聊聊tomcat的keepAlive参数

序

本文主要研究一下tomcat的keepAlive参数

maxKeepAliveRequests

org/apache/tomcat/util/net/AbstractEndpoint.java

    /*** Max keep alive requests*/private int maxKeepAliveRequests=100; // as in Apache HTTPD serverpublic int getMaxKeepAliveRequests() {// Disable keep-alive if the server socket is not boundif (bindState.isBound()) {return maxKeepAliveRequests;} else {return 1;}}public void setMaxKeepAliveRequests(int maxKeepAliveRequests) {this.maxKeepAliveRequests = maxKeepAliveRequests;}

AbstractEndpoint定义了maxKeepAliveRequests属性，默认为100

Http11Processor

org/apache/coyote/http11/Http11Processor.java

public class Http11Processor extends AbstractProcessor {private static final Log log = LogFactory.getLog(Http11Processor.class);public SocketState service(SocketWrapperBase<?> socketWrapper)throws IOException {RequestInfo rp = request.getRequestProcessor();rp.setStage(org.apache.coyote.Constants.STAGE_PARSE);// Setting up the I/OsetSocketWrapper(socketWrapper);// FlagskeepAlive = true;openSocket = false;readComplete = true;boolean keptAlive = false;SendfileState sendfileState = SendfileState.DONE;//......while (!getErrorState().isError() && keepAlive && !isAsync() && upgradeToken == null &&sendfileState == SendfileState.DONE && !protocol.isPaused()) {//......int maxKeepAliveRequests = protocol.getMaxKeepAliveRequests();if (maxKeepAliveRequests == 1) {keepAlive = false;} else if (maxKeepAliveRequests > 0 &&socketWrapper.decrementKeepAlive() <= 0) {keepAlive = false;}   //......rp.setStage(org.apache.coyote.Constants.STAGE_KEEPALIVE);sendfileState = processSendfile(socketWrapper);            }         }        }

Http11Processor的service方法会从protocol获取maxKeepAliveRequests，若为1则重置keepAlive为false，若大于0则执行socketWrapper.decrementKeepAlive()，若小于等于0则重置keepAlive为false，跳出循环，requestInfo的stage不是STAGE_KEEPALIVE

decrementKeepAlive

org/apache/tomcat/util/net/SocketWrapperBase.java

public abstract class SocketWrapperBase<E> {private static final Log log = LogFactory.getLog(SocketWrapperBase.class);protected static final StringManager sm = StringManager.getManager(SocketWrapperBase.class);private E socket;private final AbstractEndpoint<E,?> endpoint;protected final AtomicBoolean closed = new AtomicBoolean(false);// Volatile because I/O and setting the timeout values occurs on a different// thread to the thread checking the timeout.private volatile long readTimeout = -1;private volatile long writeTimeout = -1;private volatile int keepAliveLeft = 100;//......public void setKeepAliveLeft(int keepAliveLeft) { this.keepAliveLeft = keepAliveLeft; }public int decrementKeepAlive() { return (--keepAliveLeft); }}

SocketWrapperBase定义了keepAliveLeft属性，默认为100，它提供了setKeepAliveLeft方法以及decrementKeepAlive方法

