CN105989279A - Domino dedicated password cracking system based on multi-hierarchy GPU (Graphic Process Unit) cluster - Google Patents
Domino dedicated password cracking system based on multi-hierarchy GPU (Graphic Process Unit) cluster Download PDFInfo
- Publication number
- CN105989279A CN105989279A CN201510078276.6A CN201510078276A CN105989279A CN 105989279 A CN105989279 A CN 105989279A CN 201510078276 A CN201510078276 A CN 201510078276A CN 105989279 A CN105989279 A CN 105989279A
- Authority
- CN
- China
- Prior art keywords
- domino
- gpu
- node
- password
- server
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/45—Structures or tools for the administration of authentication
- G06F21/46—Structures or tools for the administration of authentication by designing passwords or checking the strength of passwords
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
Abstract
The invention discloses a Domino dedicated password cracking system based on a multi-hierarchy GPU (Graphic Process Unit) cluster. The Domino dedicated password cracking system comprises a plurality of server nodes of different hierarchies and a great quantity of GPU computational nodes of different hierarchies, wherein the server nodes are used for matters including password space division, task transmission, result summary and the like; the computational nodes are used for carrying out the GPU cracking of a Domino algorithm, the cracking process of the Domino algorithm is that all passwords distributed to the computational nodes are independently subjected to Domino computation with salt (S) in a Domino value, and a cipher is considered to be successfully cracked if the value obtained by computation is the same with a hash value (H) in the Domino value; and if the value obtained by computation is different from the hash value (H), the password is replaced to continuously try until all passwords finish being computed and compared. The domino dedicated password cracking system can stably operate and perform high performance, the overall cracking speed can achieve about four hundred million passwords/second, a decryption task submitted by a user can be reasonably decomposed to each GPU computational node to be computed, and distribution is reasonable and balanced.
Description
Technical field
The present invention relates to code-breaking techniques field, particularly relate to a kind of Domino specific password based on multi-layer GPU cluster and crack system.
Background technology
Lotus/notes is a kind of conventional email server software system, and because of its perfect in shape and function, safety is good and is used in a large number.Therefore this mailing system also becomes an important goal of investigation evidence obtaining.
For protecting the algorithm entitled Domino algorithm of user password in Lotus/notes; this is a kind of special salt adding hashing algorithm; having higher safety, user password, after this algorithm calculates, can generate a Domino value being made up of salt (S) and hashed value (H).This algorithm the most not yet finds open defect, Domino value is cracked (i.e. using a large amount of password to carry out attempting until finding real password) by the method that therefore can only rely on brute force attack, and along with the raising of people's safety consciousness, increasing user selects to use complicated password as password, and these present situations cause the amount of calculation of password cracking needs the biggest.
Domino algorithm is cracked many achievements in research, most notable software is John the ripper, on its support CPU, the Domino value to user password carries out attack based on the mode such as violence, dictionary, and the speed that cracks on monokaryon CPU was about about 150,000 password/seconds.
The development of the high-performance data disposal ability of image processor GPU (Graphic Process Unit), cracks at a high speed for Domino and provides new effective way.What another money was famous crack, and hashcat provides the support that the GPU to Domino algorithm calculates in the recent period, and its speed can reach about about 10,000,000 password/seconds of single deck tape-recorder.
To sum up, the support to Domino algorithm at present rests on unit CPU and cracks and unit GPU cracks node, cracks limited in one's ability and autgmentability is poor.
Summary of the invention
Crack for password cracking system of the prior art that speed is low, it is limited in one's ability to crack, autgmentability is poor technical problem, the invention discloses a kind of Domino specific password based on multi-layer GPU cluster and crack system.
