International audience

Nowadays, companies collect massive quantities of data to enhance their operations, often at the expense of sharing user sensible information. This data is widely used to train Deep Learning (DL) neural networks to model, classify, or recognize complex data. These activities enable companies to offer an array of services to users, such as precise advertising and optimal location services. This study explores potential solutions for preserving privacy while utilizing DL applications. To address the privacy issue, we develop a privacy-preserving framework specifically designed for fog computing environments. Unlike traditional cloud computing architectures, fog embedded architectures only share a small portion of user data with a nearby fog node, ensuring that the majority of sensitive data remains secure. Within these fog nodes, we incorporate two additional algorithms, namely Generalization and Threshold, to enhance the privacy-preserving capabilities of the framework. The first algorithm, Generalization, introduces a validation dataset within the fog nodes which not only increases the accuracy of the fog-embedded framework but also ensures that user data is preserved. The second algorithm, Threshold, is responsible for protecting user data samples and reducing the amount of information sent to the server. By combining these two algorithms, we are able to provide an additional layer of protection for user privacy while still maintaining the accuracy of the model. We conduct an evaluation to test its effectiveness using two separate datasets. In addition, we analyze them through a Feed Forward Neural Network (FFNN) and compare the results with a traditional centralized architecture to validate the effectiveness of the proposed framework. The results of our evaluation demonstrate that the proposed privacy-preserving framework, when combined with the Generalization and Threshold algorithms, can preserve up to 38.44% of user data. Additionally, we were able to extend the framework to multiple fog nodes without compromising the network’s accuracy, as we only observed a 0.1% decrease in accuracy when using the proposed architecture. This study emphasizes the importance of preserving user information while using DL applications and provides a solution that trains the desired network without violating user privacy, hence preserving their anonymity. Overall, the study highlights the potential of Federated Deep Learning to improve the accuracy and privacy of DL applications in fog computing environments.

VITAMIN-V is a 2023–2025 Horizon Europe project that aims to develop a complete RISC-V open-source software stack for cloud services with comparable performance to the cloud-dominant x86 counterpart and a powerful virtual execution environment for software development, validation, verification, and testing that considers the relevant RISC-VISA extensions for cloud deployment. VITAMIN-V will specifically support the RISC-V extensions for virtualization, cryptography, and vec-torization in three virtual environments: QEMU, gem5, and cloud FPGA prototype platforms. The project will focus on European Processor Initiative (EPI) based RISC-V designs and accelerators. VITAMIN-V will also support the ISA extensions by adding the compiler and toolchain support. Furthermore, it will develop novel software validation, verification, and testing approaches to ensure software trustworthiness. To enable the execution of complete cloud stacks, VITAMIN-V will port all necessary machine-dependent modules in relevant open-source cloud software distributions, focusing on three cloud setups. Finally, VITAMIN-V will demonstrate and benchmark these three cloud setups using relevant AI, big-data, and serverless applications. VITAMIN-V aims to match the software performance of its x86 equivalent while contributing to RISC-V open-source virtual environments, software validation, and cloud software suites.

Vitamin-V is a project funded under the Horizon Europe program for the period 2023-2025. The project aims to create a complete open-source software stack for RISC-V that can be used for cloud services. This software stack is intended to have the same level of performance as the x86 architecture, which is currently dominant in the cloud computing industry. In addition, the project aims to create a powerful virtual execution environment that can be used for software development, validation, verification, and testing. The virtual environment will consider the relevant RISC-V ISA extensions required for cloud deployment. Commercial cloud systems use hardware features currently unavailable in RISC-V virtual environments, including virtualization, cryptography, and vectorization. To address this, Vitamin-V will support these features in three virtual environments: QEMU, gem5, and cloud-FPGA prototype platforms. The project will focus on providing support for EPI-based RISC-V designs for both the main CPUs and cloud-important accelerators, such as memory compression. The project will add the compiler (LLVM-based) and toolchain support for the ISA extensions. Moreover, Vitamin-V will develop novel approaches for validating, verifying, and testing software trustworthiness. This paper focuses on the plans and visions that the Vitamin-V project has to support validation, verification, and testing for cloud applications, particularly emphasizing the hardware support that will be provided.

This special session paper introduces the Horizon Europe NEUROPULS project, which targets the development of secure and energy-efficient RISC-V interfaced neuromorphic accelerators using augmented silicon photonics technology. Our approach aims to develop an augmented silicon photonics platform, an FPGA-powered RISC-V-connected computing platform, and a complete simulation platform to demonstrate the neuromorphic accelerator capabilities. In particular, their main advantages and limitations will be addressed concerning the underpinning technology for each platform. Then, we will discuss three targeted use cases for edge-computing applications: Global National Satellite System (GNSS) anti-jamming, autonomous driving, and anomaly detection in edge devices. Finally, we will address the reliability and security aspects of the stand-alone accelerator implementation and the project use cases.