prepareResponse

org/apache/coyote/http11/Http11Processor.java

    protected final void prepareResponse() throws IOException {boolean entityBody = true;contentDelimitation = false;OutputFilter[] outputFilters = outputBuffer.getFilters();if (http09 == true) {// HTTP/0.9outputBuffer.addActiveFilter(outputFilters[Constants.IDENTITY_FILTER]);outputBuffer.commit();return;}//......// If we know that the request is bad this early, add the// Connection: close header.if (keepAlive && statusDropsConnection(statusCode)) {keepAlive = false;}if (!keepAlive) {// Avoid adding the close header twiceif (!connectionClosePresent) {headers.addValue(Constants.CONNECTION).setString(Constants.CLOSE);}} else if (!getErrorState().isError()) {if (!http11) {headers.addValue(Constants.CONNECTION).setString(Constants.KEEP_ALIVE_HEADER_VALUE_TOKEN);}if (protocol.getUseKeepAliveResponseHeader()) {boolean connectionKeepAlivePresent =isConnectionToken(request.getMimeHeaders(), Constants.KEEP_ALIVE_HEADER_VALUE_TOKEN);if (connectionKeepAlivePresent) {int keepAliveTimeout = protocol.getKeepAliveTimeout();if (keepAliveTimeout > 0) {String value = "timeout=" + keepAliveTimeout / 1000L;headers.setValue(Constants.KEEP_ALIVE_HEADER_NAME).setString(value);if (http11) {// Append if there is already a Connection header,// else create the headerMessageBytes connectionHeaderValue = headers.getValue(Constants.CONNECTION);if (connectionHeaderValue == null) {headers.addValue(Constants.CONNECTION).setString(Constants.KEEP_ALIVE_HEADER_VALUE_TOKEN);} else {connectionHeaderValue.setString(connectionHeaderValue.getString() + ", " + Constants.KEEP_ALIVE_HEADER_VALUE_TOKEN);}}}}}}//......        }    /*** Determine if we must drop the connection because of the HTTP status* code.  Use the same list of codes as Apache/httpd.*/private static boolean statusDropsConnection(int status) {return status == 400 /* SC_BAD_REQUEST */ ||status == 408 /* SC_REQUEST_TIMEOUT */ ||status == 411 /* SC_LENGTH_REQUIRED */ ||status == 413 /* SC_REQUEST_ENTITY_TOO_LARGE */ ||status == 414 /* SC_REQUEST_URI_TOO_LONG */ ||status == 500 /* SC_INTERNAL_SERVER_ERROR */ ||status == 503 /* SC_SERVICE_UNAVAILABLE */ ||status == 501 /* SC_NOT_IMPLEMENTED */;}

Http11Processor的prepareResponse方法在statusDropsConnection时会设置keepalive为false，对于为false的会判断是否已经有Connection: close的header，如果没有则给添加上

keepAliveTimeout

org/apache/tomcat/util/net/AbstractEndpoint.java

public abstract class AbstractEndpoint<S,U> {//....../*** Keepalive timeout, if not set the soTimeout is used.*/private Integer keepAliveTimeout = null;public int getKeepAliveTimeout() {if (keepAliveTimeout == null) {return getConnectionTimeout();} else {return keepAliveTimeout.intValue();}}public void setKeepAliveTimeout(int keepAliveTimeout) {this.keepAliveTimeout = Integer.valueOf(keepAliveTimeout);}/*** Socket timeout.** @return The current socket timeout for sockets created by this endpoint*/public int getConnectionTimeout() { return socketProperties.getSoTimeout(); }public void setConnectionTimeout(int soTimeout) { socketProperties.setSoTimeout(soTimeout); }//......
}

AbstractEndpoint定义了keepAliveTimeout熟悉，默认为null取的是socketProperties.getSoTimeout()的值

protocol.getKeepAliveTimeout()

org/apache/coyote/http11/Http11Processor.java

    public SocketState service(SocketWrapperBase<?> socketWrapper)throws IOException {RequestInfo rp = request.getRequestProcessor();rp.setStage(org.apache.coyote.Constants.STAGE_PARSE);// Setting up the I/OsetSocketWrapper(socketWrapper);// FlagskeepAlive = true;openSocket = false;readComplete = true;boolean keptAlive = false;SendfileState sendfileState = SendfileState.DONE;while (!getErrorState().isError() && keepAlive && !isAsync() && upgradeToken == null &&sendfileState == SendfileState.DONE && !protocol.isPaused()) {// Parsing the request headertry {if (!inputBuffer.parseRequestLine(keptAlive, protocol.getConnectionTimeout(),protocol.getKeepAliveTimeout())) {if (inputBuffer.getParsingRequestLinePhase() == -1) {return SocketState.UPGRADING;} else if (handleIncompleteRequestLineRead()) {break;}}}//......}}

