEMM) over highly tuned DLA software packages for multicore and GPGPUs which is counter-intuitive.
In this paper, we present efficient realization of Kalman Filter (KF) that can achieve up to 65% of the theoretical peak performance of underlying architecture platform. KF is realized using Modified Faddeeva Algorithm (MFA) as a basic building block
due to its versatility and REDEFINE Coarse Grained Reconfigurable Architecture (CGRA) is used as a platform for experiments since REDEFINE is capable of supporting realization of a set algorithmic compute structures at run-time on a Reconfigurable Data-path (RDP). We perform several hardware and software based optimizations in the realization of KF to achieve 116% improvement in terms of Gflops over the first realization of KF. Overall, with the presented approach for KF, 4-105x performance improvement in terms of Gflops/watt over several academically and commercially available realizations of KF is attained. In REDEFINE, we show that our implementation is scalable and the performance attained is commensurate with the underlying hardware resources
Design space exploration is an important but costly step involved in the design/deployment of custom architectures to squeeze out maximum possible performance and energy efficiency. Conventionally, optimizations require iterative sampling of the desi
gn space using simulation or heuristic tools. In this paper we investigate the possibility of learning the optimization task using machine learning and hence using the learnt model to predict optimal parameters for the design and mapping space of custom architectures, bypassing any exploration step. We use three case studies involving the optimal array design, SRAM buffer sizing, mapping, and schedule determination for systolic-array-based custom architecture design and mapping space. Within the purview of these case studies, we show that it is possible to capture the design space and train a model to generalize prediction the optimal design and mapping parameters when queried with workload and design constraints. We perform systematic design-aware and statistical analysis of the optimization space for our case studies and highlight the patterns in the design space. We formulate the architecture design and mapping as a machine learning problem that allows us to leverage existing ML models for training and inference. We design and train a custom network architecture called AIRCHITECT, which is capable of learning the architecture design space with as high as 94.3% test accuracy and predicting optimal configurations which achieve on average (GeoMean) of 99.9% the best possible performance on a test dataset with $10^5$ GEMM workloads.
We present efficient realization of Generalized Givens Rotation (GGR) based QR factorization that achieves 3-100x better performance in terms of Gflops/watt over state-of-the-art realizations on multicore, and General Purpose Graphics Processing Unit
s (GPGPUs). GGR is an improvement over classical Givens Rotation (GR) operation that can annihilate multiple elements of rows and columns of an input matrix simultaneously. GGR takes 33% lesser multiplications compared to GR. For custom implementation of GGR, we identify macro operations in GGR and realize them on a Reconfigurable Data-path (RDP) tightly coupled to pipeline of a Processing Element (PE). In PE, GGR attains speed-up of 1.1x over Modified Householder Transform (MHT) presented in the literature. For parallel realization of GGR, we use REDEFINE, a scalable massively parallel Coarse-grained Reconfigurable Architecture, and show that the speed-up attained is commensurate with the hardware resources in REDEFINE. GGR also outperforms General Matrix Multiplication (gemm) by 10% in-terms of Gflops/watt which is counter-intuitive.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
