Physical and Hardware Security 28 Physical security of processing computers and networks in a DB influences how security must be implemented. Without physical network security, the ability to perform encryption algorithms efficiently is critical. Without physical security of nodes, encryption can be used to keep data private, but data can be deleted maliciously Mullender, pp. It is a good idea to have tamperproof, tamper resistant, or tamper evident hardware.
Master Thesis Security in Distributed Databases- Ian Lee
Software Security Computers make access requests only under the control of the software programs. A program operates under the control of or in the name of a user. Thus, the accesses to a program are in all likelihood the result of requests from a user. However, programs can also be altered, that is, a program is actually a series of bits in memory, and those bits can be read, written, modified, and deleted as any data can be. Those modifications can be the result of hardware mistakes and failures, a lapse in the logic of the program, or a change influenced by some other program in the system.
Hardware errors do not happen often, 29 Unintentional or not, user errors are much more difficult to discover and prevent. Programs are stored either in files or main memory, or both. Thus, the first line of defense against program errors is memory protection, which is designed to prevent one person from deliberately or accidentally accessing files and memory assigned to another person. While these controls protect users from one another, they are far less efficient at safeguarding themselves from errors in their own program logic i. A second protection against program errors is careful and thorough software engineering, including structured and arrangement design, program analysis, and use of programming management professionals.
Such programming practices help to protect a user from unintentional errors. The third form of defense against program errors is software testing. Unfortunately, the best that testing can confirm is the presence of errors and not their absence. Nonetheless, a thoroughly tested program can give credibility to the assurance that a program is error- free. Of special interest is that the software that controls access of any subject to any object also protects itself against access by all unauthorized subjects. Nevertheless, the access control program itself represents a significant vulnerability: defeat or prevent the access control program, and the thieve can obtain unhindered access to all system services.
For this reason, on more secure computing systems, the access control function is separated among several different modules: one to control access to memory, another 30 In this way, defeating one module does not immediately open up all the DB systems resources to illegitimate uses. A related question is confirmation of the validity of the access control software itself, ensuring that it will permit all authorized users. Clearly, access control procedures are valid only if they are implemented properly. Good software engineering practices for the design and installation of the access control software are combined with explicit control over its modifications, once installed, to insure the correct and effective functioning of the access control software.
Also, if using software programs from vendors, the most updated versions should be used because they are more likely to be free of bugs. Data Security Maintaining the security of data such as a payroll file or a digitized graphical image requires consideration of the security of the entire computing DB system, including the internal data. However, computers do not understand domain names; they understand IP addresses a. IP addresses are then used to route packets around the DB network. When a computer is handed a domain name, it requests a DNS server to translate that domain name into an IP address.
Then it knows to which computer to send the packet of information.
Masters thesis powerpoint presentation
The problem with DNS system is that it lacks security. So when a computer sends a query to a DNS server and receives a response, it assumes that the response is correct and that the DNS server is honest. However, the DNS server may not be honest since hackers could have compromised it. And the reply that the computer gets from the DNS server might not have even come from the DNS server; it could have been a faked reply from an imposter. If the attackers make changes in the DNS tables the actual data that translates domains to IP addresses and vice versa , computers will automatically accept the validity of the modified tables.
Therefore, it is not difficult to imagine the kinds of computer invasions that could result8. Attackers are capable of doing all sorts of things. And DNS servers might have an automatic update procedure. So if the attacker can make a change at a few certain points, that change can propagate across the entire DB or the Internet. Cryptography is one important tool used to preserve both the confidentiality and the integrity of a DB Stein, p.
Confidential data are encrypted to prevent their disclosure to unauthorized people. One significant use of cryptography is to compute a cryptographic checksum, a function that depends upon the sum or other relationship upon every bit of a section of data and also upon the key used for the cryptographic function. For example, a weak cryptographic checksum is the parity of a string of bits; odd change to the string affects the parity. The cryptographic checksum is computed when the section of data is made and again when it is used; if the data has been changed between origin and use, the value of the checksum at time of use will certainly not match that computed at time of origin, a 33 It is recommended that users of cryptography change their encryption keys regularly.
If hackers steal or figure out the key, they can read the plaintext. Cryptography, while very powerful, is still subjected to security breaches. This reason is that cryptography is a branch of mathematics, which is logical. In the physical world, however, things can be very abstracted. Cryptography is based on hypotheses and theories.
