Limited by the breakdown of technology scaling, CPU architects are looking for creative solutions to deliver performance improvements while unable to traditionally scale up microarchitectural structures. One promising approach is engineering a microarchitecture that uses resources more efficiently, for example, by recycling them faster to relieve pressure on critical structures and making them look bigger than they are. The physical register file (PRF) is a key structure that faces severe area and power constraints that limit its (and the whole CPU's) scalability. Based on this observation, previous work proposed solutions to reduce the pressure on the PRF by reducing the time each register remains allocated. In this paper, we corroborate earlier findings that the early release of registers is a promising approach to reduce the occupancy of the PRF and we identify novel tight conditions for safely releasing registers. Based on this analysis, we design Tempranillo: an aggressive, non-speculative microarchitecture to release registers as early as possible without requiring additional recovery mechanisms. Tempranillo delivers up to 3.3 % and 11.8 % performance improvement over conventional release on singlethreaded and a 2 -way SMT CPUs, respectively. Additionally, Tempranillo requires modest storage overheads, translating into a performance improvement per KiB of storage of up to 2.6 % and 9.3 % for single-thread and 2-way SMT, respectively. Our evaluation shows that Tempranillo improves over both the state-of-the-art non-speculative and speculative proposals.

Modern datacenter applications commonly adopt a microservice software architecture, where an application is decomposed into smaller interconnected microservices communicating via the network.These microservices often operate under strict latency requirements, rendering them particularly vulnerable to microarchitectural cold effects that may arise from the interleaved execution of services on cores or power-gating cores between invocations.Previous analyses of microservices find branch mispredictions due to cold predictor resources to be a significant contributor to performance degradation, indicating that the dynamic control flow must be very similar in the set and order of executed instructions across different requests.Our analysis of control-flow similarity across requests, using static and dynamic control flow information to determine dynamic control flow reconvergence, confirms that, indeed, a large portion of requests follow similar paths.Motivated by the above findings, we propose Similarity-based Branch Prediction (SBP), a hybrid predictor architecture that enhances conventional predictors with a similarity component.SBP leverages the control-flow similarity across microservice requests to predict control flow (branch direction and target) by utilizing the control flow of past executions encoded in a reference execution trace.We realize a specific instantiation of SBP, called CHESS, which combines a conventional history-based fetch predictor, a static-hint predictor, and a similarity predictor.CHESS judiciously applies similarity prediction for branches identified as hard-to-predict through conventional prediction techniques, effectively mitigating branch predictor cold-start effects while keeping the length of the reference trace practical.Evaluation through a suite of microservices shows that CHESS reduces branch MPKI by 94% over a cold fetch predictor and 78% over a state-of-the-art predictor, while requiring a modest 18.1KB of additional storage space.This enables CHESS to deliver performance that is, on average, within 95% of a warm baseline system.

Latency-critical applications running in modern datacenters exhibit irregular request arrival patterns and are implemented using multiple services with strict latency requirements (30–250μs). These characteristics render existing energy-saving idle CPU sleep states ineffective due to the performance overhead caused by the state’s transition latency. Besides the state transition latency, another important contributor to the performance overhead of sleep states is the cold-start latency, or in other words, the time required to warm up the microarchitectural state (e.g., cache contents, branch predictor metadata) that is flushed or discarded when transitioning to a lower-power state. Both the transition latency and cold-start latency can be particularly detrimental to the performance of latency critical applications with short execution times. While prior work focuses on mitigating the effects of transition and cold-start latency by optimizing request scheduling, in this work we propose a redesign of the core C-state architecture for latency-critical applications. In particular, we introduce C6Awarm, a new Agile core C-state that drastically reduces the performance overhead caused by idle sleep state transition latency and cold-start latency while maintaining significant energy savings. C6Awarm achieves its goals by (1) implementing medium-grained power gating, (2) preserving the microarchitectural state of the core, and (3) keeping the clock generator and PLL active and locked. Our analysis for a set of microservices based on an Intel Skylake server shows that C6Awarm manages to reduce the energy consumption by up to 70% with limited performance degradation (at most 2%).