Microcontrollers running safety-critical applications with high integrity requirements must provide appropriate safety measures to manage random hardware faults. For instance, automotive safety regulations (e.g., ISO26262) impose the use of diverse redundancy for items at the highest automotive safety integrity level (ASIL), ASIL-D. In the case of computing cores, this is realized with dual core lockstep (DCLS). The advent of the RISC-VISA has made open source hardware gain popularity. However, there are few industrial open source SoCs meeting the requirements of safety-critical systems, and, to our knowledge, none of them provides lockstep cores. This paper presents the realization of a RISC-V open source lockstep core based on Gaisler's NOEL-V core for the space domain, as well as its integration in the SELENE SoC that provides a complete microcontroller synthesizable on FPGA successfully assessed against space, automotive and railway safety-critical applications in the past.

RISC-V Instruction Set Architecture (ISA) emerges as an opportunity to develop open source hardware without being subject to expensive licenses or export restrictions. A plethora of initiatives are nowadays developing systems-on-chip (SoCs) and its components based on RISC-V targeting a wide variety of markets. However, domains with safety requirements, such as avionics, space, and automotive, impose SoCs to include support to meet those requirements.This work introduces the SafeX family of components, a set of components providing SoC controllability, observability and safety measures support. These components, developed by the Barcelona Supercomputing Center with permissive open source licenses, are intended to be the basis to make SoCs meet the needs of domains with safety requirements. In particular, the SafeX components developed so far include the SafeSU (multicore statistics unit), the SafeTI (flexible and programmable traffic injector), the SafeDE and SafeSoftDR (hardware and software modules to enforce lockstep execution), and the SafeDM (module to monitor diversity across cores).

This work proposes an SRAM array with built-in real-time error detection (RTD) capabilities. Each cell in the new RTD-SRAM array computes its part of the real-time parity of an SRAM array column on-the-fly. RTD based arrays detect a fault right away after it occurs, rather than when it is read. RTD, therefore, breaks the serialization between data access and error detection and, thus, it can speed-up the access-time of arrays that use on-the-fly error detection and correction. The paper presents an analysis and optimization of an RTD-SRAM and its application to a tag array. Compared to a state-of-the-art tag array protection, the evaluated scheme has comparable error detection and correction strength and, depending on the array dimensions, the access time is reduced by 5% to 18%, energy by 20% to 40% and area up to 30%.

Soft Error Rate (SER) estimation is an important challenge for integrated circuits because of the increased vulnerability brought by technology scaling. This paper presents a methodology to estimate in early stages of the design the susceptibility of combinational circuits to particle strikes. In the core of the framework liesMASkIt, a novel approach that combines signal probabilities with technology characterization to swiftly compute the logical, electrical, and timing masking effects of the circuit under study taking into account all input combinations and pulse widths at once. Signal probabilities are estimated applying a new hybrid approach that integrates heuristics along with selective simulation of reconvergent subnetworks. The experimental results validate our proposed technique, showing a speedup of two orders of magnitude in comparison with traditional fault injection estimation with an average estimation error of 5 percent. Finally, we analyze thevulnerabilityof the Decoder, Scheduler, ALU, and FPU of an out-of-order, superscalar processor design.

Hardware acceleration circuits for cryptographic algorithms are largely deployed in a wide range of products. The HW implementations of such algorithms often suffer from a number of vulnerabilities that expose systems to several attacks, e.g., differential fault analysis (DFA). The challenge for designers is to protect cryptographic accelerators in a cost-effective and power-efficient way. In this paper, we propose a lightweight technique for protecting hardware accelerators implementing AES and SHA-2 (which are two widely used NIST standards) against DFA. The proposed technique exploits partial redundancy to first detect the occurrence of a fault and then to react to the attack by obfuscating the output values. An experimental campaign demonstrated that the overhead introduced is 8.32% for AES and 3.88% for SHA-2 in terms of area, 0.81% for AES and 12.31% for SHA-2 in terms of power with no working frequency reduction. Moreover, a comparative analysis showed that our proposal outperforms the most recent related countermeasures.

RECIPE (REliable power and time-ConstraInts-aware Predictive management of heterogeneous Exascale systems) is a recently started project funded within the H2020 FETHPC programme, which is expressly targeted at exploring new High-Performance Computing (HPC) technologies. RECIPE aims at introducing a hierarchical runtime resource management infrastructure to optimize energy efficiency and minimize the occurrence of thermal hotspots, while enforcing the time constraints imposed by the applications and ensuring reliability for both time-critical and throughput-oriented computation that run on deeply heterogeneous accelerator-based systems. This paper presents a detailed overview of RECIPE, identifying the fundamental challenges as well as the key innovations addressed by the project, which span run-time management, heterogeneous computing architectures, HPC memory/interconnection infrastructures, thermal modelling, reliability, programming models, and timing analysis. For each of these areas, the paper describes the relevant state of the art as well as the specific actions that the project will take to effectively address the identified technological challenges.

