CN105138478B - A kind of memory integrity protection method of non-equilibrium Hash tree - Google Patents
A kind of memory integrity protection method of non-equilibrium Hash tree Download PDFInfo
- Publication number
- CN105138478B CN105138478B CN201510451102.XA CN201510451102A CN105138478B CN 105138478 B CN105138478 B CN 105138478B CN 201510451102 A CN201510451102 A CN 201510451102A CN 105138478 B CN105138478 B CN 105138478B
- Authority
- CN
- China
- Prior art keywords
- data
- block
- tree
- data block
- counter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Storage Device Security (AREA)
Abstract
The present invention relates to memory completeness check field, specially a kind of memory integrity protection method of non-equilibrium Hash tree.The present invention includes：(1) it initializes；(2) non-equilibrium binary tree is built；(3) write operation；(4) read operation.Under normal circumstances, verification cost will be less than the verification cost of common balanced binary tree to this method, even if in the worst cases, the performance of this method will not be higher than the verification cost of common balanced binary tree, but identical as its verification cost.The present invention shortens path length when data authentication on the whole.
Description
Technical field
The present invention relates to memory completeness check field, specially a kind of memory integrity of non-equilibrium Hash tree
Guard method.
Technical background
With the development of science and technology, computer application is more and more universal, since the data of computer disposal can be related to very much
Secret, thus it is guaranteed that computer handle these data safety at current research hot spot.The safety of data includes
Confidentiality and integrity.The present invention pertains only to how to ensure the integrality of data herein.Attacker can be to flowing in bus
Data cheated, recombinated, Replay Attack.Integrity protection seeks to guarantee to detect that attacker implements data
Malice tampering, such as hardware piggyback attack.Focus on protection information to be allowed to from Replay Attack.Replay Attack refers to attacking
The person of hitting replaces the data being previously stored in some address location present data.Replay Attack is taken precautions against at present mainly by making
With tree-like verification scheme.According to authentication unit use method with structure tree process difference, and can be divided into Merkle Tree,
Parallel check tree PAT tri- kinds of schemes of tree and TEC-Tree.
Merkle Tree are also known as Hash Tree, are a kind of tree mechanism being used for completeness check earliest.Existed by Merkle
It is proposed the effective computational problem being mainly used in public key cryptosyst within 1980.Memory content is used it for after Blum et al. modifications
In completeness check.It establishes one tree by being iterated Hash calculation to internal storage data block, on CPU preserve root node from
And may insure the integrality of data, it can especially resist Replay Attack.The disadvantage is that data block is unable to parallel computation when updating,
Therefore postpone bigger.Parallel check tree PAT is proposed for this disadvantage, when carrying out Hash calculation every time due to this method all
A corresponding random number, the calculating of higher node are not directly dependent on level node, and upper layer node is by by the random of child node
MAC is calculated after number connection, therefore parallel computation when it realizes data update.But Merkle Tree and PAT trees are merely able to
Ensure the integrality of data, i.e. their integrity protection and Confidentiality protection scheme is separated, is proposed for this disadvantage
TEC-Tree, it also ensures the machine of data by adding redundant data into data while guaranteeing data integrity
Close property.
A common feature for these three trees is that the tree that they are established all is balanced tree, the verification path of all data blocks
Length is all identical.Although caching Hash tree CHTree has been proposed, the section top node of Merkle Tree is stored in slow
In depositing, although this method can effectively shorten check length, the high data block low with access frequency of access frequency
It is still identical to verify path length, therefore delay is still very big when average apparently data block progress Hash calculation.
Invention content
The purpose of the present invention is to provide a kind of non-equilibrium Kazakhstan for shortening the average verification path length of leaf node in tree
The memory integrity protection method of uncommon tree.