Http11Processor的service方法在keepalive为true时会执行inputBuffer.parseRequestLine(keptAlive, protocol.getConnectionTimeout(),protocol.getKeepAliveTimeout())

parseRequestLine

org/apache/coyote/http11/Http11InputBuffer.java

    /*** Read the request line. This function is meant to be used during the* HTTP request header parsing. Do NOT attempt to read the request body* using it.** @throws IOException If an exception occurs during the underlying socket* read operations, or if the given buffer is not big enough to accommodate* the whole line.** @return true if data is properly fed; false if no data is available* immediately and thread should be freed*/boolean parseRequestLine(boolean keptAlive, int connectionTimeout, int keepAliveTimeout)throws IOException {// check stateif (!parsingRequestLine) {return true;}//// Skipping blank lines//if (parsingRequestLinePhase < 2) {do {// Read new bytes if neededif (byteBuffer.position() >= byteBuffer.limit()) {if (keptAlive) {// Haven't read any request data yet so use the keep-alive// timeout.wrapper.setReadTimeout(keepAliveTimeout);}if (!fill(false)) {// A read is pending, so no longer in initial stateparsingRequestLinePhase = 1;return false;}// At least one byte of the request has been received.// Switch to the socket timeout.wrapper.setReadTimeout(connectionTimeout);}if (!keptAlive && byteBuffer.position() == 0 && byteBuffer.limit() >= CLIENT_PREFACE_START.length - 1) {boolean prefaceMatch = true;for (int i = 0; i < CLIENT_PREFACE_START.length && prefaceMatch; i++) {if (CLIENT_PREFACE_START[i] != byteBuffer.get(i)) {prefaceMatch = false;}}if (prefaceMatch) {// HTTP/2 preface matchedparsingRequestLinePhase = -1;return false;}}// Set the start time once we start reading data (even if it is// just skipping blank lines)if (request.getStartTime() < 0) {request.setStartTime(System.currentTimeMillis());}chr = byteBuffer.get();} while ((chr == Constants.CR) || (chr == Constants.LF));byteBuffer.position(byteBuffer.position() - 1);parsingRequestLineStart = byteBuffer.position();parsingRequestLinePhase = 2;}if (parsingRequestLinePhase == 2) {//// Reading the method name// Method name is a token//boolean space = false;while (!space) {// Read new bytes if neededif (byteBuffer.position() >= byteBuffer.limit()) {if (!fill(false)) // request line parsingreturn false;}// Spec says method name is a token followed by a single SP but// also be tolerant of multiple SP and/or HT.int pos = byteBuffer.position();chr = byteBuffer.get();if (chr == Constants.SP || chr == Constants.HT) {space = true;request.method().setBytes(byteBuffer.array(), parsingRequestLineStart,pos - parsingRequestLineStart);} else if (!HttpParser.isToken(chr)) {// Avoid unknown protocol triggering an additional errorrequest.protocol().setString(Constants.HTTP_11);String invalidMethodValue = parseInvalid(parsingRequestLineStart, byteBuffer);throw new IllegalArgumentException(sm.getString("iib.invalidmethod", invalidMethodValue));}}parsingRequestLinePhase = 3;}if (parsingRequestLinePhase == 3) {// Spec says single SP but also be tolerant of multiple SP and/or HTboolean space = true;while (space) {// Read new bytes if neededif (byteBuffer.position() >= byteBuffer.limit()) {if (!fill(false)) // request line parsingreturn false;}chr = byteBuffer.get();if (!(chr == Constants.SP || chr == Constants.HT)) {space = false;byteBuffer.position(byteBuffer.position() - 1);}}parsingRequestLineStart = byteBuffer.position();parsingRequestLinePhase = 4;}if (parsingRequestLinePhase == 4) {// Mark the current buffer positionint end = 0;//// Reading the URI//boolean space = false;while (!space) {// Read new bytes if neededif (byteBuffer.position() >= byteBuffer.limit()) {if (!fill(false)) // request line parsingreturn false;}int pos = byteBuffer.position();prevChr = chr;chr = byteBuffer.get();if (prevChr == Constants.CR && chr != Constants.LF) {// CR not followed by LF so not an HTTP/0.9 request and// therefore invalid. Trigger error handling.