RR-Compound: RDMA-Fused gRPC for Low Latency and High Throughput With an Easy Interface

远程直接内存访问 计算机科学 吞吐量 接口(物质) 低延迟(资本市场) 延迟(音频) 并行计算 计算机硬件 操作系统 计算机网络 电信 气泡 最大气泡压力法 无线
作者
Liang Geng,Hao Wang,Jingsong Meng,Dayi Fan,Sami Ben-Romdhane,Hari Kadayam Pichumani,Vinay Phegade,X Zhang
出处
期刊:IEEE Transactions on Parallel and Distributed Systems [Institute of Electrical and Electronics Engineers]
卷期号:35 (8): 1488-1505
标识
DOI:10.1109/tpds.2024.3404394
摘要

Advanced data centers strive for high performance and throughput, which can be achieved through the desirable merits of Remote Procedure Call (RPC) programming model and the low latency of Remote Direct Memory Access (RDMA). However, despite the widespread availability of these software and hardware utilities, they have been utilized separately for their own applications in existing production systems for many years. Although researchers have attempted to develop RDMA-enabled RPC prototypes, they often face challenges such as API discrepancies and a lack of specific features for effective integration with major production software, rendering them incompatible. This industry R&D project aims to enhance the performance of gRPC, a widely utilized RPC framework in major companies, by integrating RDMA as an internal component. Our system solution, called, combines the simple user interface and other merits of gRPC with low latency for remote data accesses. RR-Compound is fully compatible with gRPC and can serve as a seamless replacement without altering existing applications. However, to achieve low latency, high throughput, and scalability for RR-Compound, several technical challenges in managing network connections and memory space utilization must be effectively addressed. To overcome the limitations of existing connection methods, we have developed a new method called BPEV that is independent of gRPC and applicable to all RDMA systems. We have also retained the asynchronous framework of gRPC, albeit with limited buffer space in RDMA memory management. In micro-benchmarks, RR-Compound outperforms mRPC - the state-of-the-art RPC framework for a large number of connections, achieving a 14.77% increase in throughput and a 42.55% reduction in latency. Subsequently, we compare RR-Compound with gRPC over IPoIB using two real-world applications: KV-Store and TensorFlow. RR-Compound achieves up to a 2.35x increase in throughput and reduces the average latency by 46.92%.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
饭饭完成签到,获得积分10
刚刚
1秒前
1秒前
小鱼发布了新的文献求助10
2秒前
漂亮的丝袜完成签到,获得积分20
2秒前
2秒前
巴旦木发布了新的文献求助10
3秒前
韵寒发布了新的文献求助10
3秒前
大道发布了新的文献求助10
4秒前
4秒前
4秒前
4秒前
冷艳尔云完成签到,获得积分20
4秒前
韩较瘦完成签到,获得积分10
5秒前
万能图书馆应助雨辰采纳,获得10
5秒前
5秒前
cxy完成签到,获得积分20
5秒前
藏藏完成签到,获得积分10
5秒前
微雨发布了新的文献求助10
6秒前
6秒前
7秒前
科研通AI2S应助楼一笑采纳,获得10
8秒前
LYF完成签到,获得积分10
8秒前
jing完成签到,获得积分10
8秒前
9秒前
9秒前
9秒前
孙明浩完成签到 ,获得积分10
9秒前
我是老大应助Young离子采纳,获得10
9秒前
9秒前
你好发布了新的文献求助10
10秒前
Amity完成签到 ,获得积分10
11秒前
yyyyyy发布了新的文献求助10
11秒前
雨辰完成签到,获得积分10
11秒前
小胖胖完成签到,获得积分10
12秒前
12秒前
站走跑发布了新的文献求助10
12秒前
坚强的乐驹完成签到,获得积分10
13秒前
lilixia发布了新的文献求助10
13秒前
14秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Polymorphism and polytypism in crystals 1000
Signals, Systems, and Signal Processing 610
Discrete-Time Signals and Systems 610
Russian Politics Today: Stability and Fragility (2nd Edition) 500
Death Without End: Korea and the Thanatographics of War 500
Der Gleislage auf der Spur 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 纳米技术 有机化学 物理 生物化学 化学工程 计算机科学 复合材料 内科学 催化作用 光电子学 物理化学 电极 冶金 遗传学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 6083139
求助须知:如何正确求助?哪些是违规求助? 7913503
关于积分的说明 16367898
捐赠科研通 5218355
什么是DOI,文献DOI怎么找? 2789901
邀请新用户注册赠送积分活动 1772906
关于科研通互助平台的介绍 1649295