To achieve these goals, technical scheme is as follows:
Domino specific password based on multi-layer GPU cluster cracks system, it is characterized in that, the system the superiors include a server node, its lower floor can connect several GPU and calculate node and child servers node, several GPU that can connect again lower floor below described child servers node calculate node and child servers node, form multi-layer system by that analogy；In the data plane of system, user submits Domino password cracking task on the superiors' server node to, then waits that lower level node carrys out request task, once receives request, is the most according to circumstances syncopated as a task piece and gives the node of the request of sending；Each GPU calculates node when not calculating task, upwards stratum server sends the request needing calculating task, this server is if not the superiors' server, then sent request by this server to its parent server, calculate node so on up to the GPU passing to the request that sends after getting task in top-level server step by step；Calculate the result of calculation reporting oneself after node completes to calculate step by step the server node being finally aggregated into the superiors, then calculate node and send the calculating of request beginning next task sheet, circulate until the calculating task of user is fully completed with this.
GPU calculates the computer system that node is a set of CPU+GPU associated working, wherein CPU is responsible for and server communication work, crack task including acquisition, submit task computation result to, be also responsible for GPU work simultaneously and carry out cracking data and prepare and result of calculation such as compiles at the work；GPU then purely is responsible for carrying out parallel computation, and all passwords specified all are calculated comparison one time.
On GPU, it has and processes core in a large number, and each core can check a password parallel, and the cut-and-try process of password is as follows: first carries out Domino value resolving conversion, obtains salt S and hashed value H；Salt S and trial password are carried out Domino and calculates one new hashed value of generation, finally this hashed value is compared with resolving hashed value H obtained before, if the same illustrating that the password attempted is exactly correct password, if differed, not being.
The present invention includes that multiple various level server node and a large amount of various level GPU calculate node, and described server node is for matters such as the division of password space, multiplexed transport, result collect；Described calculating node cracks for the GPU carrying out Domino algorithm, it cracks process is that with the salt S in Domino value, all passwords distributing to calculate node are carried out Domino calculating respectively, if the value calculated is identical with hashed value H in Domino value, then regard as password cracking success；If the value calculated differs with hashed value H, then change password and continue to attempt to, until all of password all calculates comparison and completes.
Present invention Domino based on multi-layer GPU cluster specific password cracks system, runs through long-time total system, and Domino password is conscientiously greatly improved cracks computing capability.In the test system built, GPU calculates the quantity of node and comprises nearly 40, system can stable operation play high-performance, totally crack speed and can reach about about 400,000,000 password/seconds；The task of decryption that user submits to, it is possible to rationally decompose and carry out calculating on each GPU calculating node, distribution rationally equilibrium.The most there is not other problems by long-play in whole system.
The feature of the present invention see graphic and the most preferably embodiment the detailed description of this case and obtains and be well understood to.
Accompanying drawing explanation
Fig. 1 is that present invention Domino based on multi-layer GPU cluster specific password cracks system integrated stand composition.
Fig. 2 is that Domino specific password based on multi-layer GPU cluster cracks grid topological diagram.
Fig. 3 is that Domino algorithm password calculates comparison flow chart.
Detailed description of the invention
For the technological means making the present invention realize, creation characteristic, reach purpose and be easy to understand with effect, the present invention is expanded on further below in conjunction with specific embodiment.
See Fig. 1-3, Domino specific password based on multi-layer GPU cluster cracks system, the system the superiors include a server node, its lower floor can connect several GPU and calculate node and child servers node, several GPU that can connect again lower floor below described child servers node calculate node and child servers node, form multi-layer system by that analogy；In the data plane of system, user submits Domino password cracking task on the superiors' server node to, then waits that lower level node carrys out request task, once receives request, is the most according to circumstances syncopated as a task piece and gives the node of the request of sending；Each GPU calculates node when not calculating task, upwards stratum server sends the request needing calculating task, this server is if not the superiors' server, then sent request by this server to its parent server, calculate node so on up to the GPU passing to the request that sends after getting task in top-level server step by step；Calculate the result of calculation reporting oneself after node completes to calculate step by step the server node being finally aggregated into the superiors, then calculate node and send the calculating of request beginning next task sheet, circulate until the calculating task of user is fully completed with this.