Cross-layer reliability is becoming the preferred solution when reliability is a concern in the design of a microprocessor-based system. Nevertheless, deciding how to distribute the error management across the different layers of the system is a very complex task that requires the support of dedicated frameworks for cross-layer reliability analysis. This paper proposes SyRA, a system-level cross-layer early reliability analysis framework for radiation induced soft errors in memory arrays of microprocessor-based systems. The framework exploits a multi-level hybrid Bayesian model to describe the target system and takes advantage of Bayesian inference to estimate different reliability metrics. SyRA implements several mechanisms and features to deal with the complexity of realistic models and implements a complete tool-chain that scales efficiently with the complexity of the system. The simulation time is significantly lower than micro-architecture level or RTL fault-injection experiments with an accuracy high enough to take effective design decisions. To demonstrate the capability of SyRA, we analyzed the reliability of a set of microprocessor-based systems characterized by different microprocessor architectures (i.e., Intel x86, ARM Cortex-A15, ARM Cortex-A9) running both the Linux operating system or bare metal in the presence of single bit upsets caused by radiation induced soft errors. Each system under analysis executes different software workloads both from benchmark suites and from real applications.

With the advent of the slowdown in DRAM capacitor scaling [1] and the increased reliability problems of traditional 6T SRAM memories [2], industry and academia have looked for alternative memory cells. Among those, gain-cells have attracted significant attention due to their smaller size (compared to SRAM) and non-destructive read operation (compared to DRAM) as well as considerable low power and reasonable robustness. This paper first summarizes the available evidences of SRAM and eDRAM in commercial and test chips. Then, it analyzes the performance, reliability and scaling of eDRAM gain-cells in 10 and 7 nm FinFET technology; as well as above and below VT(i.e. sub-threshold).

Early reliability assessment of hardware structures using microarchitecture level simulators can effectively guide major error protection decisions in microprocessor design. Statistical fault injection on microarchitectural structures modeled in performance simulators is an accurate method to measure their Architectural Vulnerability Factor (AVF) but requires excessively long campaigns to obtain high statistical significance.

Integrated circuits are getting increasingly vulnerable to soft errors; as a consequence, soft error rate (SER) estimation has become an important and very challenging goal. In this work, a novel approach for SER estimation of combinational circuits is presented. The proposed framework is divided in two stages. First, signal probabilities are computed via a hybrid approach combining heuristics and selective simulation of reconvergent subnets. In the second stage, signal probabilities are used to compute the vulnerability of all the gates in a combinational block using a backward-traversing algorithm that takes into account logical, electrical and timing masking factors. Experimental results show that our signal probability estimation approach, in comparison with similar techniques in the literature, reduces inaccuracy by 96% while adding minimal execution time overhead. In addition, results indicate that our framework is two orders of magnitude faster than traditional Monte Carlo-based fault injection with minor loss in accuracy in both signal probability and SER estimation (average error of 5%).

Estimating the failure probabilities of SRAM memory cells using Monte Carlo or Importance Sampling techniques is expensive in the number of SPICE simulations needed. This paper presents a methodology for estimating the dynamic margin failure probabilities by building a surrogate model of the dynamic margin using Gaussian Process regression. Additive kernel functions that can extrapolate the margin values from the simulated samples are presented. These proposed kernel functions decrease the out-of-sample error of the surrogate model for a 6T cell by 32% compared with a six-dimensional universal kernel such as a Radial-Basis-Function kernel (RBF). Finally, the failure probability values predicted by a surrogate model built using 1250 SPICE simulations are reported and compared with Monte Carlo analysis with 106samples. The results show a relative error of 30% at 0.4V (predicted value of 4×10-6for the Monte Carlo estimate of 3×10-6) and a relative error of 172% at 0.3V (predicted value of 3×10-5for the Monte Carlo estimate of 1.1×10-5) for the dynamic read margin. These accuracy numbers are similar to those reported in previous proposals while the reduction in SPICE simulations is between 4× and 23× relative to these proposals and 800× compared to Monte Carlo method.

Bio-medical wearable devices restricted to their small-capacity embedded-battery require energy-efficiency of the highest order. However, minimum-energy point (MEP) at sub-threshold voltages is unattainable with SRAM memory, which fails to hold below 0.3V because of its vanishing noise margins. This paper examines minimum-energy operation of 2T and 3T1D e-DRAM gain cells as an alternative to SRAM at 32nm technology node with different design points: up-sizing transistors, using high-Vth transistors, read/write wordline assists and temperature. First, the e-DRAM cells are evaluated without considering any process variations. The design-space is explored by creating a kriging meta-model to reduce the number of simulations. Finally, a full-factorial statistical analysis of e-DRAM cells is performed in presence of threshold voltage variations. The effect on mean MEP is also reported.

