This paper presents and characterizes the Princeton Application Repository for Shared-Memory Computers (PARSEC), a benchmark suite for studies of Chip-Multiprocessors (CMPs). Previous available benchmarks for multiprocessors have focused on high-performance computing applications and used a limited number of synchronization methods. PARSEC includes emerging applications in recognition, mining and synthesis (RMS) as well as systems applications which mimic large-scale multithreaded commercial programs. Our characterization shows that the benchmark suite covers a wide spectrum of working sets, locality, data sharing, synchronization and off-chip traffic. The benchmark suite has been made available to the public.

/pdf/the-parsec-benchmark-suite-characterization-and-jiajft2rem.pdf

The PARSEC benchmark suite: characterization and architectural implications

This paper presents and characterizes Rodinia, a benchmark suite for heterogeneous computing. To help architects study emerging platforms such as GPUs (Graphics Processing Units), Rodinia includes applications and kernels which target multi-core CPU and GPU platforms. The choice of applications is inspired by Berkeley's dwarf taxonomy. Our characterization shows that the Rodinia benchmarks cover a wide range of parallel communication patterns, synchronization techniques and power consumption, and has led to some important architectural insight, such as the growing importance of memory-bandwidth limitations and the consequent importance of data layout.

/pdf/rodinia-a-benchmark-suite-for-heterogeneous-computing-gib3ik78jh.pdf

Rodinia: A benchmark suite for heterogeneous computing

Newness and distinctiveness is claimed in the features of ornamentation as shown inside the broken line circle in the accompanying representation.

Display screen or portion thereof with graphical user interface

International Technology Roadmap for Semiconductors 2003の要求清浄度について － シリコンウエハ表面と雰囲気環境に要求される清浄度, 分析方法の現状について －

Deep convolutional neural networks (CNNs) are widely used in modern AI systems for their superior accuracy but at the cost of high computational complexity. The complexity comes from the need to simultaneously process hundreds of filters and channels in the high-dimensional convolutions, which involve a significant amount of data movement. Although highly-parallel compute paradigms, such as SIMD/SIMT, effectively address the computation requirement to achieve high throughput, energy consumption still remains high as data movement can be more expensive than computation. Accordingly, finding a dataflow that supports parallel processing with minimal data movement cost is crucial to achieving energy-efficient CNN processing without compromising accuracy.In this paper, we present a novel dataflow, called row-stationary (RS), that minimizes data movement energy consumption on a spatial architecture. This is realized by exploiting local data reuse of filter weights and feature map pixels, i.e., activations, in the high-dimensional convolutions, and minimizing data movement of partial sum accumulations. Unlike dataflows used in existing designs, which only reduce certain types of data movement, the proposed RS dataflow can adapt to different CNN shape configurations and reduces all types of data movement through maximally utilizing the processing engine (PE) local storage, direct inter-PE communication and spatial parallelism. To evaluate the energy efficiency of the different dataflows, we propose an analysis framework that compares energy cost under the same hardware area and processing parallelism constraints. Experiments using the CNN configurations of AlexNet show that the proposed RS dataflow is more energy efficient than existing dataflows in both convolutional (1.4× to 2.5×) and fully-connected layers (at least 1.3× for batch size larger than 16). The RS dataflow has also been demonstrated on a fabricated chip, which verifies our energy analysis.

Eyeriss: a spatial architecture for energy-efficient dataflow for convolutional neural networks

The commonly used LRU replacement policy is susceptible to thrashing for memory-intensive workloads that have a working set greater than the available cache size. For such applications, the majority of lines traverse from the MRU position to the LRU position without receiving any cache hits, resulting in inefficient use of cache space. Cache performance can be improved if some fraction of the working set is retained in the cache so that at least that fraction of the working set can contribute to cache hits.We show that simple changes to the insertion policy can significantly reduce cache misses for memory-intensive workloads. We propose the LRU Insertion Policy (LIP) which places the incoming line in the LRU position instead of the MRU position. LIP protects the cache from thrashing and results in close to optimal hitrate for applications that have a cyclic reference pattern. We also propose the Bimodal Insertion Policy (BIP) as an enhancement of LIP that adapts to changes in the working set while maintaining the thrashing protection of LIP. We finally propose a Dynamic Insertion Policy (DIP) to choose between BIP and the traditional LRU policy depending on which policy incurs fewer misses. The proposed insertion policies do not require any change to the existing cache structure, are trivial to implement, and have a storage requirement of less than two bytes. We show that DIP reduces the average MPKI of the baseline 1MB 16-way L2 cache by 21%, bridging two-thirds of the gap between LRU and OPT.

