Dimitra Papagiannopoulou

NBTI-aware data allocation strategies for scratchpad memory based embedded systems

While performance and power continue to be important metrics for embedded systems, as CMOS technologies continue to shrink, new metrics such as variability and reliability have emerged as limiting factors in the design of modern embedded systems. In particular, the reliability impact of pMOS negative bias temperature instability (NBTI) has become a serious concern. Recent works have shown how conventional leakage optimization techniques can help mitigate NBTI-induced aging effects on cache memories. In this paper we focus specifically on scratchpad memory (SPM) and present novel software approaches as a means of alleviating the NBTI-induced aging effects. In particular, we demonstrate how intelligent software directed data allocation strategies can extend the lifetime of partitioned SPMs by means of distributing the idleness across the memory sub-banks.

Speculative synchronization for coherence-free embedded NUMA architectures

High-end embedded systems, like their general-purpose counterparts, are turning to many-core cluster-based shared-memory architectures that provide a shared memory abstraction subject to non-uniform memory access (NUMA) costs. In order to keep the cores and memory hierarchy simple, many-core embedded systems tend to employ simple, scratchpad-like memories, rather than hardware managed caches that require some form of cache coherence management. These “coherence-free” systems still require some means to synchronize memory accesses and guarantee memory consistency. Conventional lock-based approaches may be employed to accomplish the synchronization, but may lead to both useability and performance issues. Instead, speculative synchronization, such as hardware transactional memory, may be a more attractive approach. However, hardware speculative techniques traditionally rely on the underlying cache-coherence protocol to synchronize memory accesses among the cores. The lack of a cache-coherence protocol adds new challenges in the design of hardware speculative support. In this paper, we present a new scheme for hardware transactional memory support within a cluster-based NUMA system that lacks an underlying cache-coherence protocol. To the best of our knowledge, this is the first design for speculative synchronization for this type of architecture. Through a set of benchmark experiments using our simulation platform, we show that our design can achieve significant performance improvements over traditional lock-based schemes.

Flexible data allocation for scratch-pad memories to reduce NBTI effects

Negative Bias Temperature Instability (NBTI) is a major reliability issue in nanoscale VLSI systems. Previous work has shown how the exploitation of conventional optimization techniques can reduce the NBTI-induced aging in cache memories. Other works have proposed approaches that incorporate software directed data allocation strategies to partially recover from NBTI-induced aging in Scratchpad Memories (SPM). In this paper, we extend the existing software approach in order to enhance the memory allocation flexibility and make it more appropriate for real embedded applications. Simulation results demonstrate how our proposed data allocation strategies can help mitigate the NBTI-induced aging effects, as well as reduce the leakage energy consumption on scratch-pad memories.

NBTI-Aware Data Allocation Strategies for Scratchpad Based Embedded Systems

The push to embed reliable and low-power memories architectures into modern systems-on-chip is driving the EDA community to develop new design techniques and circuit solutions that can concurrently optimize aging effects due to Negative Bias Temperature Instability (NBTI), and static power consumption due to leakage mechanisms. While recent works have shown how conventional leakage optimization techniques can help mitigate NBTI-induced aging effects on cache memories, in this paper we focus specifically on scratchpad memory (SPM) and present novel software approaches as a means of alleviating the NBTI-induced aging effects. In particular, we demonstrate how intelligent software directed data allocation strategies can extend the lifetime of partitioned SPMs by means of distributing the idleness across the memory sub-banks.

Energy-Efficient and High-Performance Lock Speculation Hardware for Embedded Multicore Systems

Embedded systems are becoming increasingly common in everyday life and like their general-purpose counterparts, they have shifted towards shared memory multicore architectures. However, they are much more resource constrained, and as they often run on batteries, energy efficiency becomes critically important. In such systems, achieving high concurrency is a key demand for delivering satisfactory performance at low energy cost. In order to achieve this high concurrency, consistency across the shared memory hierarchy must be accomplished in a cost-effective manner in terms of performance, energy, and implementation complexity. In this article, we propose Embedded-Spec, a hardware solution for supporting transparent lock speculation, without the requirement for special supporting instructions. Using this approach, we evaluate the energy consumption and performance of a suite of benchmarks, exploring a range of contention management and retry policies. We conclude that for resource-constrained platforms, lock speculation can provide real benefits in terms of improved concurrency and energy efficiency, as long as the underlying hardware support is carefully configured.