System reliability estimation during early design phases facilitates informed decisions for the integration of effective protection mechanisms against different classes of hardware faults. When not all system abstraction layers (technology, circuit, microarchitecture, software) are factored in such an estimation model, the delivered reliability reports must be excessively pessimistic and thus lead to unacceptably expensive, over-designed systems. We propose a scalable, cross-layer methodology and supporting suite of tools for accurate but fast estimations of computing systems reliability. The backbone of the methodology is a component-based Bayesian model, which effectively calculates system reliability based on the masking probabilities of individual hardware and software components considering their complex interactions. Our detailed experimental evaluation for different technologies, microarchitectures, and benchmarks demonstrates that the proposed model delivers very accurate reliability estimations (FIT rates) compared to statistically significant but slow fault injection campaigns at the microarchitecture level.

In this paper, we analyze the suitability of implementing embedded DRAM (eDRAM) cells on FinFET technology compared to classical planar MOSFETs. The results show a significant improvement in overall cell performance for multi-gate devices. While pFinFET-based memories showed better cell behavior and variability robustness, mixed n/pFinFET cells had the highest working frequency and a negligible impact on degradation. Finally, we show that a multiple fin-height strategy can be used to reduce the layout area of the eDRAM cells (>10%).

With the growing importance of parametric (process and environmental) variations in advanced technologies, it has become a serious challenge to design reliable, fast and low-power embedded memories. Adopting a variation-aware design paradigm requires a holistic perspective of memory-wide metrics such as yield, power and performance. However, accurate estimation of such metrics is largely dependent on circuit implementation styles, technology parameters and architecture-level specifics. In this paper, we propose a fully automated tool - INFORMER - that helps high-level designers estimate memory reliability metrics rapidly and accurately. The tool relies on accurate circuit-level simulations of failure mechanisms such as soft-errors and parametric failures. The statistics obtained can then help couple low-level metrics with higher-level design choices. A new technique for rapid estimation of low-probability failure events is also proposed. We present three use-cases of our prototype tool to demonstrate its diverse capabilities in autonomously guiding large SRAM based robust memory designs.

Recent technology trends has turned DRAMs into an interesting candidate to substitute traditional SRAM-based on-chip memory structures (i.e. register file, cache memories). Nevertheless, a major problem to introduce these cells is that they lose their state (i.e. value) over time, and they have to be refreshed. This paper proposes the implementation of coherent caches with DRAM cells. Furthermore, we propose to use the coherence state to tune the refresh overhead. According to our analysis, an average of up to 57% of refresh energy can be saved. Also, comparing to the caches implemented in SRAMs total energy savings are on average up to 39% depending of the refresh policy with a performance loss below 8%.

Estimating extremely low SRAM failure-probabilities by conventional Monte Carlo (MC) approach requires hundreds-of-thousands simulations making it an impractical approach. To alleviate this problem, failure-probability estimation methods with a smaller number of simulations have recently been proposed, most notably variants of consecutive mean-shift based Importance Sampling (IS). In this method, a large amount of time is spent simulating data points that will eventually be discarded in favor of other data-points with minimum norm. This can potentially increase the simulation time by orders of magnitude. To solve this very important limitation, in this paper, we introduce SSFB: a novel SRAM failure-probability estimation method that has much better cognizance of the data points compared to conventional approaches. The proposed method starts with radial simulation of a single point and reduces discarded simulations by: a) random sampling-only-when it reaches a failure boundary and after that continues again with radial simulation of a chosen point, and b) random sampling is performed-only-within a specific failure-range which decreases in each iteration. The proposed method is also scalable to higher dimensions (more input variables) as sampling is done on the surface of the hyper-sphere, rather than within-the-hypersphere as other techniques do. Our results show that using our method we can achieve an overall 40x reduction in simulations compared to consecutive mean-shift IS methods while remaining within the 0.01-Sigma accuracy.

Negative bias temperature instability (NBTI) is a major cause of concern for chip designers because of its inherent ability to drastically reduce silicon reliability over the lifetime of the processor. Coupled with statistical variations of process parameters, it can potentially render systems dysfunctional in certain scenarios. Data caches suffer the most from such phenomenon because of the unbalanced duty cycle ratio of SRAM cells and maximum intrinsic susceptibility to process variations. In this paper, we propose a novel NBTI-aware technique, invert-Read-Modify-Write (iRMW) that can improve the functional yield of the data cache significantly over its lifetime. Using architecture-level benchmarks, we first analyse the impact of activity factor and workload variation on NBTI-induced failures in data caches. iRMW is then used as a means to balance the duty cycle by alternating between recovery and stress cycle upon successive read accesses to the cache line. The highly transient nature of the data stored in L1 data cache aides this process of recovery upon using iRMW. A unique feature of iRMW is its intelligent use of low-leakage & NBTI-tolerant embedded-DRAM cells as an alternative to SRAM-cells for storing important state information. Our experiments conducted using SPEC2006 and PhysicsBench workloads show that on-average the cache failure probability can be reduced by 22%, 33% and 36% after two, four and eight years of processor usage respectively. In addition to being extremely power-frugal, use of eDRAM reduces total area footprint of iRMW tremendously.