// Avoid unknown protocol triggering an additional errorrequest.protocol().setString(Constants.HTTP_11);String invalidRequestTarget = parseInvalid(parsingRequestLineStart, byteBuffer);throw new IllegalArgumentException(sm.getString("iib.invalidRequestTarget", invalidRequestTarget));}if (chr == Constants.SP || chr == Constants.HT) {space = true;end = pos;} else if (chr == Constants.CR) {// HTTP/0.9 style request. CR is optional. LF is not.} else if (chr == Constants.LF) {// HTTP/0.9 style request// Stop this processing loopspace = true;// Set blank protocol (indicates HTTP/0.9)request.protocol().setString("");// Skip the protocol processingparsingRequestLinePhase = 7;if (prevChr == Constants.CR) {end = pos - 1;} else {end = pos;}} else if (chr == Constants.QUESTION && parsingRequestLineQPos == -1) {parsingRequestLineQPos = pos;} else if (parsingRequestLineQPos != -1 && !httpParser.isQueryRelaxed(chr)) {// Avoid unknown protocol triggering an additional errorrequest.protocol().setString(Constants.HTTP_11);// %nn decoding will be checked at the point of decodingString invalidRequestTarget = parseInvalid(parsingRequestLineStart, byteBuffer);throw new IllegalArgumentException(sm.getString("iib.invalidRequestTarget", invalidRequestTarget));} else if (httpParser.isNotRequestTargetRelaxed(chr)) {// Avoid unknown protocol triggering an additional errorrequest.protocol().setString(Constants.HTTP_11);// This is a general check that aims to catch problems early// Detailed checking of each part of the request target will// happen in Http11Processor#prepareRequest()String invalidRequestTarget = parseInvalid(parsingRequestLineStart, byteBuffer);throw new IllegalArgumentException(sm.getString("iib.invalidRequestTarget", invalidRequestTarget));}}if (parsingRequestLineQPos >= 0) {request.queryString().setBytes(byteBuffer.array(), parsingRequestLineQPos + 1,end - parsingRequestLineQPos - 1);request.requestURI().setBytes(byteBuffer.array(), parsingRequestLineStart,parsingRequestLineQPos - parsingRequestLineStart);} else {request.requestURI().setBytes(byteBuffer.array(), parsingRequestLineStart,end - parsingRequestLineStart);}// HTTP/0.9 processing jumps to stage 7.// Don't want to overwrite that here.if (parsingRequestLinePhase == 4) {parsingRequestLinePhase = 5;}}if (parsingRequestLinePhase == 5) {// Spec says single SP but also be tolerant of multiple and/or HTboolean space = true;while (space) {// Read new bytes if neededif (byteBuffer.position() >= byteBuffer.limit()) {if (!fill(false)) // request line parsingreturn false;}byte chr = byteBuffer.get();if (!(chr == Constants.SP || chr == Constants.HT)) {space = false;byteBuffer.position(byteBuffer.position() - 1);}}parsingRequestLineStart = byteBuffer.position();parsingRequestLinePhase = 6;// Mark the current buffer positionend = 0;}if (parsingRequestLinePhase == 6) {//// Reading the protocol// Protocol is always "HTTP/" DIGIT "." DIGIT//while (!parsingRequestLineEol) {// Read new bytes if neededif (byteBuffer.position() >= byteBuffer.limit()) {if (!fill(false)) // request line parsingreturn false;}int pos = byteBuffer.position();prevChr = chr;chr = byteBuffer.get();if (chr == Constants.CR) {// Possible end of request line. Need LF next.} else if (prevChr == Constants.CR && chr == Constants.LF) {end = pos - 1;parsingRequestLineEol = true;} else if (!HttpParser.isHttpProtocol(chr)) {String invalidProtocol = parseInvalid(parsingRequestLineStart, byteBuffer);throw new IllegalArgumentException(sm.getString("iib.invalidHttpProtocol", invalidProtocol));}}if ((end - parsingRequestLineStart) > 0) {request.protocol().setBytes(byteBuffer.array(), parsingRequestLineStart,end - parsingRequestLineStart);parsingRequestLinePhase = 7;}// If no protocol is found, the ISE below will be triggered.}if (parsingRequestLinePhase == 7) {// Parsing is complete. Return and clean-up.parsingRequestLine = false;parsingRequestLinePhase = 0;parsingRequestLineEol = false;parsingRequestLineStart = 0;return true;}throw new IllegalStateException(sm.getString("iib.invalidPhase", Integer.valueOf(parsingRequestLinePhase)));}