The object of the present invention is achieved like this：
(1) it initializes
(1.1) memory is divided into the data block data_block[ of same size;i], there are one counters for each data block
counter[i]To record the read/write number to the data block；When initial, counter[i]=0, when processor reads and writes data block
data_block[i]When, Counter Value counter[i]=counter[i]+1；Block data structure uses 3 linked list types：
(1.2) a timer Time is set；After timer reaches the time of setting, a non-equilibrium y-bend is built
Tree, while the counter counter of all data block data_block is all reset, it counts again；
(2) non-equilibrium binary tree is built
(2.1) using the read/write number of n data block in memory as weights, then n weights { w is obtained1,w2,w3,w4,
w5,……wn, each weights constitute a tree T for there was only root nodei, then the set F={ T of n tree are obtained1,T2,T3,T4,
T5,……Tn}；
(2.2) two minimum tree T of weights are chosen from set FiAnd Tj, a new tree T is built as left and right child,
The weights of new tree T nodes are two child nodes weights sums；T is deleted from set FiAnd Tj, and tree T is added in set F；
(2.3) step (2.2) is repeated, until only remaining one tree in set F；
(3) write operation
(3.1) when CPU is to memory writing data blocks data_block[i]When, refresh counter counter[i]=
counter[i]+1；
(3.2) data block data_block[ is found;i]Corresponding leaf node pointer；
(3.3) data block corresponding to data block and sibling is connected, the Hash after data block connection is recalculated
Value hash updates the cryptographic Hash of father node, repeats (3.1)-(3.3) until root node；
(3.4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-
Hash tree is balanced, otherwise, write operation terminates；
(4) read operation
(4.1) as CPU rdma read data blocks data_block[i]When, update its counter counter[i]=
counter[i]+1；
(4.2) data block data_block[ is found;i]Corresponding leaf node pointer；
(4.3) data block corresponding to data block and sibling is connected, the Hash after data block connection is then calculated
Value hash is compared Hash result with the cryptographic Hash of father node, repeats this process until root node, if last calculate
Root node cryptographic Hash it is identical as the cryptographic Hash of the root node stored in CPU, then illustrate that data are correct, do not usurped
Change, CPU can use data；Conversely, then illustrating that data are tampered, and send out alarm；
(4.4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-
Hash tree is balanced, otherwise, read operation terminates.
The beneficial effects of the present invention are：
The present invention shortens path length when data authentication on the whole.
During building non-equilibrium binary tree, it may appear that 3 kinds of situations, the first is the general feelings enumerated in attached drawing 1
Condition, in this case, it has proved that, methods herein is smaller than the common balanced binary tree check length that is averaged.Except general feelings
Outside condition, two kinds of extreme cases are there is also during building non-equilibrium binary tree, are respectively：
(1) tree of structure is identical with original balanced binary tree.In this case their verification cost is identical.
(2) the non-equilibrium binary tree built is single branch binary tree.The verification cost of non-equilibrium binary tree in this case
It is less than the verification cost of common balanced binary tree.
In conclusion this method is under normal circumstances, verification cost will be less than the verification cost of common balanced binary tree,
Even if in the worst cases, the performance of this method will not be higher than common balanced binary tree verification cost, but with it
It is identical to verify cost.
Description of the drawings
Fig. 1 is balanced binary tree and non-equilibrium binary tree.
Fig. 2 is to build non-equilibrium binary tree step.
Fig. 3 is that data step is written into memory.
Fig. 4 is that data step is read from memory.
Specific implementation mode
The present invention is described further below in conjunction with the accompanying drawings.
The present invention relates to memory completeness check fields, are a kind of non-equilibrium Hash trees based on principle of locality
Memory integrity protection method.Replay Attack in order to prevent mostly uses greatly the integrality of tree construction protection data at present.At present
The tree used is all balance N-ary tree, and level of all leaf nodes in tree is all identical, i.e., the verification path of all data blocks
Length is identical.According to the principle of locality of program, certain data access frequencies are high whithin a period of time, certain some data is visited
Ask that frequency is low.Therefore the present invention is improved Hash tree, establishes a non-equilibrium binary tree, the section for keeping those access frequencys high
Point is close from root node, and the node for keeping those access frequencys low is remote from root node.The high Data block check path of access frequency in this way
It is short, the low Data block check path length of access frequency.Path is verified to shorten on the whole, reduces delay.
