基于Hash树热点窗口的存储器完整性校验方法  被引量:4

Memory Integrity Verifying Based on Hash Tree Hot-Window

在线阅读下载全文

作  者:侯方勇[1] 王志英[1] 刘真[1] 

机构地区:[1]国防科学技术大学计算机学院,长沙410073

出  处:《计算机学报》2004年第11期1471-1479,共9页Chinese Journal of Computers

基  金:国家自然科学基金 (90 10 40 2 5 );国家"八六三"高技术研究发展计划项目基金 (2 0 0 3AA14 2 0 80 )资助 .

摘  要:该文描述了一种存储器完整性校验的硬件方案 .首先 ,介绍了存储器完整性校验的情况 ;然后 ,提出了基于Hash树热点窗口的存储器完整性校验方法 .该方法将存储器访问的局部性特征与Hash树完整性校验原理相结合 ,通过缩短多数情况下校验所需路径的长度以及对节点实施恰当的缓冲 ,来有效减小完整性检验的代价 .该文描述了该方法的原理、算法与系统实现 ,并进行了模拟 .理论分析和模拟结果表明该方法在应用上是高性能的 ,同时也是低代价的 .This paper presents a hardware scheme to realize memory integrity verifying. At first, this paper introduces the general situation of memory integrity verification, points out its important meaning for computing security, gives an overview of main related techniques, emphasizes on analyzing the characteristics and the problems of these techniques. Then, this paper brings forward a new approach basing on hot-windows to optimize Hash tree to realize memory integrity verification. This approach integrates the principle of Hash tree based integrity checking with the local characteristic of memory accessing. Through shortening the length of needed checking path, and correctly buffering portion of internal nodes of Hash tree, this approach can reduce the cost of memory integrity verifying effectively, and achieves some distinct advantages. This paper elaborates the principle, algorithms and system realization of this new memory integrity verification hardware scheme, also gives some tests. Theoretical analysis explains that applying this hardware scheme to verify memory integrity is high performance, as well as low cost. The experimental simulation result shows that such a mechanism has less than 15% performance penalty for many cases, even can achieve 5% performance overhead with potential improvement.

关 键 词:存储器 完整性校验 HASH树 性能 校验方法 

分 类 号:TP302[自动化与计算机技术—计算机系统结构]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象