/pdf/adaptive-insertion-policies-for-high-performance-caching-y9mqbjmlop.pdf

Adaptive insertion policies for high performance caching

Practical cache replacement policies attempt to emulate optimal replacement by predicting the re-reference interval of a cache block. The commonly used LRU replacement policy always predicts a near-immediate re-reference interval on cache hits and misses. Applications that exhibit a distant re-reference interval perform badly under LRU. Such applications usually have a working-set larger than the cache or have frequent bursts of references to non-temporal data (called scans). To improve the performance of such workloads, this paper proposes cache replacement using Re-reference Interval Prediction (RRIP). We propose Static RRIP (SRRIP) that is scan-resistant and Dynamic RRIP (DRRIP) that is both scan-resistant and thrash-resistant. Both RRIP policies require only 2-bits per cache block and easily integrate into existing LRU approximations found in modern processors. Our evaluations using PC games, multimedia, server and SPEC CPU2006 workloads on a single-core processor with a 2MB last-level cache (LLC) show that both SRRIP and DRRIP outperform LRU replacement on the throughput metric by an average of 4% and 10% respectively. Our evaluations with over 1000 multi-programmed workloads on a 4-core CMP with an 8MB shared LLC show that SRRIP and DRRIP outperform LRU replacement on the throughput metric by an average of 7% and 9% respectively. We also show that RRIP outperforms LFU, the state-of the art scan-resistant replacement algorithm to-date. For the cache configurations under study, RRIP requires 2X less hardware than LRU and 2.5X less hardware than LFU.

High performance cache replacement using re-reference interval prediction (RRIP)

As memory accesses become slower with respect to the processor and consume more power with increasing memory size, the focus of memory performance and power consumption has become increasingly important. With the trend to develop multi-threaded, multi-core processors, the demands on the memory system will continue to scale. However, determining the optimal memory system configuration is non-trivial. The memory system performance is sensitive to a large number of parameters. Each of these parameters take on a number of values and interact in fashions that make overall trends difficult to discern. A comparison of the memory system architectures becomes even harder when we add the dimensions of power consumption and manufacturing cost. Unfortunately, there is a lack of tools in the public-domain that support such studies. Therefore, we introduce DRAMsim, a detailed and highly-configurable C-based memory system simulator to fill this gap. DRAMsim implements detailed timing models for a variety of existing memories, including SDRAM, DDR, DDR2, DRDRAM and FB-DIMM, with the capability to easily vary their parameters. It also models the power consumption of SDRAM and its derivatives. It can be used as a standalone simulator or as part of a more comprehensive system-level model. We have successfully integrated DRAMsim into a variety of simulators including MASE [15], Sim-alpha [14], BOCHS[2] and GEMS[13]. The simulator can be downloaded from www.ece.umd.edu/dramsim.

/pdf/dramsim-a-memory-system-simulator-5g7v9o19xh.pdf

DRAMsim: a memory system simulator

Single-ISA heterogeneous multi-core processors are typically composed of small (e.g., in-order) power-efficient cores and big (e.g., out-of-order) high-performance cores. The effectiveness of heterogeneous multi-cores depends on how well a scheduler can map workloads onto the most appropriate core type. In general, small cores can achieve good performance if the workload inherently has high levels of ILP. On the other hand, big cores provide good performance if the workload exhibits high levels of MLP or requires the ILP to be extracted dynamically. This paper proposes Performance Impact Estimation (PIE) as a mechanism to predict which workload-to-core mapping is likely to provide the best performance. PIE collects CPI stack, MLP and ILP profile information, and estimates performance if the workload were to run on a different core type. Dynamic PIE adjusts the scheduling at runtime and thereby exploits fine-grained time-varying execution behavior. We show that PIE requires limited hardware support and can improve system performance by an average of 5.5% over recent state-of-the-art scheduling proposals and by 8.7% over a sampling-based scheduling policy.

/pdf/scheduling-heterogeneous-multi-cores-through-performance-3yw32x6qfk.pdf

Scheduling heterogeneous multi-cores through Performance Impact Estimation (PIE)

Chip Multiprocessors (CMPs) allow different applications to concurrently execute on a single chip. When applications with differing demands for memory compete for a shared cache, the conventional LRU replacement policy can significantly degrade cache performance when the aggregate working set size is greater than the shared cache. In such cases, shared cache performance can be significantly improved by preserving the entire working set of applications that can co-exist in the cache and preserving some portion of the working set of the remaining applications. This paper investigates the use of adaptive insertion policies to manage shared caches. We show that directly extending the recently proposed dynamic insertion policy (DIP) is inadequate for shared caches since DIP is unaware of the characteristics of individual applications. We propose Thread-Aware Dynamic Insertion Policy (TADIP) that can take into account the memory requirements of each of the concurrently executing applications. Our evaluation with multi-programmed workloads for 2-core, 4-core, 8-core, and 16-core CMPs show that a TADIP-managed shared cache improves overall throughput by as much as 94%, 64%, 26%, and 16% respectively (on average 14%, 18%, 15%, and 17%) over the baseline LRU policy. The performance benefit of TADIP is 2.6x compared to DIP and 1.3x compared to the recently proposed Utility-based Cache Partitioning (UCP) scheme. We also show that a TADIP-managed shared cache provides performance benefits similar to doubling the size of an LRU-managed cache. Furthermore, TADIP requires a total storage overhead of less than two bytes per core, does not require changes to the existing cache structure, and performs similar to LRU for LRU friendly workloads.

Aamer Jaleel

Papers

Adaptive insertion policies for high performance caching

High performance cache replacement using re-reference interval prediction (RRIP)

DRAMsim: a memory system simulator

Scheduling heterogeneous multi-cores through Performance Impact Estimation (PIE)

Adaptive insertion policies for managing shared caches