Speculative execution, the base on which modern high-performance general-purpose CPUs are built on, has recently been shown to enable a slew of security attacks. All these attacks are centered around a common set of behaviors: During speculative execution, the architectural state of the system is kept unmodified, until the speculation can be verified. In the event that a misspeculation occurs, then anything that can affect the architectural state is reverted (squashed) and re-executed correctly. However, the same is not true for the microarchitectural state. Normally invisible to the user, changes to the microarchitectural state can be observed through various sidechannels, with timing differences caused by the memory hierarchy being one of the most common and easy to exploit. The speculative side-channels can then be exploited to perform attacks that can bypass software and hardware checks in order to leak information. These attacks, out of which the most infamous are perhaps Spectre and Meltdown, have led to a frantic search for solutions.In this work, we present our own solution for reducing the microarchitectural state-changes caused by speculative execution in the memory hierarchy. It is based on the observation that if we only allow accesses that hit in the L1 data cache to proceed, then we can easily hide any microarchitectural changes until after the speculation has been verified. At the same time, we propose to prevent stalls by value predicting the loads that miss in the L1. Value prediction, though speculative, constitutes an invisible form of speculation, not seen outside the core. We evaluate our solution and show that we can prevent observable microarchitectural changes in the memory hierarchy while keeping the performance and energy costs at 11% and 7%, respectively. In comparison, the current state of the art solution, InvisiSpec, incurs a 46% performance loss and a 51% energy increase.

Matrix-matrix multiplication is a key computational kernel for numerous applications in science and engineering, with ample parallelism and data locality that lends itself well to high-performance implementations. Many matrix multiplicationdependent applications can use reduced-precision integer or fixedpoint representations to increase their performance and energy efficiency while still offering adequate quality of results. However, precision requirements may vary between different application phases or depend on input data, rendering constant-precision solutions ineffective. We present BISMO, a vectorized bitserial matrix multiplication overlay for reconfigurable computing. BISMO utilizes the excellent binary-operation performance of FPGAs to offer a matrix multiplication performance that scales with required precision and parallelism. We characterize the resource usage and performance of BISMO across a range of parameters to build a hardware cost model, and demonstrate a peak performance of 6.5 TOPS on the Xilinx PYNQ-Z1 board.

No abstract

No abstract

To enhance the performance of memory-bound applications, hardware designs have been developed to hide memory latency, such as the out-of-order (OoO) execution engine, at the price of increased energy consumption. Contemporary processor cores span a wide range of performance and energy efficiency options: from fast and power-hungry OoO processors to efficient, but slower in-order processors. The more memory-bound an application is, the more aggressive the OoO execution engine has to be to hide memory latency.This proposal targets the middle ground, as seen in a simple OoO core, which strikes a good balance between performance and energy efficiency and currently dominates the market for mobile, hand-held devices and high-end embedded systems. We show that these simple, more energy-efficient OoO cores, equipped with the appropriate compile-time support, considerably boost the performance of single-threaded execution and reach new levels of performance for memorybound applications.Clairvoyance generates code that is able to hide memory latency and better utilize the OoO engine, thus delivering higher performance at lower energy. To this end, Clairvoyance overcomes restrictions which yielded conventional compile-time techniques impractical: (i) statically unknown dependencies, (ii) insufficient independent instructions, and (iii) register pressure. Thus, Clairvoyance achieves a geomean execution time improvement of 7% for memory-bound applications with a conservative approach and 13% with a speculative but safe approach, on top of standard O3 optimizations, while maintaining compute-bound applications' high-performance.

Many of the important services running on data centres are latency-critical, time-varying, and demand strict user satisfaction. Stringent tail-latency targets for colocated services and increasing system complexity make it challenging to reduce the power consumption of data centres. Data centres typically sacrifice server efficiency to maintain tail-latency targets resulting in an increased total cost of ownership. This paper introduces Twig, a scalable quality-of-service (QoS) aware task manager for latency-critical services co-located on a server system. Twig successfully leverages deep reinforcement learning to characterise tail latency using hardware performance counters and to drive energy-efficient task management decisions in data centres. We evaluate Twig on a typical data centre server managing four widely used latency-critical services. Our results show that Twig outperforms prior works in reducing energy usage by up to 38% while achieving up to 99% QoS guarantee for latency-critical services.

The modified-Booth algorithm is extensively used for high-speed multiplier circuits. Once, when array multipliers were used, the reduced number of generated partial products significantly improved multiplier performance. In designs based on reduction trees with logarithmic logic depth, however, the reduced number of partial products has a limited impact on overall performance. The Baugh-Wooley algorithm is a different scheme for signed multiplication, but is not so widely adopted because it may be complicated to deploy on irregular reduction trees. We use the Baugh-Wooley algorithm in our High Performance Multiplier (HPM) tree, which combines a regular layout with a logarithmic logic depth. We show for a range of operator bitwidths that, when implemented in 130-nm and 65-nm process technologies, the Baugh-Wooley multipliers exhibit comparable delay, less power dissipation and smaller area foot-print than modified-Booth multipliers.

Abstract-Due to performance reasons, all ways in setassociative level-one (L1) data caches are accessed in parallel for load operations even though the requested data can only reside in one of the ways. Thus, a significant amount of energy is wasted when loads are performed. We propose a speculation technique that performs the tag comparison in parallel with the address calculation, leading to the access of only one way during the following cycle on successful speculations. The technique incurs no execution time penalty, has an insignificant area overhead, and does not require any customized SRAM implementation. Assuming a 16kB 4-way set-associative L1 data cache implemented in a 65-nm process technology, our evaluation based on 20 different MiBench benchmarks shows that the proposed technique on average leads to a 24% data cache energy reduction.