Address translation is a performance bottleneck in data-intensive workloads due to large datasets and irregular access patterns that lead to frequent high-latency page table walks (PTWs). PTWs can be reduced by using (i) large hardware TLBs or (ii) large software-managed TLBs. Unfortunately, both solutions have significant drawbacks: increased access latency, power and area (for hardware TLBs), and costly memory accesses, the need for large contiguous memory blocks, and complex OS modifications (for software-managed TLBs).

Modern user-facing applications deployed in datacenters use a distributed system architecture that exacerbates the latency requirements of their constituent microservices (30-250$\mu$s). Existing CPU power-saving techniques degrade the performance of these applications due to the long transition latency (order of 100$\mu$s) to wake up from a deep CPU idle state (C-state). For this reason, server vendors recommend only enabling shallow core C-states (e.g., CC1) for idle CPU cores, thus preventing the system from entering deep package C-states (e.g., PC6) when all CPU cores are idle. This choice, however, impairs server energy proportionality since power-hungry resources (e.g., IOs, uncore, DRAM) remain active even when there is no active core to use them. As we show, it is common for all cores to be idle due to the low average utilization (e.g., 5-20%) of datacenter servers running user-facing applications. We propose to reap this opportunity with AgilePkgC (APC), a new package C-state architecture that improves the energy proportionality of server processors running latency-critical applications. APC implements PC 1A (package C l agile), a new deep package C-state that a system can enter once all cores are in a shallow C-state (i.e., CC1) and has a nanosecond-scale transition latency. PC 1A is based on four key techniques. First, a hardware-based agile power management unit (APMU) rapidly detects when all cores enter a shallow core C-state (CC1) and triggers the system-level power savings control flow. Second, an IO Standby Mode (IOSM) places IO interfaces (e.g., PCIe, DMI, UPI, DRAM) in shallow (nanosecond-scale transition latency) low-power modes. Third, a CLM Retention (CLMR) mode rapidly reduces the CLM (Cache-and-home-agent, Last-level-cache, and Mesh network-on-chip) domain’s voltage to its retention level, drastically reducing its power consumption. Fourth, APC keeps all system PLLs active in PC 1A to allow nanosecond-scale exit latency by avoiding PLL re-locking overhead. Combining these techniques enables significant power savings while requiring less than 200ns transition latency, $\gt250\times$ faster than existing deep package C-states (e.g., PC6), making PC 1A practical for datacenter servers. Our evaluation based on an Intel Skylake-based server shows that APC reduces the energy consumption of Memcached by up to 41% (25% on average) with <0.1% performance degradation. APC provides similar benefits for other representative workloads.

User-facing applications running in modern datacenters exhibit irregular request patterns and are implemented using a multitude of services with tight latency requirements (30–250$\mu$s). These characteristics render existing energy-conserving techniques ineffective when processors are idle due to the long transition time (order of 100$\mu$s) from a deep CPU core idle power state (C-state). While prior works propose management techniques to mitigate this inefficiency, we tackle it at its root with AgileWatts (AW): a new deep CPU core C-state architecture optimized for datacenter server processors targeting latency-sensitive applications.AW drastically reduces the transition latency from deep CPU core idle power states while retaining most of their power savings based on three key ideas. First, AW eliminates the latency (several microseconds) of savinglrestoring the core context when powering-off/-on the core in a deep idle state by i) implementing medium-grained power-gates, carefully distributed across the CPU core, and ii) reraining context in the power-ungated domain. Second, AW eliminates rhe flush latency (several tens of microseconds) of the LllL2 caches when entering a deep idle state by keeping LllL2 content power-ungated. A small control logic also remains ungated to serve cache coherence traffic. AW implements cache sleep-mode and leakage reduction for the power-ungated domain by lowering a core’s voltage to the minimum operational level. Third, using a state-of-the-art power efficient all-digital phase-locked loop (ADPLL) clock generator, AW keeps the PLL active and locked during the idle state, cutting microseconds of wake-up latency at negligible power cost.Our evaluation with an accurate industrial-grade simulator calibrated against an Intel Skylake server shows that AW reduces the energy consumprion of Memcached by up to 71% (35% on average) with<1% end-to-end performance degradation. We observe similar trends for other evaluated services (MySQL and Kafka). AW’s new deep C-states C6A and C6AE reduce transition-time by up to 900$\times$ as compared to the deepest existing idle state C6, while consuming only 7% and 5% of the active state (C0) power, respectively.

