A key question for the microprocessor research and design community is whether scaling multicores will provide the performance and value needed to scale down many more technology generations. To provide a quantitative answer to this question, a comprehensive study that projects the speedup potential of future multicores and examines the underutilization of integration capacity-dark silicon-is timely and crucial.

/pdf/dark-silicon-and-the-end-of-multicore-scaling-fxij8qeoty.pdf

Dark Silicon and the End of Multicore Scaling

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 －

Since 2005, processor designers have increased core counts to exploit Moore's Law scaling, rather than focusing on single-core performance. The failure of Dennard scaling, to which the shift to multicore parts is partially a response, may soon limit multicore scaling just as single-core scaling has been curtailed. This paper models multicore scaling limits by combining device scaling, single-core scaling, and multicore scaling to measure the speedup potential for a set of parallel workloads for the next five technology generations. For device scaling, we use both the ITRS projections and a set of more conservative device scaling parameters. To model single-core scaling, we combine measurements from over 150 processors to derive Pareto-optimal frontiers for area/performance and power/performance. Finally, to model multicore scaling, we build a detailed performance model of upper-bound performance and lower-bound core power. The multicore designs we study include single-threaded CPU-like and massively threaded GPU-like multicore chip organizations with symmetric, asymmetric, dynamic, and composed topologies. The study shows that regardless of chip organization and topology, multicore scaling is power limited to a degree not widely appreciated by the computing community. Even at 22 nm (just one year from now), 21% of a fixed-size chip must be powered off, and at 8 nm, this number grows to more than 50%. Through 2024, only 7.9x average speedup is possible across commonly used parallel workloads, leaving a nearly 24-fold gap from a target of doubled performance per generation.

/pdf/dark-silicon-and-the-end-of-multicore-scaling-goghgnng6o.pdf

Dark silicon and the end of multicore scaling

Processing-in-memory (PIM) is a promising solution to address the "memory wall" challenges for future computer systems. Prior proposed PIM architectures put additional computation logic in or near memory. The emerging metal-oxide resistive random access memory (ReRAM) has showed its potential to be used for main memory. Moreover, with its crossbar array structure, ReRAM can perform matrix-vector multiplication efficiently, and has been widely studied to accelerate neural network (NN) applications. In this work, we propose a novel PIM architecture, called PRIME, to accelerate NN applications in ReRAM based main memory. In PRIME, a portion of ReRAM crossbar arrays can be configured as accelerators for NN applications or as normal memory for a larger memory space. We provide microarchitecture and circuit designs to enable the morphable functions with an insignificant area overhead. We also design a software/hardware interface for software developers to implement various NNs on PRIME. Benefiting from both the PIM architecture and the efficiency of using ReRAM for NN computation, PRIME distinguishes itself from prior work on NN acceleration, with significant performance improvement and energy saving. Our experimental results show that, compared with a state-of-the-art neural processing unit design, PRIME improves the performance by ~2360× and the energy consumption by ~895×, across the evaluated machine learning benchmarks.

PRIME: a novel processing-in-memory architecture for neural network computation in ReRAM-based main memory

This paper investigates the problem of partitioning a shared cache between multiple concurrently executing applications. The commonly used LRU policy implicitly partitions a shared cache on a demand basis, giving more cache resources to the application that has a high demand and fewer cache resources to the application that has a low demand. However, a higher demand for cache resources does not always correlate with a higher performance from additional cache resources. It is beneficial for performance to invest cache resources in the application that benefits more from the cache resources rather than in the application that has more demand for the cache resources. This paper proposes utility-based cache partitioning (UCP), a low-overhead, runtime mechanism that partitions a shared cache between multiple applications depending on the reduction in cache misses that each application is likely to obtain for a given amount of cache resources. The proposed mechanism monitors each application at runtime using a novel, cost-effective, hardware circuit that requires less than 2kB of storage. The information collected by the monitoring circuits is used by a partitioning algorithm to decide the amount of cache resources allocated to each application. Our evaluation, with 20 multiprogrammed workloads, shows that UCP improves performance of a dual-core system by up to 23% and on average 11% over LRU-based cache partitioning.

/pdf/utility-based-cache-partitioning-a-low-overhead-high-4t50s5xud2.pdf

Utility-Based Cache Partitioning: A Low-Overhead, High-Performance, Runtime Mechanism to Partition Shared Caches

Contention for critical sections can reduce performance and scalability by causing thread serialization. The proposed accelerated critical sections mechanism reduces this limitation. ACS executes critical sections on the high-performance core of an asymmetric chip multiprocessor (ACMP), which can execute them faster than the smaller cores can.

/pdf/accelerating-critical-section-execution-with-asymmetric-3tk7wxunna.pdf

Accelerating Critical Section Execution with Asymmetric Multicore Architectures

Dynamic Random Access Memory (DRAM) cells rely on periodic refresh operations to maintain data integrity. As the capacity of DRAM memories has increased, so has the amount of time consumed in doing refresh. Refresh operations contend with read operations, which increases read latency and reduces system performance. We show that eliminating latency penalty due to refresh can improve average performance by 7.2p. However, simply doing intelligent scheduling of refresh operations is ineffective at obtaining significant performance improvement.This article provides an alternative and scalable option to reduce the latency penalty due to refresh. It exploits the property that each refresh operation in a typical DRAM device internally refreshes multiple DRAM rows in JEDEC-based distributed refresh mode. Therefore, a refresh operation has well-defined points at which it can potentially be Paused to service a pending read request. Leveraging this property, we propose Refresh Pausing, a solution that is highly effective at alleviating the contention from refresh operations. It provides an average performance improvement of 5.1p for 8Gb devices and becomes even more effective for future high-density technologies. We also show that Refresh Pausing significantly outperforms the recently proposed Elastic Refresh scheme.

https://dl.acm.org/doi/pdf/10.1145/2579669

Refresh pausing in DRAM memory systems

A memory storage structure includes a memory storage device, and a first meta-structure having a first size and operating at a first speed. The first speed is faster than a second speed for storing meta-information based on information stored in a memory. A second meta-structure is hierarchically associated with the first meta-structure. The second meta-structure has a second size larger than the first size and operates at the second speed such that faster and more accurate prefetching is provided by coaction of the first and second meta-structures. A method is provided to assemble the meta-information in the first meta-structure and copy this information to the second meta-structure, and prefetching the stored information from the second meta-structure to the first meta-structure ahead of its use.

Context look ahead storage structures

A memory system and method includes a cache having a filtered portion and an unfiltered portion. The unfiltered portion is divided into block sized components, and the filtered portion is divided into sub-block sized components. Blocks evicted from the unfiltered portion have selected sub-blocks thereof cached in the filtered portion for servicing requests.

Apparatus and method for filtering unused sub-blocks in cache memories

Adaptive write leveling in limited lifetime memory devices including performing a method for monitoring a write data stream that includes write line addresses. A property of the write data stream is detected and a write leveling process is adapted in response to the detected property. The write leveling process is applied to the write data stream to generate physical addresses from the write line addresses.

Moinuddin K. Qureshi

Papers

Accelerating Critical Section Execution with Asymmetric Multicore Architectures

Refresh pausing in DRAM memory systems

Context look ahead storage structures

Apparatus and method for filtering unused sub-blocks in cache memories

Adaptive Wear Leveling via Monitoring the Properties of Memory Reference Stream