Thrifty-malloc: A HW/SW codesign for the dynamic management of hardware transactional memory in embedded multicore systems

We present thrifty-malloc: a transaction-friendly dynamic memory manager for high-end embedded multicore systems. The manager combines modularity, ease-of-use and hardware transactional memory (HTM) compatibility in a lightweight and memory-effcient design. Thrifty-malloc is easy to deploy and configure for non-expert programmers, yet provides good performance with low memory overhead for highly-parallel embedded applications running on massively parallel processor arrays (MPPAs) or many-core architectures. In addition, the transparent mechanisms that increase our managers resilience to unpredictable dynamic situations incur a low timing overhead in comparison to established techniques.

Transparent and energy-efficient speculation on NUMA architectures for embedded MPSoCs

High-end embedded systems such as smart phones, game consoles, GPS-enabled automotive systems, and home entertainment centers, are becoming ubiquitous. Like their general-purpose counterparts, and for many of the same energy-related reasons, embedded systems are turning to multicore architectures. Moreover, as the demand for more compute-intensive capabilities for embedded systems increases, these multicore architectures will evolve into many-core systems for improved performance or performance/area/Watt. These systems are often organized as cluster based Non-Uniform Memory Access (NUMA) architectures that provide the programmer with a shared-memory abstraction, with the cost of sharing memory (in terms of performance, energy, and complexity) varying substantially depending on the locations of the communicating processes. This paper investigates one of the principal challenges presented by these emerging NUMA architectures for embedded systems: providing efficient, energy-effective and convenient mechanisms for synchronization and communication. In this paper, we propose an initial solution based on hardware support for speculative synchronization.

Hardware Transactional Memory Exploration in Coherence-Free Many-Core Architectures

High-end embedded systems, like their general-purpose counterparts, are turning to many-core cluster-based shared-memory architectures that provide a shared memory abstraction subject to non-uniform memory access costs. In order to keep the cores and memory hierarchy simple, many-core embedded systems tend to employ simple, scratchpad-like memories, rather than hardware managed caches that require some form of cache coherence management. These “coherence-free” systems still require some means to synchronize memory accesses and guarantee memory consistency. Conventional lock-based approaches may be employed to accomplish the synchronization, but may lead to both usability and performance issues. Instead, speculative synchronization, such as hardware transactional memory, may be a more attractive approach. However, hardware speculative techniques traditionally rely on the underlying cache-coherence protocol to synchronize memory accesses among the cores. The lack of a cache-coherence protocol adds new challenges in the design of hardware speculative support. In this article, we present a new scheme for hardware transactional memory (HTM) support within a cluster-based, many-core embedded system that lacks an underlying cache-coherence protocol. We propose two alternative data versioning implementations for the HTM support, Full-Mirroring and Distributed Logging and we conduct a performance comparison between them. To the best of our knowledge, these are the first designs for speculative synchronization for this type of architecture. Through a set of benchmark experiments using our simulation platform, we show that our designs can achieve significant performance improvements over traditional lock-based schemes.

Evaluating critical bits in arithmetic operations due to timing violations

Various error models are being used in simulation of voltage-scaled arithmetic units to examine application-level tolerance of timing violations. The selection of an error model needs further consideration, as differences in error models drastically affect the performance of the application. Specifically, floating point arithmetic units (FPUs) have architectural characteristics that characterize its behavior. We examine the architecture of FPUs and design a new error model, which we call Critical Bit. We run selected benchmark applications with Critical Bit and other widely used error injection models to demonstrate the differences.