Http11InputBuffer的parseRequestLine方法对于读不到请求数据时会设置SocketWrapperBase的readTimeout为keepAliveTimeout

小结

tomcat提供了maxKeepAliveRequests及keepAliveTimeout两个keepalive相关的参数，其中在SocketWrapperBase维护了keepAliveLeft及readTimeout属性，Http11Processor的service方法会在protocol的maxKeepAliveRequests大于0则执行socketWrapper.decrementKeepAlive()，若小于等于0则重置keepAlive为false；Http11Processor的prepareResponse方法在statusDropsConnection时会设置keepalive为false，对于为false的会判断是否已经有Connection: close的header，如果没有则给添加上。

Http11Processor的service方法在keepalive为true时会执行inputBuffer.parseRequestLine，对于读不到请求数据时会设置SocketWrapperBase的readTimeout为keepAliveTimeout(默认为null取的是socketProperties.getSoTimeout()的值)。

本文来自互联网用户投稿，该文观点仅代表作者本人，不代表本站立场。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如若转载，请注明出处：http://www.mzph.cn/news/195693.shtml

如若内容造成侵权/违法违规/事实不符，请联系多彩编程网进行投诉反馈email:809451989@qq.com，一经查实，立即删除！

相关文章

嘉里大荣物流与极智嘉再度携手，合作助力物流服务高效升级

嘉里大荣物流与极智嘉再度携手，合作助力物流服务高效升级

近日&#xff0c;全球仓储机器人引领者极智嘉(Geek)与3PL知名企业嘉里大荣物流联合宣布&#xff0c;双方再度携手&#xff0c;6周内共建全新自动化订单履行中心&#xff0c;赋能国际时尚运动品牌New Balance加速B2B和B2C订单交付&#xff0c;为其客户提供更高效便捷的物流服务。…
阅读更多...
江科大STM32学习笔记（上）

江科大STM32学习笔记（上）

STM32F103xx 前言外设篇GPIO输出GPIO位结构GPIO模式外设的GPIO配置查看实战1&#xff1a; 如何进行基本的GPIO输入输出 OLED显示屏及调试Keil的调试模式演示 EXTI外部中断NVIC基本结构EXTI结构代码实战2&#xff1a;如何使用中断和对射式红外传感器&#xff06;旋转编码器 TIM&…
阅读更多...
【flink番外篇】1、flink的23种常用算子介绍及详细示例（1）- map、flatmap和filter

【flink番外篇】1、flink的23种常用算子介绍及详细示例（1）- map、flatmap和filter

Flink 系列文章 1、Flink 专栏等系列综合文章链接 文章目录 Flink 系列文章一、Flink的23种算子说明及示例1、maven依赖2、java bean3、map4、flatmap5、Filter 本文主要介绍Flink 的3种常用的operator&#xff08;map、flatmap和filter&#xff09;及以具体可运行示例进行说明…
阅读更多...
Mysql的索引详解

Mysql的索引详解

1.索引的分类 1.按照功能来分&#xff0c;可以分为主键索引、唯一索引、普通索引、全文索引 2.按照索引字段个数来分&#xff0c;可以分为单列索引、联合索引 3.按照物理实现方式来分&#xff0c;可以聚簇索引、非聚簇索引 2.适合添加索引的场景 1.具有唯一性约束的字段。 2…
阅读更多...
医药行业：轻松学会超低温冰箱技能