Most modern processors use dynamic voltage and frequency scaling (DVFS) for power management. DVFS allows to optimize power consumption by scaling voltage and frequency depending on performance demand. Previous research has indicated that this frequency scaling might pose a security threat in the form of a covert channel, which could leak sensitive information. However, an analysis able to determine whether DVFS is a serious security issue is still missing. In this paper, we conduct a detailed analysis of the threat potential of a DVFS-based covert channel. We investigate two multicore platforms representative of modern laptops and hand-held devices. Furthermore, we develop a channel model to determine an upper bound to the channel capacity, which is in the order of 1 bit per channel use. Last, we perform an experimental analysis using a novel transceiver implementation. The neural network-based receiver yields packet error rates between 1% and 8% at average throughputs of up to 1.83 and 1.20 bps for platforms representative of laptops and hand-held devices, respectively. Considering the well-known small message criterion, our results show that a relevant covert channel can be established by exploiting the behavior of computing systems with DVFS.

Modern multicore processors feature easily accessible temperature sensors that provide useful information for dynamic thermal management. These sensors were recently shown to be a potential security threat, since otherwise isolated applications can exploit them to establish a thermal covert channel and leak restricted information. Previous research showed experiments that document the feasibility of (low-rate) communication over this channel, but did not further analyze its fundamental characteristics. For this reason, the important questions of quantifying the channel capacity and achievable rates remain unanswered.

Contention on the shared Last-Level Cache (LLC) can have a fundamental negative impact on the performance of applications executed on modern multicores. An interesting software approach to address LLC contention issues is based on page coloring , which is a software technique that attempts to achieve performance isolation by partitioning a shared cache through careful memory management. The key assumption of traditional page coloring is that the cache is physically addressed. However, recent multicore architectures (e.g., Intel Sandy Bridge and later) switched from a physical addressing scheme to a more complex scheme that involves a hash function. Traditional page coloring is ineffective on these recent architectures. In this article, we extend page coloring to work on these recent architectures by proposing a mechanism able to handle their hash-based LLC addressing scheme. Just as for traditional page coloring, the goal of this new mechanism is to deliver performance isolation by avoiding contention on the LLC, thus enabling predictable performance. We implement this mechanism in the Linux kernel, and evaluate it using several benchmarks from the SPEC CPU2006 and PARSEC 3.0 suites. Our results show that our solution is able to deliver performance isolation to concurrently running applications by enforcing partitioning of a Sandy Bridge LLC, which traditional page coloring techniques are not able to handle.

Latency-critical workloads (e.g., web search), common in datacenters, require stable tail (e.g., 95th percentile) latencies of a few milliseconds. Servers running these workloads are kept lightly loaded to meet these stringent latency targets. This low utilization wastes billions of dollars in energy and equipment annually.

Efficiently exploiting multicore processors for real-time applications is challenging because jobs that run concurrently on different cores can interfere on shared resources, severely complicating precise timing analysis. We propose a new scheduling model called Isolation Scheduling (IS), IS provides a framework to exploiting multicores for real-time applications where tasks are grouped in classes. IS enforces mutually exclusive execution among different task classes, thus avoiding inter-class interference by construction. We show that IS encompasses several recent advances in real-time scheduling as special cases and we propose global and partitioned scheduling algorithms based on this model. Specific results are provided if the task classes correspond to different safety criticality levels.