In order to shorten the average verification path length of leaf node in tree, the present invention proposes a kind of depositing for non-equilibrium Hash tree
Reservoir completeness protection method.According to the principle of locality of program, the access frequency of data whithin a period of time in memory is
Different, the access frequency of some data wants high, and the access frequency of some data wants low.Therefore, method proposed by the present invention
It is to establish a non-equilibrium binary tree, the very high data block of those nearest access frequencys is placed on the place close from root node, phase
Instead, those data blocks being not frequently used are placed on the place remote from root node.Shorten the average verification path length of data block.
(number wherein on leaf node is access times) as shown in Fig. 1.
The present invention is divided into memory the data block data_block[ of same size;i], and set for each data block
Set a counter counter[i], for recording the read/write number to data block.It is deposited while storing data block in memory
The pointer of corresponding leaf node is stored up, leaf node is the cryptographic Hash of data block.One timer Time is set, timer is worked as
After Time reaches setting time, non-equilibrium Hash tree is built using the cryptographic Hash of internal storage data block as leaf node, it is mobile every time
When leaf node, to change the pointer of corresponding leaf node, and retain the root node of the Hash tree in CPU, for complete
Property verification.
The concrete operations for building non-equilibrium Hash tree are as follows：
(1) using the read/write number of n data block in memory as weights, then n weights { w is obtained1,w2,w3,w4,
w5,……wn, each weights constitute a tree T for there was only root nodei, then the set F={ T of n tree are obtained1,T2,T3,T4,
T5,……Tn}。
(2) two minimum tree T of weights are chosen from set FiAnd Tj, built using them as left and right child one it is new
T is set, the new weights for setting T nodes are two child nodes weights sums.T is deleted from set FiAnd Tj, and set F is added in tree T
In.
(3) step (2) is repeated, until only remaining one tree in set F.
When processor into memory writing data blocks data_block[i]When, update the data the access times counter of block
[i]With entire Hash tree.Specific operation is as follows：
(1) when CPU is to memory writing data blocks data_block[i]When, update its counter counter[i]=
counter[i]+1。
(2) and then data block data_block[ is found;i]Corresponding leaf node pointer.
(3) data block corresponding to data block and sibling is connected, the cryptographic Hash after data block connection is recalculated
Hash updates the cryptographic Hash of father node, repeats this process until root node.
(4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-flat
Weigh Hash tree, and otherwise, write operation terminates.
When processor reads data block data_block[ from memory;i]When, update its counter counter[i], simultaneously
Integrity check is carried out to data.Specific operation is as follows：
(1) as CPU rdma read data blocks data_block[i]When, update its counter counter[i]=counter
[i]+1。
(2) and then data block data_block[ is found;i]Corresponding leaf node pointer.
(3) data block corresponding to data block and sibling is connected, the cryptographic Hash after data block connection is then calculated
Hash is compared Hash result with the cryptographic Hash of father node, repeats this process until root node, if finally calculate
The cryptographic Hash of root node is identical as the cryptographic Hash of the root node stored in CPU, then illustrates that data are correct, be not tampered with,
CPU can use data；Conversely, then illustrating that data are tampered, and send out alarm.
(4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-flat
Weigh Hash tree, and otherwise, read operation terminates.
Cum rights check length (the sum of the cum rights check lengths of all leaf nodes in tree) using tree compares as performance
Standard, indicated with WPL, i.e.,Wherein weight wiFor the read/write number of each leaf node, liFor
The check length of leaf node, i.e. path top set number.N is the number of node.
According to the read-write frequency of data block, rather than number is read and write to rebuild Hash tree.
It is first provided before introducing present disclosure defined below：
Weights：The read/write number of data block, is indicated with w.