The big challenge that we face today for designing resilient memories is the huge number of simulations needed to arrive at a good estimate of memory's yield. A lot of work has come up recently focusing on the reduction of these simulations. The majority of these methods have focused on using different Markov Chain Monte Carlo (MCMC) methods, most notably Importance Sampling. SRAMs, though, have an interesting property of failure monotonicity which implies that given a known failure point in SRAM's parameter space all points with larger variations will also be failure points. Our work REEM (Region Estimation by Exploiting Monotonicity), thus, focuses on exploiting the SRAM's failure monotonicity property for faster estimation of the Failure/Non-Failure regions. The usual MCMC methods can then be used without needing actual spice simulations. Our results show that using our method we can achieve an overall 10× reduction in simulations compared to traditional Importance Sampling methods.

Evaluation of computing systems reliability must be accurate enough to provide hints for the required fault protection mechanisms that will guarantee correctness of operation at acceptance costs. To be useful, reliability evaluation must be performed early enough in the design cycle when, however, the available details of the system are largely unknown. This inherent contradiction in terms: early vs. accurate, requires a cross-layer approach for reliability evaluation. Different layers of abstraction contribute differently in the overall system reliability; if this contribution can be assessed independently, the reliability of the system can be evaluated at the early stages of the design. We review the state-of-the-art in the area and discuss corresponding challenges .

Low-power modes in modern microprocessors rely on low frequencies and low voltages to reduce the energy budget. Nevertheless, manufacturing induced parameter variations can make SRAM cells unreliable producing hard errors at supply voltages below Vccmin.

The heavily-threaded data processing demands of streaming multiprocessors (SM) in a GPGPU require a large register file (RF). The fast increasing size of the RF makes the area cost and power consumption unaffordable for traditional SRAM designs in the future technologies. In this paper, we propose to use embedded-DRAM (eDRAM) as an alternative in future GPGPUs. Compared with SRAM, eDRAM provides higher density and lower leakage power. However, the limited data retention time in eDRAM poses new challenges. Periodic refresh operations are needed to maintain data integrity. This is exacerbated with the scaling of eDRAM density, process variations and temperature. Unlike conventional CPUs which make use of multi-ported RF, most of the RFs in modern GPGPU are heavily banked but not multi-ported to reduce the hardware cost. This provides a unique opportunity to hide the refresh overhead. We propose two different eDRAM implementations based on 3T1D and 1T1C memory cells. To mitigate the impact of periodic refresh, we propose two novel refresh solutions using bank bubble and bank walk-through. Plus, for the 1T1C RF, we design an interleaved bank organization together with an intelligent warp scheduling strategy to reduce the impact of the destructive reads. The analysis shows that our schemes present better energy efficiency, scalability and variation tolerance than traditional SRAM-based designs.

The widespread adoption of chip multiprocessors in recent years has increased the number of applications simultaneously accessing DRAM memories. Therefore, memory access patterns have also changed and this has reduced row buffer locality significantly, degrading performance and energy efficiency. Furthermore, concurrent execution of applications also has shown the need of performance isolation among threads in the memory controller to enforce a quality of service in virtualized environments. Existing DRAM memories, however, enforce a tradeoff between throughput and isolation. To solve these problems, this paper proposes the addition of Thread Row Buffers (TRBs) to DRAM memories. TRBs keep an active row per thread, thereby increasing DRAM efficiency by avoiding alternate accesses to a limited number of rows and allowing the implementation of a memory scheduler not bound to the throughput-isolation tradeoff. Thread Row Buffers with Service Partitioning (TRB-SP) increase the row hit-rate by 38 percent with respect to FR-FCFS and by 11 percent with respect to Cache DRAM. This, in turn, increases overall performance by 17 and 7 percent, respectively. TRB-SP is also able to reduce the standard deviation of the memory access time of an application by 40 percent over FR-FCFS, 31 percent over PAR-BS, and 42 percent over Cache DRAM.

In view of device scaling issues, embedded DRAM (eDRAM) technology is being considered as a strong alternative to conventional SRAM for use in on-chip memories. Memory cells designed using eDRAM technology in addition to being logic-compatible, are variation tolerant and immune to noise present at low supply voltages. However, two major causes of concern are the data retention capability which is worsened by parameter variations leading to frequent data refreshes (resulting in large dynamic power overhead) and the transient reduction of stored charge increasing soft-error (SE) susceptibility. In this paper, we present a novel variation-tolerant 4T-DRAM cell whose power consumption is 20.4% lower when compared to a similar sized eDRAM cell. The retention time on-average is improved by 2.04X while incurring a delay overhead of 3% on the read-access time. Most importantly, using a soft-error (SE) rate analysis tool, we have confirmed that the cell sensitivity to SEs is reduced by 56% on-average in a natural working environment.