医药行业：轻松学会超低温冰箱技能

超低温冰箱在医疗、科研和生物领域中扮演着至关重要的角色&#xff0c;用于存储和保护对温度极为敏感的样品和药品。 然而&#xff0c;由于这些冰箱内的温度波动可能导致样品的损坏&#xff0c;因此对超低温冰箱的监控变得至关重要。 客户案例 医疗研究机构 上海某医疗研究机…
阅读更多...
YOLOv7独家原创改进：创新自研CPMS注意力，多尺度通道注意力具+多尺度深度可分离卷积空间注意力，全面升级CBAM

YOLOv7独家原创改进：创新自研CPMS注意力，多尺度通道注意力具+多尺度深度可分离卷积空间注意力，全面升级CBAM

💡💡💡本文自研创新改进：自研CPMS, 多尺度通道注意力具+多尺度深度可分离卷积空间注意力,全面升级CBAM 1）作为注意力CPMS使用； 推荐指数：五星 CPMS | 亲测在多个数据集能够实现涨点，对标CBAM。 收录 YOLOv7原创自研 https://blog.csdn.net/m0_63774211/ca…
阅读更多...
外包干了4年，技术退步太明显了。。。。。

外包干了4年，技术退步太明显了。。。。。

先说一下自己的情况&#xff0c;本科生生&#xff0c;18年通过校招进入武汉某软件公司&#xff0c;干了接近4年的功能测试&#xff0c;今年国庆&#xff0c;感觉自己不能够在这样下去了&#xff0c;长时间呆在一个舒适的环境会让一个人堕落!而我已经在一个企业干了四年的功能测…
阅读更多...
Docker容器（一）概述

Docker容器（一）概述

一、虚拟化概述 1.1引⼊虚拟化技术的必要性 服务器只有5%的时间是在⼯作的&#xff1b;在其它时间服务器都处于“休眠”状态. 虚拟化前 每台主机⼀个操作系统; 软硬件紧密结合; 在同⼀个主机上运⾏多个应⽤程序通常会遭遇冲突; 系统的资源利⽤率低; 硬件成本⾼昂⽽且不够灵活…
阅读更多...
金蝶云星空单据体明细权限和表单插件操作事件的先后顺序

金蝶云星空单据体明细权限和表单插件操作事件的先后顺序

文章目录 金蝶云星空单据体明细权限和表单插件操作事件的先后顺序顺序说明结论 金蝶云星空单据体明细权限和表单插件操作事件的先后顺序 顺序说明 先分录菜单单击事件EntryBarItemClick 再验权 后表单操作执行事件BeforeDoOperation 结论 如果是需要鉴权通过才允许操作的逻辑…
阅读更多...
SELinux refpolicy详解（10）

SELinux refpolicy详解（10）

接前一篇文章：SELinux refpolicy详解（9） 三、refpolicy内容详解 上一回（上几回）讲解完了refpolicy源码根目录下的build.conf文件。本回继续讲解源码根目录下的其它文件。第三个来讲解在README文件中提到的INSTALL文件。 3. INSTALL 文件路径：refpolicy源码根目录/INST…
阅读更多...
万界星空科技仓库管理wms系统

万界星空科技仓库管理wms系统

企业在管理库存时&#xff0c;尤其是生产制造企业&#xff0c;使用传统方式比如纸笔、Excel 管理库存&#xff0c;由于工具和信息化存在局限&#xff0c;导致在管理库存时出现如下问题&#xff1a; 1、通过纸笔记录出入库申请&#xff0c;人为手动计算易出错&#xff0c;数据易…
阅读更多...
全球与中国HDPE管道市场：增长趋势、竞争格局与前景展望

