《嵌入式网络协议引擎ETA加速网络应用程序处理.ppt》由会员分享，可在线阅读，更多相关《嵌入式网络协议引擎ETA加速网络应用程序处理.ppt（18页珍藏版）》请在三一文库上搜索。
1、,嵌入式网络协议引擎(ETA)加速网络应用程序处理,作者：林明峰,日期：2003-12,概要,今天的网络环境和网络操作系统 Linux 的tcp/ip stack 的性能 嵌入内核的tcp 加速模块 ETA的现实应用,今天的网络环境和操作系统,10G的以太网络带来什么,通用处理器必须每67奈秒处理一个数据包 超过通用处理器处理极限 协议占用太多时间,NP是唯一的解决方案吗?,Intel Ixp1200带来好处 网络处理器对软件系统结果带来的问题 灵活性和高性能的矛盾,系统中断 内存拷贝 缓冲区管理 系统呼叫 Tcp/ip协议 应用程序只能占用3%,谁占用的cpu的时间.,Linux 的tcp/
2、ip协议,内核的resource manager (ckrm),控制 diskIO 的资源分配 控制网络的IO分配 目前还不成熟 http:/ 固定一个cpu来专门处理协议 制定一个异步的通讯接口 新的socket库来利用ETA,ETA System Architecture,LAN,Storage,IPC,ETA Host Interface,IP Storage Driver,File System,Kernel Applications,User Socket Applications,Socket Proxy,Network stack Virtualized, asynchronou
3、s queuing and event handling Engine Architecture & platform integration,Network Fabric,Packet Processing Engine,Direct Transport Interface,ETA Packet Processing Engine,NIC,Application (Kernel or User),Adaptation Layer,DTI Event Queue,Anonymous Buffer Pool,Shared Host Memory,App Buffers,NIC,DTI Opera
4、tion Model,DTI operations: Connection requests (Connect, Listen, Bind, Accept, Close, ) Data transfer requests (Send, Receive) Misc. operations (Set/Get Options,),ETA Test Environment,2P SMP Profile,Processing requirements in multiple components TCP/IP is the largest single component, but is small c
5、ompared to total The copy overhead is required to support legacy (synchronous) socket semantics Interrupts and system calls are required in order to time-share the CPU resources,ETA Profile (1 host CPU + 1 PPE),Processing times are compressed Idle time represents CPU resource that is usable for applications Asynchronous queuing interface avoids copy overhead Interrupts avoided by not time-sharing CPU System calls avoided by ETA queuing model,ETA,2P SMP,Profile Comparisons,Normalized CPU Usage,ETA的应用,基于内核级别的ids程序 独立于操作系统的包获取 接管网卡,