The cum rights check length of node：The product weighed on check length and node between node to root vertex is used
WPL expressions, i.e. WPLi=wi*Ii。
The cum rights check length of tree：The sum of the cum rights check length of all leaf nodes, is indicated, i.e., with WPL in tree
Memory is divided into the data block data_block[ of same size;i], and it is arranged one for each data block
Counter counter[i], for recording the read/write number to data block.Storage is corresponding while storing data block in memory
Leaf node pointer, leaf node be data block cryptographic Hash.One timer Time is set, when timer Time is reached
After setting time, non-equilibrium Hash tree is built using the cryptographic Hash of internal storage data block as leaf node, moves leaf node every time
When, to change the pointer of corresponding leaf node, and retain the root node of the Hash tree in CPU, be used for completeness check.
When building non-equilibrium Hash tree, according to n given weights { w1,w2,w3,w4,w5,……wn, every time from centre
The data block for selecting two weights minimums, using them as the left and right subtree of binary tree, the weights of father node are weighed for child nodes
The sum of value, while the knot removal for the two weights minimum selected, this newly-generated binary tree is added as new tree
Enter into above-mentioned set.It repeats the above process, until there was only one tree.
Memory integrity checking method is realized by following procedure：
1) it initializes
(1) memory is divided into the data block data_block[ of same size;i], there are one counters for each data block
counter[i]To record the read/write number to the data block.When initial, counter[i]=0, when processor reads and writes data block
data_block[i]When, its Counter Value counter[i]=counter[i]+1.
(2) data structure uses 3 linked list types, as follows：
Node
{
long int counter；// read/write number, the weights as data block
long int hash；// cryptographic Hash
Struct node*lchild；// left child
Struct node*rchild；// right child
Struct node*parent；// father node
}node；
(3) a timer Time is set.After timer reaches the time of setting, a non-equilibrium binary tree is built,
The counter counter of all data block data_block is all reset simultaneously, them is made to count again.
2) non-equilibrium binary tree is built
The flow that processor builds non-equilibrium binary tree is as shown in Fig. 2, and concrete operations are as follows：
(1) using the read/write number of n data block in memory as weights, then n weights { w is obtained1,w2,w3,w4,
w5,……wn, each weights constitute a tree T for there was only root nodei, then the set F={ T of n tree are obtained1,T2,T3,T4,
T5,……Tn}。
(2) two minimum tree T of weights are chosen from set FiAnd Tj, built using them as left and right child one it is new
T is set, the new weights for setting T nodes are two child nodes weights sums.T is deleted from set FiAnd Tj, and set F is added in tree T
In.
(3) step (2) is repeated, until only remaining one tree in set F.
3) write operation
When processor into memory writing data blocks data_block[i]When, update the data the access times counter of block
[i]With entire Hash tree.The flow that processor carries out memory write operation is as shown in Fig. 3, and specific operation is as follows：
(1) when CPU is to memory writing data blocks data_block[i]When, update its counter counter[i]=
counter[i]+1。
(2) and then data block data_block[ is found;i]Corresponding leaf node pointer.
(3) data block corresponding to data block and sibling is connected, the cryptographic Hash after data block connection is recalculated
Hash updates the cryptographic Hash of father node, repeats this process until root node.
(4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-flat
Weigh Hash tree, and otherwise, write operation terminates.
4) read operation
When processor reads data block data_block[ from memory;i]When, update its counter counter[i], simultaneously
Integrity check is carried out to data.The flow that processor carries out memory read operation is as shown in Fig. 4, and specific operation is such as
Under：
(1) as CPU rdma read data blocks data_block[i]When, update its counter counter[i]=counter
[i]+1。
(2) and then data block data_block[ is found;i]Corresponding leaf node pointer.
(3) data block corresponding to data block and sibling is connected, the cryptographic Hash after data block connection is then calculated
Hash is compared Hash result with the cryptographic Hash of father node, repeats this process until root node, if finally calculate
The cryptographic Hash of root node is identical as the cryptographic Hash of the root node stored in CPU, then illustrates that data are correct, be not tampered with,
CPU can use data；Conversely, then illustrating that data are tampered, and send out alarm.
