首页> 外文期刊>Journal of systems architecture >A novel approach for implementing high-speed and long-distance networking protocols in a limited memory embedded kernel
【24h】

A novel approach for implementing high-speed and long-distance networking protocols in a limited memory embedded kernel

机译:在有限内存嵌入式内核中实现高速和长距离网络协议的新方法

获取原文
获取原文并翻译 | 示例
           

摘要

STREAMS kernel mechanisms are being used to implement networking protocols in limited memory embedded systems. The current approaches for STREAMS-based networking protocols implementation suffer from some shortcomings when these approaches are used to implement high-speed and long distance WAN protocols that require large transport windows. Accumulation of duplicate copies of large transport layer windows in the subnet layer cause unnecessary hogging of kernel memory resources. This memory hogging leads to performance degradation since shortage of buffers forces the protocols to operate at smaller window sizes. The reason for this overhead is that the protocols cannot be implemented efficiently due to strict layering scheme of STREAMS. We have devised new kernel mechanisms to provide solution to this problem. We have defined new mapping mechanisms between protocol layers and have coupled these mechanisms with novel "event-based' flow control mechanisms. These mechanisms provided appropriate flow control handling in the kernel that led to significant reduction in memory buffers hogging. This makes embedded systems handle large windows efficiently.
机译:STREAMS内核机制用于在有限内存嵌入式系统中实现网络协议。当这些方法用于实现需要大传输窗口的高速和长距离WAN协议时,当前基于STREAMS的网络协议实现方法存在一些缺陷。子网层中大型传输层窗口的重复副本的积累会导致不必要的占用内核内存资源。这种内存占用导致性能下降,因为缓冲区不足会迫使协议以较小的窗口大小运行。这种开销的原因是由于严格的STREAMS分层方案,协议无法有效实现。我们已经设计了新的内核机制来提供此问题的解决方案。我们在协议层之间定义了新的映射机制,并将这些机制与新颖的“基于事件”的流控制机制相结合。这些机制在内核中提供了适当的流控制处理,从而显着减少了内存缓冲区的占用。大窗户有效。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号