GPU calculates the computer system that node is a set of CPU+GPU associated working, wherein CPU is responsible for and server communication work, crack task including acquisition, submit task computation result to, be also responsible for GPU work simultaneously and carry out cracking data and prepare and result of calculation such as compiles at the work；GPU then purely is responsible for carrying out parallel computation, and all passwords specified all are calculated comparison one time.
On GPU, it has and processes core in a large number, and each core can check a password parallel, and the cut-and-try process of password is as follows: first carries out Domino value resolving conversion, obtains salt S and hashed value H；Salt S and trial password are carried out Domino and calculates one new hashed value of generation, finally this hashed value is compared with resolving hashed value H obtained before, if the same illustrating that the password attempted is exactly correct password, if differed, not being.
The ultimate principle of the present invention, principal character and advantages of the present invention have more than been shown and described.Skilled person will appreciate that of the industry; the present invention is not restricted to the described embodiments; the principle of the simply present invention described in above-described embodiment and description; the present invention also has various changes and modifications without departing from the spirit and scope of the present invention, and these changes and improvements both fall within the range of claimed invention.The protection domain of application claims is defined by appending claims and equivalent thereof.
Claims (3)
1. Domino specific password based on multi-layer GPU cluster cracks system, it is characterized in that, the system the superiors include a server node, its lower floor can connect several GPU and calculate node and child servers node, several GPU that can connect again lower floor below described child servers node calculate node and child servers node, form multi-layer system by that analogy；In the data plane of system, user submits Domino password cracking task on the superiors' server node to, then waits that lower level node carrys out request task, once receives request, is the most according to circumstances syncopated as a task piece and gives the node of the request of sending；Each GPU calculates node when not calculating task, upwards stratum server sends the request needing calculating task, this server is if not the superiors' server, then sent request by this server to its parent server, calculate node so on up to the GPU passing to the request that sends after getting task in top-level server step by step；Calculate the result of calculation reporting oneself after node completes to calculate step by step the server node being finally aggregated into the superiors, then calculate node and send the calculating of request beginning next task sheet, circulate until the calculating task of user is fully completed with this.
Domino specific password based on multi-layer GPU cluster the most according to claim 1 cracks system, it is characterized in that, described GPU calculates the computer system that node is a set of CPU+GPU associated working, wherein CPU is responsible for and server communication work, crack task including acquisition, submit task computation result to, be also responsible for GPU work simultaneously and carry out cracking data and prepare and result of calculation such as compiles at the work；GPU then purely is responsible for carrying out parallel computation, and all passwords specified all are calculated comparison one time.
Domino specific password based on multi-layer GPU cluster the most according to claim 1 cracks system, it is characterized in that, on GPU, it has and processes core in a large number, each core can check a password parallel, the cut-and-try process of password is as follows: first carries out Domino value resolving conversion, obtains salt S and hashed value H；Salt S and trial password are carried out Domino and calculates one new hashed value of generation, finally this hashed value is compared with resolving hashed value H obtained before, if the same illustrating that the password attempted is exactly correct password, if differed, not being.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510078276.6A CN105989279A (en) | 2015-02-13 | 2015-02-13 | Domino dedicated password cracking system based on multi-hierarchy GPU (Graphic Process Unit) cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510078276.6A CN105989279A (en) | 2015-02-13 | 2015-02-13 | Domino dedicated password cracking system based on multi-hierarchy GPU (Graphic Process Unit) cluster |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105989279A true CN105989279A (en) | 2016-10-05 |
Family
ID=57041291