In this paper, we present the dynamic 3T memory cell for future 10nm tri-gate FinFETs as a potential replacement for classical 6T SRAM cell for implementation in high speed cache memories. We investigate read access time, retention time, and static power consumption of the cell when it is exposed to the effects of process and environmental variations. Process variations are extracted from the ITRS predictions and they are modeled at device level. For simulation, we use 10nm SOI tri-gate FinFET BSIM-CMG model card developed by the University of Glasgow, Device Modeling Group. When compared to the classical 6T SRAM, 3T cell has 40% smaller area, leakage is reduced up to 14 times while access time is approximately the same. In order to achieve higher retention times, we propose several cell extensions which, at the same time, enable post-fabrication/run-time adaptability.

Due to increased leakage currents and variability, classical bulk technology is reaching its scaling limits and some alternatives must be found. FinFETs are one of those alternatives. Through their 3D structure, they achieve better channel control which is the key to scalability. However, some sources of variability still remain. The impact of this technology shift on SRAM and DRAM memories is analyzed in this work.

Current trends in CMPs indicate that the core count will increase in the near future. One of the main performance limiters of these forthcoming microarchitectures is the latency and high demand of the on-chip network and the off-chip memory communication. One of the main trade-offs when searching an optimal cache hierarchy is the sharing degree of cache space and its on-die distribution. Several techniques have appeared recently that optimize these parameters to get a better performance. This work provides some insight in the most promising configurations for tiled microarchitectures and shows the advantages and limitations of each of them in terms of performance and energy efficiency. This paper extends previous works by providing a complete study that evaluates different network topologies, single and multithreaded benchmarks, and single and multiprogrammed execution. In all these studies, the Distributed Cooperative Caching shows to be a promising alternative to traditional configurations for chip multiprocessors, providing a scalable and energy efficient solution.

Memory circuits are playing a key role in complex multicore systems with both data and instructions storage and mailbox communication functions. There is a general concern that conventional SRAM cell based on the 6T structure could exhibit serious limitations in future CMOS technologies due to the instability caused by transistor mismatching as well as for leakage consumption reasons. For L1 data caches the new cell 3T1D DRAM is considered a potential candidate to substitute 6T SRAMs. We first evaluate the impact of the positive bias temperature instability, PBTI, on the access and retention time of the 3T1D memory cell implemented with 45 nm technology. Then, we consider all sources of variations and the effect of the degradation caused by the aging of the device on the yield at system level.

In this paper, we propose a dynamically tunable fine-grain body biasing mechanism to reduce standby leakage power in first level data-caches under process variations. Accessed physical arrays are forward body biased (FBB) to improve latency while idle (unaccessed) arrays are reverse body biased (RBB) for reducing standby leakage power. The bias voltage to be applied is computed at design time and updated at run-time to counter the negative effects of process variations. This ensures that under all scenarios, the cache will consume the lowest leakage power for the target access latency computed at design-time. A sensor-like hardware mechanism measures the variation in latency and leakage at run-time and this measurement is used to update the bias voltage. The backbone of the hardware used for measurement is a three-transistor one-diode(3T1D)DRAM cell embedded into a regular cache array. By measuring the access and retention time of the 3T1D cell, we show that it is possible to classify cache arrays based on run-time latency/leakage profiles. Our technique reduces leakage energy consumption and access latency of the cache on an average by 20% & 18% respectively. Finally we show that our technique will improve parametric yield by a maximum of 38% for worst-case scenario.

The TRAMS (Terascale Reliable Adaptive MEMORY Systems) project addresses in an evolutionary way the ultimate CMOS scaling technologies and paves the way for revolutionary, most promising beyond-CMOS technologies. In this abstract we show the significant variability levels of future 18 and 13nm device bulk-CMOS technologies as well as its dramatic effect on the yield of memory cells, and what kind of circuit solution would be required to maintain the current yield level. Later, we discuss the impact of errors at the system level, and different approaches at system level to adapt the heterogeneous systems to user's requirements.

With every process generation, the problem of variability in physical parameters and environmental conditions poses a great challenge to the design of fast and reliable circuits. Propagation delays which decide circuit performance are likely to suffer the most from this phenomena. While Statistical static timing analysis (SSTA) is used extensively for this purpose, it does not account for dynamic conditions during operation. In this paper, we present a multivariate regression based technique that computes the propagation delay of circuits subject to manufacturing process variations in the presence of temporal variations like temperature. It can be used to predict the dynamic behavior of circuits under changing operating conditions. The median error between the proposed model and circuit-level simulations is below 5%. With this model, we ran a study of the effect of temperature on access time delays for 500 cache samples. The study was run in 0.557 seconds, compared to the 20h and 4min of the SPICE simulation achieving a speedup of over 1×105. As a case study, we show that the access times of caches can vary as much as 2.03× at high temperatures in future technologies under process variations.