Ideally, the pay-as-you-go model of Infrastructure as a Service (IaaS) clouds should enable users to rent just enough resources (e.g., CPU or memory bandwidth) to fulfill their service level objectives (SLOs). Achieving this goal is hard on current IaaS offers, which require users to explicitly specify the amount of resources to reserve; this requirement is nontrivial for users, because estimating the amount of resources needed to attain application-level SLOs is often complex, especially when resources are virtualized and the service provider colocates virtual machines (VMs) on host nodes. For this reason, users who deploy VMs subject to SLOs are usually prone to overprovisioning resources, thus resulting in inflated business costs. This article tackles this issue with AutoPro : a runtime system that enhances IaaS clouds with automated and fine-grained resource provisioning based on performance SLOs. Our main contribution with AutoPro is filling the gap between application-level performance SLOs and allocation of a contended resource, without requiring explicit reservations from users. In this article, we focus on CPU bandwidth allocation to throughput-driven, compute-intensive multithreaded applications colocated on a multicore processor; we show that a theoretically sound, yet simple, control strategy can enable automated fine-grained allocation of this contended resource, without the need for offline profiling. Additionally, AutoPro helps service providers optimize infrastructure utilization by provisioning idle resources to best-effort workloads, so as to maximize node-level utilization. Our extensive experimental evaluation confirms that AutoPro is able to automatically determine and enforce allocations to meet performance SLOs while maximizing node-level utilization by supporting batch workloads on a best-effort basis.

Motivation While the pay-as-you-go model of Infrastructure-as-a-Service (IaaS) clouds is more flexible than an in-house IT infrastructure, it still has a resource-based interface towards users, who can rent virtual computing resources over relatively long time scales. There is a fundamental mismatch between this resource-based interface and what users really care about: performance.

Motivation and Contribution The commodity multicores that power cloud infrastructures hide memory latency through deep memory hierarchies, with the last-level cache (LLC) usually shared among cores. While a shared LLC improves utilization of on-chip resources, it may also lead to unpredictable performance of colocated virtual machines (VMs) as a result of unanticipated contention. Past research showed that the operating system page allocator can favor performance predictability on a physically-addressed shared LLC through page coloring [4, 8, 9]: a software technique that can work on commodity multicores, unlike hardware approaches [2, 7]. The main drawback of page coloring is the high cost of modifying allocations (i.e., recoloring), making this technique almost impractical for applications with varying memory footprints [6].

Traditionally, hypervisors, operating systems, and runtime systems have been providing an abstraction layer over the bare-metal hardware. Traditional abstractions, however, do not consider for non-functional requirements such as system-level constraints or users' objectives. As these requirements are gaining increasing importance, researchers are looking into making user-specified and system-level objectives first-class citizens in the computer systems' realm.

Adaptive Resource-Centric Computing (ARCC) enables a simultaneous mix of high-throughput parallel, real-time, and interactive applications through automatic discovery of the correct mix of resource assignments necessary to achieve application requirements. This approach, embodied in the Tessellation manycore operating system, distributes resources to QoS domains called cells. Tessellation separates global decisions about the allocation of resources to cells from application-specific scheduling of resources within cells. We examine the implementation of ARCC in the Tessellation OS, highlight Tessellation's ability to provide predictable performance, and investigate the performance of Tessellation services within cells.

In this paper, we present Metronome: a framework to enhance commodity operating systems with self-adaptive capabilities. The Metronome framework features two distinct components: Heart Rate Monitor (HRM) and Performance--Aware Fair Scheduler (PAFS). HRM is an active monitoring infrastructure implementing the observe phase of a self--adaptive computing system Observe--Decide--Act (ODA) control loop, while PAFS is an adaptation policy implementing the decide and act phases of the control loop. Metronome was designed and developed looking towards multi--core processors; therefore, its experimental evaluation has been carried on with the PARSEC 2.1 benchmark suite.

Traditionally, hardware circuits are realized according to techniques that follow the classical phases of design and testing. A completely new approach in the creation of hardware circuits has been proposed---the Evolvable Hardware (EHW) paradigm, which bases the circuit synthesis on a goal-oriented evolutionary process inspired by biological evolution in Nature. FPGA-based approaches have emerged as the main architectural solution to implement EHW systems. Various EHW systems have been proposed by researchers but most of them, being based on outdated chips, do not take advantage of the interesting features introduced in newer FPGAs. This article describes a project named Hardware Evolution over Reconfigurable Architectures (HERA), which aims at creating a complete and performance-oriented framework for the evolution of digital circuits, leveraging the reconfiguration technology available in FPGAs. The project is described from its birth to its current state, presenting its evolutionary technique tailored for FPGA-based circuits and the most recent enhancements to improve the scalability with respect to problem size. The developed EHW system outperforms the state of the art, proving its effectiveness in evolving both standard benchmarks and more complex real-world applications.