面向OBS的缓存技术研究-计算机系统结构专业毕业论文.docxVIP

面向OBS的缓存技术研究-计算机系统结构专业毕业论文.docx

  1. 1、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。。
  2. 2、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载
  3. 3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
  4. 4、该文档为VIP文档,如果想要下载,成为VIP会员后,下载免费。
  5. 5、成为VIP后,下载本文档将扣除1次下载权益。下载后,不支持退款、换文档。如有疑问请联系我们
  6. 6、成为VIP后,您将拥有八大权益,权益包括:VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
  7. 7、VIP文档为合作方或网友上传,每下载1次, 网站将根据用户上传文档的质量评分、类型等,对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档
查看更多
II II Abstract? Storage systems have faced more and more challenges with the development of computer technique and networks, including higher I/O speech, larger storage capacity, security, expansibility and usability. The OBS (Object Based storage System) fits in with the next generation internet. The model of data access is object interface for OBS, there is an exclusive ID for every data object, constructing a flat three-level namespace. The OBS make up of computer node, object-based storage device(OSD), object-based file system, metadata server and network. The object-based file system (OBFS) is the major factor of affecting the performance with OBS. Making full use of the difference among the memory speed, network bandwidth, and disk bandwidth, building caches in the file system is one of the major methods to enhance the file system performance. In order to improve the performance of OBFS, a hierarchical cache system has been designed, including metadata cache level zero, cache level one in computer node, cache level two in OSD. Cache level zero is present at computer node. With strategy of improving metadata access protocol, the metadata cache reduces the delay of data accessing and one communication with metadata server once the cache hits. The cache has high hit-ratio, because of the metadata hold little space. The cache level one in computer node is implemented using shared memory. The application processes attaches the shared memory to its own memory space, then they can access the cache directly. The shared memory cache reduces the times of copying data between tow processes, and realizes data truly shared among the processes in a node. In order to improve the whole cache hit-ratio for cache level two, we design a replace algorithm called MQU. The algorithm classifies the blocks which often be accessed together as an access group, and replaces the blocks of an access group in bulk. MQU ??This research is supported by the National 973 Project Foundation (20

您可能关注的文档

文档评论(0)

jianzhongdahong + 关注
实名认证
文档贡献者

该用户很懒,什么也没介绍

1亿VIP精品文档

相关文档