However, in order to recognize the conclusions, the premises, the models, and the relationship between the theories and the reality, must be accepted. And that is sometimes very difficult to achieve.
- Guidelines for inclusion of a thesis in the Thesis Repository.
- Masters thesis powerpoint presentation;
- Master of Science in Informatics - Hellenic American University.
- Master Thesis In Database Management!
- blood in macbeth thesis?
- essays on madame bovary.
People do not always follow the rules. Sometimes they do just the opposite. Hardware and software can be the same way. They break down, misbehave, and fail. Also, it does not matter how good the cryptography is or what the key length is; weak passwords will help hackers to break into the DB. For example, hackers can use L0phtcrack, a NT password-auditing tool that can test an encrypted password against an 8-megabyte dictionary of popular passwords in seconds.
Some e-mails use Cryptography, which performs two valuable functions. It provides a digital signature for authenticity and encryption for privacy. In 34 The sender obtains the public key of the receiver. A bulk encryption key is generated, and then the sensitive data is encrypted with this key.
The message is now ready to be delivered. The receiver uses the private key in order to gain access to the bulk encryption key. The receiver then uses the bulk encryption key to return the document to uncover the plaintext Todd, , p. Digital Signatures Digital signature techniques are also used by e-mail. This helps to assure both the sender and the receiver that the message has not been tampered with during transmission. When the user indicates through the e-mail interface that the message should have a digital signature, the private key is used to mix the message and produce the message digest.
The document and the message digest are then sent to the receiver. The e-mail interface will indicate to the receiver that the message contains a digital signature. The document is divided up by the generation of a bit9 number of the receiver. If the decrypted digital signature matches the generated bit number, the receiver knows that the sender is really the person who is indicated on the message and that the body of the message has not been breached before the receiver has gotten it.
Mathematically, the ability to sign a message using public encryption depends on the fact that the encryption and decryption algorithms are the inverse of one another, that is: 9 If the key is n bits long, then there are 2n possible keys. So, if the key is bits long, there are trillions of possible keys. Of course, the physical world is more complicated. Just as one does not encrypt messages with public-key encryption algorithms one encrypts a message key , one also does not sign messages directly. Instead, this person takes a one-way hash of a message and then signs the hash.
Again, signing the hash is a few orders of magnitude faster, and there can be mathematical security problems with signing messages directly. Also, most digital signature algorithms do not actually encrypt the messages that are signed. This signature is affixed to the message.
The other end makes another calculation based on the message, the signature, and the public key to verify the signature. Even without the private key, the hacker can verify the signature. It is standard practice to encrypt passwords, but it is possible also to encrypt data and messages, and this encryption may well be desirable in a DDBMS environment.
In fact, the dominating issue is not the difficulty of breaking the code, but rather the security of the encryption keys Schneier, p. DES uses a bit key and the algorithm is available on an LSI chip that is capable of processing text at a rate of over one megabit per second Stein, p. The idea here is to assign two keys to each user: an encryption key for encrypting plaintext, and a decryption key for deciphering ciphertext.
It is nearly impossible to deduce the decryption key from the encryption key Stein, p. It has to act as a gatekeeper. It keeps intruders out and internal users in.
It has to figure out which bits are harmful and deny them entry. It has to do this without 37 Also once attackers bypass the firewall into the DBs, the firewall is no longer a safeguard.
CS Faculty and Their Research Interests
Since about 70 percent of all computer attacks come from the outside10 , firewalls are worth considering for most businesses. Hackers can use Trojan horses to penetrate firewalls, exploit some kind of bug in the DB that will open a connection between the hacker outside the firewall and the computer inside the firewall. If it all works, the hacker gets inside. Early firewalls were once known as packet filters. The firewall would look at each packet header coming in and decide whether to admit it or drop it, depending on a myriad of rules by the programs. Eventually firewall technology got better.
sdc.ascensiondental.com/caq-app-gratis.php Still, firewalls only have so long a memory inside, and slow and persistent attacks can often get through. For further protection, some companies have two firewall systems: one for the outside world, and another one with more restrictions against the insiders.
Related master thesis presentation on database management
Copyright 2019 - All Right Reserved