(4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-flat
Weigh Hash tree, and otherwise, read operation terminates.
Here is the specific implementation content of the present invention：
The content of present invention realizes algorithm content proposed by the invention using simulator SimpleScalar tool, below
The specific implementation pseudocode of each step algorithm is introduced respectively.
1, non-equilibrium binary tree is built
The flow that processor builds non-equilibrium binary tree is as shown in Fig. 2, and concrete operations are as follows：
(1) using the read/write number of n data block in memory as weights, then n weights { w is obtained1,w2,w3,w4,
w5,……wn, each weights constitute a tree T for there was only root nodei, then the set F={ T of n tree are obtained1,T2,T3,T4,
T5,……Tn}。
(2) two minimum tree T of weights are chosen from set FiAnd Tj, built using them as left and right child one it is new
T is set, the new weights for setting T nodes are two child nodes weights sums.T is deleted from set FiAnd Tj, set T and be added in set F.
(3) step (2) is repeated, until only remaining one tree in set F.
The specific implementation code of non-equilibrium binary tree is built as shown in algorithm 1.
2, data are written into memory
When processor into memory writing data blocks data_block[i]When, update the data the access times counter of block
[i]With entire Hash tree.The flow that processor carries out memory write operation is as shown in Fig. 3, and specific operation is as follows：
(1) when CPU is to memory writing data blocks data_block[i]When, update its counter counter[i]=
counter[i]+1。
(2) and then data block data_block[ is found;i]Corresponding leaf node pointer.
(3) data block corresponding to this data block and sibling is attached, recalculates data block and connects it
Cryptographic Hash hash afterwards updates the cryptographic Hash of father node, repeats this process until root node.
(4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-flat
Weigh Hash tree, and otherwise, write operation terminates.
Specific implementation process into memory when one data block of write-in is as shown in algorithm 2.
3, data are read from memory
When processor reads data block data_block[ from memory;i]When, update its counter counter[i], simultaneously
Integrity check is carried out to data.The flow that processor carries out memory read operation is as shown in Fig. 4, and specific operation is such as
Under：
(1) data block data_block[ is read from memory as CPU;i]When, update its counter counter[i]=
counter[i]+1。
(2) and then data block data_block[ is found;i]Corresponding leaf node pointer.
(3) data block corresponding to this data block and sibling is attached, then calculates data block and connects it
Cryptographic Hash hash afterwards is compared Hash result with the cryptographic Hash of father node, repeats this process until root node, such as
The cryptographic Hash of root node after fruit finally calculating is identical as the Hash result of the root node stored on CPU, then illustrates that data are just
True, it is not tampered with, CPU can use data；Conversely, then illustrating that data are tampered, and send out alarm.
(4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-flat
Weigh Hash tree, and otherwise, read operation terminates.
Specific implementation process from memory when one data block of reading is as shown in algorithm 3.