Next generation tiled microarchitectures are going to be limited by off-chip misses and by on-chip network usage. Furthermore, these platforms will run an heterogeneous mix of applications with very different memory needs, leading to significant optimization opportunities. Existing adaptive memory hierarchies use either centralized structures that limit the scalability or software based resource allocation that increases programming complexity.

Estimation of static and dynamic energy of caches is critical for high-performance low-power designs. Commercial CAD tools performing energy estimation statically are not aware of the changing operating and environmental conditions which makes the problem of energy estimation more dynamic in nature. It is worsened by process induced variations of low level parameters like threshold voltage and channel length. In this paper we present MODEST, a proposal for estimating the static and dynamic energy of caches taking into account spatial variations of physical parameters, temporal changes of supply voltage and environmental factors like temperature. It can be used to estimate the energy of different blocks of a cache based on a combination empirical data and analytical equations. The observed maximum and median error between MODEST and HSPICE energy-estimates for 22,500 samples is around 7.8% and 0.5% respectively. As a case study, using MODEST, we propose a two step iterative optimization procedure involving Dual-V th assignment and standby supply voltage minimization for reclaiming energy-constrained caches. The observed energy reduction is around 50.8% for the most-leaky Cache. A speed-up of 750X over conventional hard-coded implementation for such optimizations is achieved.

This paper evaluates several techniques to save leakage in CMP L2 caches by selectively switching off the less used lines. We primarily focus on private snoopy L2 caches. In this case, coherence must be enforced in all situations and specially when a line is turned off to save power. In particular, we introduce three techniques: the first one turns off the cache lines by using the coherence protocol invalidations, the second one is an implementation of a cache decay technique specific for coherent caches, the third one is a performance-optimized decay-based technique for coherent caches. Experimental results, carried out by using accurate performance/thermal/energy models, show that appreciable power savings can be achieved by properly designing a leakage optimization technique. We target a CMP composed of 4 cores and 1 to 8 MB of total cache. For 4 MB, the proposed techniques show a 13%, 30%, and 21% energy reduction, respectively, at the cost of 0%, 8%, and 2% performance loss. For other cache sizes the behavior is qualitatively similar.

SRAM and DRAM cells have been the predominant technologies used to implement memory cells in computer systems, each one having its advantages and shortcomings. SRAM cells are faster and require no refresh since reads are not destructive. In contrast, DRAM cells provide higher density and minimal leakage energy since there are no paths within the cell from Vdd to ground. Recently, DRAM cells have been embedded in logic-based technology, thus overcoming the speed limit of typical DRAM cells.

This paper presents the Distributed Cooperative Caching, a scalable and energy-efficient scheme to manage chip multiprocessor (CMP) cache resources. The proposed configuration is based in the Cooperative Caching framework [3] but it is intended for large scale CMPs. Both centralized and distributed configurations have the advantage of combining the benefits of private and shared caches. In our proposal, the Coherence Engine has been redesigned to allow its partitioning and thus, eliminate the size constraints imposed by the duplication of all tags. At the same time, a global replacement mechanism has been added to improve the usage of cache space. Our framework uses several Distributed Coherence Engines spread across all the nodes to improve scalability. The distribution permits a better balance of the network traffic over the entire chip avoiding bottlenecks and increasing performance for a 32-core CMP by 21% over a traditional shared memory configuration and by 57% over the Cooperative Caching scheme.

Multicore architectures have been ruling the recent microprocessor design trend. This is due to different reasons: better performance, thread-level parallelism bounds in modern applications, ILP diminishing returns, better thermal/power scaling (many small cores dissipate less than a large and complex one), and the ease and reuse of design. This paper presents a thorough evaluation of multicore architectures. The architecture that we target is composed of a configurable number of cores, a memory hierarchy consisting of private L1, shared/private L2, and a shared bus interconnect. We consider a benchmark set composed of several parallel shared memory applications. We explore the design space related to the number of cores, L2 cache size, and processor complexity, showing the behavior of the different configurations/applications with respect to performance, energy consumption, and temperature. Design trade-offs are analyzed, stressing the interdependency of the metrics and design factors. In particular, we evaluate several chip floorplans. Their power/thermal characteristics are analyzed, showing the importance of considering thermal effects at the architectural level to achieve the best design choice.