全球与中国HDPE管道市场：增长趋势、竞争格局与前景展望

快速成长的人口、快速的经济成长和工业发展增加了对可靠供水系统的需求。工业需要为制造流程、冷却系统和卫生目的提供可靠的水供应。随着国家的发展&#xff0c;它们更加重视基础设施&#xff0c;包括供水系统&#xff0c;以支持工业成长。HDPE管道广泛应用于饮用水和灌溉的配…
阅读更多...
Mybatis简介

Mybatis简介

1、MyBatis历史 MyBatis最初是Apache的一个开源项目iBatis, 2010年6月这个项目由Apache Software Foundation迁移到了Google Code。随着开发团队转投Google Code旗下&#xff0c; iBatis3.x正式更名为MyBatis。代码于2013年11月迁移到Github。iBatis一词来源于“internet”和“…
阅读更多...
医院不良事件报告系统源码带鱼骨图分析

医院不良事件报告系统源码带鱼骨图分析

医院不良事件上报系统通过 “事前的人员知识培训管理和制度落地促进”、“事中的事件上报和跟进处理”、 以及 “事后的原因分析和工作持续优化”&#xff0c;结合预存上百套已正在使用的模板&#xff0c;帮助医院从对护理事件、药品事件、医疗器械事件、医院感染事件、输血事件…
阅读更多...
CentOS系列：【Linux】CentOS7操作系统安装nginx实战（多种方法，超详细）

CentOS系列：【Linux】CentOS7操作系统安装nginx实战（多种方法，超详细）

CentOS7操作系统安装nginx实战&#xff08;多种方法&#xff0c;超详细&#xff09; 一. 实验环境二. 使用yum安装nginx2.1 添加yum源2.1.1 使用官网提供的源地址&#xff08;方法一&#xff09;1. 找到官网的源2. 使用rpm -ivh 进行安装3. 安装完成之后查看源&#xff1a; 2.1…
阅读更多...
三十七、XA模式

三十七、XA模式

目录 一、XA模式原理 在XA模式中&#xff0c;一个事务通常被分为两个阶段&#xff1a; 二、Seata的XA模式 RM一阶段的工作&#xff1a; TC一阶段的工作&#xff1a; RM二阶段的工作&#xff1a; 1、优点&#xff1a; 2、缺点&#xff1a; 三、实现XA模式 1、修改yml文…
阅读更多...
数据结构实验任务六 ：基于 Dijsktra 算法的最短路径求解

数据结构实验任务六 ：基于 Dijsktra 算法的最短路径求解

本次代码为实验六:基于 Dijsktra 算法的最短路径求解实现。本实验的重点在于对于Dijsktra算法的理解。有关Dijsktra的资料可以参考有关博文&#xff1a; 图论&#xff1a;Dijkstra算法——最详细的分析&#xff0c;图文并茂&#xff0c;一次看懂&#xff01;-CSDN博客 以下附上…
阅读更多...
572 - Oil Deposits （UVA）

572 - Oil Deposits （UVA）

题目链接如下&#xff1a; Online Judge 这道题我一开始用连通集来做的&#xff0c;搜了些测试样例都能过&#xff0c;但不知道为什么就是WA。后来写了个DFS。本来想把连通集的写法改好了再一并记录下来的&#xff0c;但花了好几个钟头怎么都调不过&#xff0c;很无奈。 DFS…
阅读更多...
Vision Transformer

Vision Transformer

Vision Transformer详解-CSDN博客 视频&#xff1a;11.1 Vision Transformer(vit)网络详解_哔哩哔哩_bilibili Vision Transformer学习笔记_linear projection of flattened patches-CSDN博客 一、embedding 层 对于标准的Transformer模块&#xff0c;要求输入的是token (向量…
阅读更多...
win10使用copilot（尝试中）

win10使用copilot（尝试中）

一、 Microsoft account | Sign In or Create Your Account Today – Microsoft 一路next全部点好【1】 二、 查看当前win10的版本&#xff0c;cmd输入命令winver 三、 修改区域为美国 四、更新和安全 Reference 【1】完美&#xff5c;在 Win10 强行开启 Win11 的独有功能…
阅读更多...
最新文章

Copyright @ 2022~2023