。
Claims (1)
1. a kind of memory integrity protection method of non-equilibrium Hash tree, which is characterized in that include the following steps：
1) it initializes
(1.1) memory is divided into the data block data_block[ of same size;i], there are one counters for each data block
counter[i]To record the read/write number to the data block；When initial, counter[i]=0, when processor reads and writes data block
data_block[i]When, its Counter Value counter[i]=counter[i]+1；
(1.2) data structure uses 3 linked list types, as follows：
Node
{
long int counter；// read/write number, the weights as data block
long int hash；// cryptographic Hash
Struct node*lchild；// left child
Struct node*rchild；// right child
Struct node*parent；// father node
}node；
(1.3) a timer Time is set；After timer reaches the time of setting, a non-equilibrium binary tree is built, together
When the counter counter of all data block data_block is all reset, so that them is counted again；
The present invention realizes proposed algorithm content using simulator SimpleScalar tool, introduces each separately below
The specific implementation pseudocode of step algorithm；
2) non-equilibrium binary tree is built
Processor builds the flow of non-equilibrium binary tree, and concrete operations are as follows：
(2.1) using the read/write number of n data block in memory as weights, then n weights { w is obtained1,w2,w3,w4,w5,……
wn, each weights constitute a tree T for there was only root nodei, then the set F={ T of n tree are obtained1,T2,T3,T4,T5,……
Tn}；
(2.2) two minimum tree T of weights are chosen from set FiAnd Tj, a new tree is built using them as left and right child
T, the new weights for setting T nodes are two child nodes weights sums；T is deleted from set FiAnd Tj, set T and be added in set F；
(2.3) step (2.2) is repeated, until only remaining one tree in set F；
The specific implementation code of non-equilibrium binary tree is built as shown in algorithm 1；
Algorithm 1：Build non-equilibrium binary tree algorithm
Input：Data block node data_block[n]
Output：Non-equilibrium binary tree T
Algorithm description：
3) data are written into memory
When processor into memory writing data blocks data_block[i]When, update the data the access times counter[ of block;i]
With entire Hash tree；Processor carries out memory the flow of write operation, and specific operation is as follows：
(3.1) when CPU is to memory writing data blocks data_block[i]When, update its counter counter[i]=
counter[i]+1；
(3.2) and then data block data_block[ is found;i]Corresponding leaf node pointer；
(3.3) data block corresponding to this data block and sibling is attached, recalculate data block connection after
Cryptographic Hash hash, update the cryptographic Hash of father node, repeat this process until root node；
(3.4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-equilibrium
Hash tree, otherwise, write operation terminate；
Specific implementation process into memory when one data block of write-in is as shown in algorithm 2；
Algorithm 2：Data block algorithm is write into memory
Input：Data block node data_block[n], leaf node array of pointers node[n], set F
Output：Binary tree T in F
Algorithm description：
4) data are read from memory
When processor reads data block data_block[ from memory;i]When, update its counter counter[i], while logarithm
According to progress integrity check；The detailed process that processor carries out memory read operation is as follows：
(4.1) data block data_block[ is read from memory as CPU;i]When, update its counter counter[i]=
counter[i]+1；
(4.2) and then data block data_block[ is found;i]Corresponding leaf node pointer；
(4.3) data block corresponding to this data block and sibling is attached, then after the connection of calculating data block
Cryptographic Hash hash, Hash result is compared with the cryptographic Hash of father node, repeat this process until root node, if
The cryptographic Hash of root node after finally calculating is identical as the Hash result of the root node stored on CPU, then illustrates that data are correct
, it is not tampered with, CPU can use data；Conversely, then illustrating that data are tampered, and send out alarm；
(4.4) it checks whether timer Time reaches setting time, if having arrived at setting time, re-establishes non-equilibrium
Hash tree, otherwise, read operation terminate；
Specific implementation process from memory when one data block of reading is as shown in algorithm 3；
Algorithm 3：Data block algorithm is read from memory
Input：Data block node data_block[n], leaf node array of pointers node[n], set F
Output：Binary tree T in F
Algorithm description：
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510451102.XA CN105138478B (en) | 2015-07-28 | 2015-07-28 | A kind of memory integrity protection method of non-equilibrium Hash tree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510451102.XA CN105138478B (en) | 2015-07-28 | 2015-07-28 | A kind of memory integrity protection method of non-equilibrium Hash tree |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105138478A CN105138478A (en) | 2015-12-09 |
