中圖分類號(hào)： TP391 文獻(xiàn)標(biāo)識(shí)碼： A DOI：10.16157/j.issn.0258-7998.212432 中文引用格式： 劉世超,，楊斌，劉衛(wèi)國(guó). 高性能高可用Redis客戶端的設(shè)計(jì)與實(shí)現(xiàn)[J].電子技術(shù)應(yīng)用,，2022,，48(1)：46-52，58. 英文引用格式： Liu Shichao,，Yang Bin,，Liu Weiguo. Design and implementation of high-performance and high-availability redis client[J]. Application of Electronic Technique，2022,，48(1)：46-52,，58.
Design and implementation of high-performance and high-availability redis client
Liu Shichao1，2,，Yang Bin1,，2，Liu Weiguo1,，2
1.School of Software,，Shandong University，Jinan 250101,，China；2.National Supercomputing Center in Wuxi,，Wuxi 214072,，China
Abstract： Redis is an unstructured database based on memory storage. It is known for high I/O(Input/Output) performance and high response speed. It plays an important role in data buffering, message queues, key-value storage and other scenarios. Among the many clients it supports, the C/C++ client Hiredis is particularly widely used. This article did an in-depth analysis of the Hiredis library and found that its pipeline function has high overhead, improper instruction storage, and memory confusion problems. Based on this, this article designs and implements a C/C++-oriented high-performance and high-availability Redis client on a 32-core X86 architecture processor and a 64 GB memory Linux server. It improves the performance of processing a large number of instructions and solves the problem of memory confusion in complex scenarios through memory pre-allocation and memory isolation. After testing, the new client has improved instruction execution efficiency by 3～7 times, while also ensuring memory safety and accuracy in complex scenarios.
Key words : Redis,；pipeline；Hiredis,；memory confusion,；performance optimization
0 引言
隨著互聯(lián)網(wǎng)飛速發(fā)展以及大規(guī)模應(yīng)用的不斷涌現(xiàn)，目前已經(jīng)步入了大數(shù)據(jù)時(shí)代,。非結(jié)構(gòu)化數(shù)據(jù)逐漸替代了傳統(tǒng)結(jié)構(gòu)化數(shù)據(jù)并迅速占據(jù)了主導(dǎo)地位,，為了管理形式多樣的非結(jié)構(gòu)化數(shù)據(jù)，涌現(xiàn)了諸如MongoDB[1],、InfluxDB[2],、Elasticsearch[3]等十分有代表性的數(shù)據(jù)庫(kù)。這些數(shù)據(jù)庫(kù)雖然針對(duì)非結(jié)構(gòu)化數(shù)據(jù)的存取做了很多優(yōu)化,，但是受限于硬盤(Hard Disk Drive,，HDD)等底層存儲(chǔ)介質(zhì)，往往無(wú)法滿足高性能場(chǎng)景的需求,。
為了提高性能,，以Redis為代表的內(nèi)存數(shù)據(jù)庫(kù)應(yīng)運(yùn)而生。Redis是一個(gè)非結(jié)構(gòu)化數(shù)據(jù)庫(kù),，支持使用非結(jié)構(gòu)化語(yǔ)言(Not-only Structured Query Language,，NoSQL)查詢。同時(shí),，Redis通過(guò)I/O(Input/Output)多路復(fù)用和DRAM(Dynamic Random Access Memory)提供了高吞吐,、高并發(fā)和低時(shí)延的服務(wù)，在數(shù)據(jù)緩沖,、消息隊(duì)列,、Key-Value存儲(chǔ)等場(chǎng)景都發(fā)揮了重要的作用。
