Integrated 3D-stacked server designs for increasing physical density of key-value stores
Integrated 3D-stacked server designs for increasing physical density of key-value stores
复制标题
集成 3D 堆叠服务器设计，用于提高键值存储的物理密度
DOI：
10.1145/2541940.2541951
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
T. Mudge
中科院分区：
文献类型：
--
作者：
Anthony Gutierrez;Michael Cieslak;Bharan Giridhar;R. Dreslinski;L. Ceze;T. Mudge
自Web 2.0时代开始以来，键值存储（如Memcached）一直用于扩展Web服务。数据中心的真实的资产是昂贵的，我们采访过的几位行业专家建议，他们的数据中心空间的很大一部分用于关键价值存储。尽管它被广泛使用，但在硬件专业化方面几乎没有提高Memcached的效率和密度;它目前部署在包含高端CPU的商品服务器上，这些CPU旨在提取尽可能多的并行级别。然而，乱序CPU在运行Memcached时效率低下。为了解决Memcached的效率问题，我们提出了两种使用3D堆栈来提高数据存储效率的架构。我们的第一个3D架构Mercury由ARM Cortex-A7内核堆栈组成，具有4GB的DRAM和1000。我们的第二个架构Iridium用NAND Flash取代DRAM，以提高密度。我们通过模拟来探索在使用3D堆栈的服务器上运行Memcached的潜在效率优势，以将低功耗CPU与CPU和内存紧密集成。通过Mercury，我们证明，与运行优化的Memcached的最先进的服务器相比，密度可以提高2.9倍，能效提高4.9倍，吞吐量提高10倍，每GB吞吐量提高3.5倍。通过Iridium，我们发现密度可以提高14倍，功率效率提高2.4倍，吞吐量提高5.2倍，同时仍然满足大多数请求的延迟要求。
Key-value stores, such as Memcached, have been used to scale web services since the beginning of the Web 2.0 era. Data center real estate is expensive, and several industry experts we have spoken to have suggested that a significant portion of their data center space is devoted to key value stores. Despite its wide-spread use, there is little in the way of hardware specialization for increasing the efficiency and density of Memcached; it is currently deployed on commodity servers that contain high-end CPUs designed to extract as much instruction-level parallelism as possible. Out-of-order CPUs, however have been shown to be inefficient when running Memcached. To address Memcached efficiency issues, we propose two architectures using 3D stacking to increase data storage efficiency. Our first 3D architecture, Mercury, consists of stacks of ARM Cortex-A7 cores with 4GB of DRAM, as well as NICs. Our second architecture, Iridium, replaces DRAM with NAND Flash to improve density. We explore, through simulation, the potential efficiency benefits of running Memcached on servers that use 3D-stacking to closely integrate low-power CPUs with NICs and memory. With Mercury we demonstrate that density may be improved by 2.9X, power efficiency by 4.9X, throughput by 10X, and throughput per GB by 3.5X over a state-of-the-art server running optimized Memcached. With Iridium we show that density may be increased by 14X, power efficiency by 2.4X, and throughput by 5.2X, while still meeting latency requirements for a majority of requests.