| CN105138478B true CN105138478B (en) | 2018-10-26 |
Family
ID=54723831
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510451102.XA Active CN105138478B (en) | 2015-07-28 | 2015-07-28 | A kind of memory integrity protection method of non-equilibrium Hash tree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105138478B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3442160A1 (en) * | 2017-08-07 | 2019-02-13 | Siemens Aktiengesellschaft | Pruning of authentication trees |
| CN109428822B (en) * | 2017-09-01 | 2020-10-09 | 华为技术有限公司 | Name searching method and network equipment |
| US10866963B2 (en) * | 2017-12-28 | 2020-12-15 | Dropbox, Inc. | File system authentication |
| CN108595980B (en) * | 2018-05-02 | 2023-01-24 | 广州品唯软件有限公司 | Method and device for protecting commodity traceability information |
| CN109495446B (en) * | 2018-10-02 | 2020-12-22 | 复旦大学 | Order-preserving encryption algorithm based on balanced ordering tree storage structure |
| CN109445763B (en) * | 2018-10-25 | 2022-02-01 | 长沙新弘软件有限公司 | Unbalanced binary tree construction method based on binary boundary value calculation |
| CN111898164B (en) * | 2020-07-02 | 2024-03-29 | 武汉纺织大学 | Data integrity auditing method supporting label block chain storage and query |
| CN112651054B (en) * | 2020-12-30 | 2022-10-14 | 海光信息技术股份有限公司 | Memory data integrity protection method and device and electronic equipment |
| CN113111391B (en) * | 2021-04-09 | 2022-07-08 | 支付宝(杭州)信息技术有限公司 | Method for memory integrity protection and memory controller |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853190A (en) * | 2010-06-04 | 2010-10-06 | 华中科技大学 | Data integrity verification method suitable for embedded processor |
| CN101976322A (en) * | 2010-11-11 | 2011-02-16 | 清华大学 | Safety metadata management method based on integrality checking |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8578116B2 (en) * | 2006-03-28 | 2013-11-05 | Broadcom Corporation | System and method for memory data protection with secure pad memory |
-
2015
- 2015-07-28 CN CN201510451102.XA patent/CN105138478B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853190A (en) * | 2010-06-04 | 2010-10-06 | 华中科技大学 | Data integrity verification method suitable for embedded processor |
| CN101976322A (en) * | 2010-11-11 | 2011-02-16 | 清华大学 | Safety metadata management method based on integrality checking |
Non-Patent Citations (1)
| Title |
|---|
| 《非对称hash树存储器完整性保护方法》;马海峰等;《小型微型计算机系统》;20141031;第35卷(第10期);第2341-2346页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105138478A (en) | 2015-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105138478B (en) | A kind of memory integrity protection method of non-equilibrium Hash tree | |
| JP6804668B2 (en) | Block data validation method and equipment | |
| Stefanov et al. | Iris: A scalable cloud file system with efficient integrity checks | |
| Goodrich et al. | Invertible bloom lookup tables | |
| ES2837476T3 (en) | Methods and apparatus for a distributed database within a network | |
| Goodrich et al. | Oblivious RAM simulation with efficient worst-case access overhead | |
| JP2022110150A (en) | Blockchain implemented counting system and method for use in secure voting and distribution | |
| CN103136243B (en) | File system duplicate removal method based on cloud storage and device | |
| TW202037108A (en) | System and method for parallel-processing blockchain transactions | |
| CN105022968B (en) | A kind of integrity checking method of internal storage data | |
| JP2020513176A5 (en) | ||
| US10733061B2 (en) | Hybrid data storage system with private storage cloud and public storage cloud | |
| US20210049715A1 (en) | Blockchain-based data procesing method, apparatus, and electronic device | |
| TW202101248A (en) | Hierarchical storage method and device based on block chain and electronic equipment | |
| US9749418B2 (en) | Efficient dynamic proofs of retrievability | |
| JP2018018348A (en) | System for creating block chain, and program | |
| Huang et al. | Outsourced private information retrieval | |
| CN104778123A (en) | Method and device for detecting system performance | |
| CN106991148A (en) | It is a kind of to support the full database authentication system and method for updating operation | |
| Qiao et al. | Lightweight integrity auditing of edge data for distributed edge computing scenarios | |
| CN110419029A (en) | The method of data content is partially updated in distributed storage network | |
| CN106612274A (en) | Homogeneity-based shared data verification algorithm in cloud computing | |
| US9876638B2 (en) | System and method for compressed level-ordered edge sequence encoding | |
| CN112181790A (en) | Capacity statistical method and system of storage equipment and related components | |
| CN117111854A (en) | Data storage method, device and medium based on distributed encryption storage |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |