Scale-out datacenters mandate high per-server throughput to get the maximum benefit from the large TCO investment. Emerging applications (e.g., data serving and web search) that run in these datacenters operate on vast datasets that are not accommodated by on-die caches of existing server chips. Large caches reduce the die area available for cores and lower performance through long access latency when instructions are fetched. Performance on scale-out workloads is maximized through a modestly-sized last-level cache that captures the instruction footprint at the lowest possible access latency. In this work, we introduce a methodology for designing scalable and efficient scale-out server processors. Based on a metric of performance-density, we facilitate the design of optimal multi-core configurations, called pods. Each pod is a complete server that tightly couples a number of cores to a small last-level cache using a fast interconnect. Replicating the pod to fill the die area yields processors which have optimal performance density, leading to maximum per-chip throughput. Moreover, as each pod is a stand-alone server, scale-out processors avoid the expense of global (i.e., interpod) interconnect and coherence. These features synergistically maximize throughput, lower design complexity, and improve technology scalability. In 20nm technology, scaleout chips improve throughput by 5x-6.5x over conventional and by 1.6x-1.9x over emerging tiled organizations.

Emerging scale-out workloads require extensive amounts of computational resources. However, data centers using modern server hardware face physical constraints in space and power, limiting further expansion and calling for improvements in the computational density per server and in the per-operation energy. Continuing to improve the computational resources of the cloud while staying within physical constraints mandates optimizing server efficiency to ensure that server hardware closely matches the needs of scale-out workloads.In this work, we introduce CloudSuite, a benchmark suite of emerging scale-out workloads. We use performance counters on modern servers to study scale-out workloads, finding that today's predominant processor micro-architecture is inefficient for running these workloads. We find that inefficiency comes from the mismatch between the workload needs and modern processors, particularly in the organization of instruction and data memory systems and the processor core micro-architecture. Moreover, while today's predominant micro-architecture is inefficient when executing scale-out workloads, we find that continuing the current trends will further exacerbate the inefficiency in the future. In this work, we identify the key micro-architectural needs of scale-out workloads, calling for a change in the trajectory of server processors that would lead to improved computational density and power efficiency in data centers. Categories and Subject Descriptors C.4 [Performance of Systems]: Performance of Systems -Design studiesGeneral Terms Design, Measurement, Performance• Instruction-and memory-level parallelism in scale-out workloads is low. Modern aggressive out-of-order cores are excessively complex, consuming power and on-chip area without providing performance benefits to scale-out workloads.• Data working sets of scale-out workloads considerably exceed the capacity of on-chip caches. Processor real-estate and power are misspent on large last-level caches that do not contribute to improved scale-out workload performance.

Emerging GPU applications exhibit increasingly high computation demands which has led GPU manufacturers to build GPUs with an increasingly large number of streaming multiprocessors (SMs). Providing data to the SMs at high bandwidth puts significant pressure on the memory hierarchy and the Network-on-Chip (NoC). Current GPUs typically partition the memory-side last-level cache (LLC) in equally-sized slices that are shared by all SMs. Although a shared LLC typically results in a lower miss rate, we find that for workloads with high degrees of data sharing across SMs, a private LLC leads to a significant performance advantage because of increased bandwidth to replicated cache lines across different LLC slices.In this paper, we propose adaptive memory-side last-level GPU caching to boost performance for sharing-intensive workloads that need high bandwidth to read-only shared data. Adaptive caching leverages a lightweight performance model that balances increased LLC bandwidth against increased miss rate under private caching. In addition to improving performance for sharing-intensive workloads, adaptive caching also saves energy in a (co-designed) hierarchical two-stage crossbar NoC by power-gating and bypassing the second stage if the LLC is configured as a private cache. Our experimental results using 17 GPU workloads show that adaptive caching improves performance by 28.1% on average (up to 38.1%) compared to a shared LLC for sharing-intensive workloads. In addition, adaptive caching reduces NoC energy by 26.6% on average (up to 29.7%) and total system energy by 6.1% on average (up to 27.2%) when configured as a private cache. Finally, we demonstrate through a GPU NoC design space exploration that a hierarchical two-stage crossbar is both more power-and area-efficient than full and concentrated crossbars with the same bisection bandwidth, thus providing a low-cost cooperative solution to exploit workload sharing behavior in memory-side last-level caches. CCS CONCEPTS• Computer systems organization → Single instruction, multiple data.

Runahead execution improves processor performance by accurately prefetching long-latency memory accesses. When a long-latency load causes the instruction window to fill up and halt the pipeline, the processor enters runahead mode and keeps speculatively executing code to trigger accurate prefetches. A recent improvement tracks the chain of instructions that leads to the long-latency load, stores it in a runahead buffer, and executes only this chain during runahead execution, with the purpose of generating more prefetch requests. Unfortunately, all prior runahead proposals have shortcomings that limit performance and energy efficiency because they release processor state when entering runahead mode and then need to re-fill the pipeline to restart normal operation. Moreover, runahead buffer limits prefetch coverage by tracking only a single chain of instructions that leads to the same long-latency load.We propose precise runahead execution (PRE) which builds on the key observation that when entering runahead mode, the processor has enough issue queue and physical register file resources to speculatively execute instructions. This mitigates the need to release and re-fill processor state in the ROB, issue queue, and physical register file. In addition, PRE preexecutes only those instructions in runahead mode that lead to full-window stalls, using a novel register renaming mechanism to quickly free physical registers in runahead mode, further improving efficiency and effectiveness. Finally, PRE optionally buffers decoded runahead micro-ops in the front-end to save energy. Our experimental evaluation using a set of memoryintensive applications shows that PRE achieves an additional 18.2% performance improvement over the recent runahead proposals while at the same time reducing energy consumption by 6.8%.

No abstract