Process variations will greatly impact the stability, leakage power consumption, and performance of future microprocessors. These variations are especially detrimental to 6T SRAM (6-transistor static memory) structures and will become critical with continued technology scaling. In this paper, we propose new on-chip memory architectures based on novel 3T1D DRAM (3-transistor, 1-diode dynamic memory) cells. We provide a detailed comparison between 6T and 3T1D designs in the context of a L1 data cache. The effects of physical device variation on a 3T1D cache can be lumped into variation of data retention times. This paper proposes a range of cache refresh and placement schemes that are sensitive to retention time, and we show that most of the retention time variations can be masked by the microarchitecture when using these schemes. We have performed detailed circuit and architectural simulations assuming different degrees of variability in advanced technology nodes, and we show that the resulting memory architecture can tolerate large process variations with little or even no impact on performance when compared to ideal 6T SRAM designs. Furthermore, these designs are robust to memory cell stability issues and can achieve large power savings. These advantages make the new memory architectures a promising choice for on-chip variation-tolerant cache structures required for next generation microprocessors.

Multicore architectures are ruling the recent microprocessor design trend. This is due to different reasons: better performance, thread-level parallelism bounds in modern applications, ILP diminishing returns, better thermal/power scaling (many small cores dissipate less than a large and complex one); and, ease and reuse of design.This paper presents a thorough evaluation of multicore architectures. The architecture we target is composed of a configurable number of cores, a memory hierarchy consisting of private L1 and L2, and a shared bus interconnect. We consider parallel shared memory applications. We explore the design space related to the number of cores, L2 cache size and processor complexity, showing the behavior of the different configurations/applications with respect to performance, energy consumption and temperature. Design tradeoffs are analyzed, stressing the interdependency of the metrics and design factors. In particular, we evaluate several chip floorplans. Their power/thermal characteristics are analyzed and they show the importance of considering thermal effects at the architectural level to achieve the best design choice.

Operand gating is a technique for improving processor energy efficiency by gating off sections of the data path that are unneeded by short-precision (narrow) operands. A method for implementing software-controlled power gating is proposed and evaluated. The instruction set architecture (ISA) is enhanced to include opcodes that specify operand widths (if not already included in the ISA). A compiler or a binary translator uses statically available information to determine initial value ranges. The technique is enhanced through a profile-based analysis that results in the specialization of certain code regions for a given value range. After the analysis, instruction opcodes are assigned using the minimum required width. To evaluate this technique the Alpha instruction set is enhanced to include opcodes for 8, 16, and 32 bit operands. Applying the proposed software technique to the Speclnt95 benchmarks results in energy-delay savings of 14%. When combined with previously proposed hardware-based techniques, the energy-delay benefit is 28%.

The issue logic of dynamically scheduled superscalar processors is one of their most complex and power-consuming parts. In this paper we present alternative issue-logic designs that are much simpler than the traditional scheme while they retain most of its ability to exploit ILP. These alternative schemes are based on the observation that most values produced by a program are used by very few instructions, and the latencies of most operation are deterministic.

Clustered microarchitectures are an effective approach to reducing the penalties caused by wire delays inside a chip. Current superscalar processors have in fact a two-cluster microarchitecture with a naive code partitioning approach: integer instructions are allocated to one cluster and floating-point instructions to the other. This partitioning scheme is simple and results in no communications between the two clusters (just through memory) but it is in general far from optimal because she workload is not evenly distributed most of the time. In fact, when the processor is running integer programs, the workload is extremely unbalanced since the FP cluster is not used at all. In this work we investigate run-time mechanisms that dynamically distribute the instructions of a program among these two clusters. By optimizing the trade-off between inter-cluster communication penalty and workload balance, the proposed schemes can achieve an average speed-up of 36% for the SpecInt95 benchmark suite.

One of the main concerns in today's processor design is the issue logic. Instruction-level parallelism is usually favored by an out-of-order issue mechanism where instructions can issue independently of the program order. The out-of-order scheme yields the best performance but at the same time introduces important hardware costs such as an associative look-up, which might be prohibitive for wide issue processors with large instruction windows. This associative search may slow-down the clock-rate and it has an important impact on power consumption. In this work, two new issue schemes that reduce the hardware complexity of the issue logic with minimal impact on the average number of instructions executed per cycle are presented.

Data, addresses, and instructions are compressed by maintaining only significant bytes with two or three extension bits appended to indicate the significant byte positions. This significance compression method is integrated into a 5-stage pipeline, with the extension bits flowing down the pipeline to enable pipeline operations only for the significant bytes. Consequently, register logic and cache activity (and dynamic power) are substantially reduced. An initial trace-driven study shows reduction in activity of approximately 30-40% for each pipeline stage. Several pipeline organizations are studied. A byte serial pipeline is the simplest implementation, but suffers a CPI (cycles per instruction) increase of 79% compared with a conventional 32-bit pipeline. Widening certain pipeline stages in order to balance processing bandwidth leads to an implementation with a CPI 24% higher than the baseline 32-bit design. Finally, full-width pipeline stages with operand gating achieve a CPI within 2-6% of the baseline 32-bit pipeline.