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510078276.6A Pending CN105989279A (en) | 2015-02-13 | 2015-02-13 | Domino dedicated password cracking system based on multi-hierarchy GPU (Graphic Process Unit) cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105989279A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107566348A (en) * | 2017-08-11 | 2018-01-09 | 广州慧睿思通信息科技有限公司 | One kind is based on the distributed shifty decryption systems of FPGA and its method |
| CN113094696A (en) * | 2021-06-09 | 2021-07-09 | 中国电子信息产业集团有限公司第六研究所 | Password cracking effect evaluation method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103684754A (en) * | 2013-12-03 | 2014-03-26 | 中国电子科技集团公司第三十研究所 | WPA shared key cracking system based on GPU cluster |
| CN103716381A (en) * | 2013-12-12 | 2014-04-09 | 华为技术有限公司 | Control method and management node of distributed system |
| CN104283948A (en) * | 2014-09-26 | 2015-01-14 | 东软集团股份有限公司 | Server cluster system and load balancing implementation method thereof |
-
2015
- 2015-02-13 CN CN201510078276.6A patent/CN105989279A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103684754A (en) * | 2013-12-03 | 2014-03-26 | 中国电子科技集团公司第三十研究所 | WPA shared key cracking system based on GPU cluster |
| CN103716381A (en) * | 2013-12-12 | 2014-04-09 | 华为技术有限公司 | Control method and management node of distributed system |
| CN104283948A (en) * | 2014-09-26 | 2015-01-14 | 东软集团股份有限公司 | Server cluster system and load balancing implementation method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 姜峰: "基于分布式的GPU密码破译平台的研究与实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107566348A (en) * | 2017-08-11 | 2018-01-09 | 广州慧睿思通信息科技有限公司 | One kind is based on the distributed shifty decryption systems of FPGA and its method |
| CN113094696A (en) * | 2021-06-09 | 2021-07-09 | 中国电子信息产业集团有限公司第六研究所 | Password cracking effect evaluation method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhu et al. | The dynamic privacy-preserving mechanisms for online dynamic social networks | |
| CN111898137A (en) | Private data processing method, equipment and system for federated learning | |
| CN108595517A (en) | A kind of extensive document similarity detection method | |
| CN105224606A (en) | A kind of disposal route of user ID and device | |
| CN104077280A (en) | Community discovery parallelization method, community discovery parallelization system, host node equipment and computing node equipment | |
| WO2022142001A1 (en) | Target object evaluation method based on multi-score card fusion, and related device therefor | |
| Apostal et al. | Password recovery using MPI and CUDA | |
| JP6553816B2 (en) | User data sharing method and apparatus | |
| Zhang et al. | Reverse attack: Black-box attacks on collaborative recommendation | |
| CN105989279A (en) | Domino dedicated password cracking system based on multi-hierarchy GPU (Graphic Process Unit) cluster | |
| Ait-Saada et al. | How to leverage a multi-layered transformer language model for text clustering: an ensemble approach | |
| Bei et al. | Non-recursive cluster-scale graph interacted model for click-through rate prediction | |
| CN112925914A (en) | Data security classification method, system, device and storage medium | |
| CN108509531B (en) | Spark platform-based uncertain data set frequent item mining method | |
| Akter et al. | WeightGrad: Geo-distributed data analysis using quantization for faster convergence and better accuracy | |
| CN104077320A (en) | Method and device for generating to-be-published information | |
| Cai et al. | Association rule mining method based on weighted frequent pattern tree in mobile computing environment | |
| CN110223125B (en) | User position obtaining method under node position kernel-edge profit algorithm | |
| Hinge et al. | MuGDAD: Multilevel graph drawing algorithm in a distributed architecture | |
| CN108763400B (en) | Object dividing method and device based on object behaviors and theme preferences | |
| Brugger et al. | A custom computing system for finding similarties in complex networks | |
| CN112580070A (en) | RBAC role hierarchical auxiliary construction method based on prefix dictionary tree | |
| JP2017182377A (en) | Information processing device, control method, and program | |
| Chandra et al. | Modeling temporal and spatial features of collaboration network | |
| JP2020095437A (en) | Clustering device, clustering method, and clustering program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20170704 Address after: 200231, No. fifth, No. 600 Middle Road, Shanghai, Xuhui District Applicant after: SHANGHAI MODERN GENERAL RECOGNITION TECHNOLOGY CO., LTD. Address before: 201112 Shanghai Minhang District city Pujiang town LIANHANG Road No. 1588 Applicant before: Shanghai general identification technology research institute |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161005 |
|
| RJ01 | Rejection of invention patent application after publication |