1. What are the two main performance characteristics one looks for in server technology?
What are the main types of servers,and how do they differ from one another?

- Scalability ¶é� server ¢Í§¤Ø³ËÁ´¤ÇÒÁÊÒÁÒöŧ¡çãËéãª� scalable technology ¤Ø³ÊÒÁÒö᷹·Õ� server hardware ·ÕèÁÕÍÂÙèä´é´éÇ computer·ÕèÁÕ¢¹Ò´ãË­è¡ÇèÒáµèÁÕª¹Ô´à´ÕÂǡѹä´éËÅѧ¨Ò¡·Õè¤Ø³·Ó¡ÒÃà¤Å×è͹ÂéÒÂâ»Ãá¡Ã� áÅ�data files àÃÕºÃéÍÂáÅéǤس¡çÊÒÁÒö·Õè¨ÐºÃÔ¡ÒâéÍÁÙÅãËé¡Ñ� user ÁÒ¡¢Öé¹ÍÕ¡ä´é·Ñ¹·Õ
- Reliability äÁèÇèÒ¨Ðà»ç¹ server ËÃ×Í webserver ¨Ðµéͧ·Ó§Ò¹ä´éµÅÍ´äÁèÁÕ¡ÒÃàÊÕÂËÒÂâ´Â·Ó§Ò¹µÅÍ� 24 ªÁ.

Server áºè§ä´� 4 »ÃÐàÀ· ¤×�
1. HIGH-END INTEL-BASED PERSONAL COMPUTER à»ç¹ PC Ẻ¸ÃÃÁ´� «Öè§ÁÕ¤ÇÒÁàÃçÇÁÒ¡à¾ÃÒÐãª� intel microprocessors, RAM ¢¹Ò´ãË�, disk drive ·ÕèÁÕ¢¹Ò´ãË­� àÃÒäÁèÊÒÁÒö·Õè¨Ðãªéà¤Ã×èͧª¹Ô´¹Õé¡Ñ� Window 3.1,95,98 à¾ÃÒ�©Ð¹Ñ鹤س¨ÐµéͧµÔ´µÑé� server operating system (SOS) ·ÕèÍ͡ẺÁÒÊÓËÃѺ¡ÒÃãª� server ºÒ§¤ÃÑé§àÃÒàÃÕ¡ server operating system Çè� network operating system ÃÇÁ¶Ö§ Window NT Server, Novell Netware, IntranetWare áÅ� UNIX àÇÍÃìªÑè¹µèÒ§æ ·Õè·Ó§Ò¹º¹ intel processor
2.INTEL-BASED SUPERSERVERS à¤Ã×èÍ� Intel-based superserver ¹Ñ鹫Öè§ÁÕ¤ÇÒÁÊÒÁÒö㹡ÒûÃÐÁÇżÅÊÙ§¡ÇèÒà¤Ã×èÍ� superserver
3.RISC SUPERSERVER à»ç¹ workstation server ·Õèãªé RISC microprocessor ·ÕèÁÕ¤ÇÒÁÊÒÁÒö㹡Ò÷ӧҹ·ÕèÊÙ§¡Çè� intel processor ·ÕèÊÙ§·ÕèÊØ´
4.MAINFRAME SERVERS ¤ÇÒÁàÃçÇÊÙ§ÊØ´ IBM àÃÕÂ� mainframe ·ÕèÁÕ¢¹Ò´ãË­è·ÕèÊØ´ÇèÒ "enterprise servers"

2.Distinguish between symmetric multiprocessing,server clustering,load balancing,and geographical decentralization. Which of these improves scalability? Which of these improves reliability?

¥workstation superserver ÁÒ¡ÁÒÂ
- Server Clustering ¡ÒÃÃÇÁ¡ÅØèÁ¢Í§ server Server Clustering «Öè§ã¹ server ËÅÒÂ� µÑǨзӡÒöÙ� link à¢éÒ´éÇ¡Ñ� ÁÕ 2 à˵ؼÅ㹡Ò÷� server ãË� scalability 㹡ÒÃãªé server cluster ¹Ñé¹ÁÕ¤ÇÒÁÊÒÁÒö㹡ÒûÃÐÁÇżÅÊÙ§¡ÇèÒ server à¤Ã×èͧà´ÕÂÇ à˵ؼŷÕ� 2 ¤×� ¤ÇÒÁ¹èÒàª×èͶÖ�(reliability) ¶é� server µÑÇ˹Öè§ã¹ cluster à¡Ô´àÊÕ¡ç¨ÐäÁè¶Ù¡¢Ñ´¨Ñ§ËÇÐ㹡ÒúÃÔ¡ÒÃ�ÍÁÙÅ
- Load Balancing server clustering ãËéà¡Ô´»ÃÐ⪹� ¤×� «×éÍ server ÁÒËÅÒÂ� à¤Ã×èÍ�(ËÃ×Í cluster) áÅеèÍà¤Ã×èͧàËÅèÒ¹Õé´éÇ� load balancing router
- Geographical Decentralization ÍÕ¡·Ò§·Õè¨Ð·ÓãËé¡Ò÷ӧҹ㹠server à¾ÔèÁ¢Ö鹡ç¤×Í geographical decentralization ÈÙ¹Âì¡ÅÒ§¢Í§ server áµèÅ� à¤Ã×èͧã� server ·ÑèÇâÅ¡¡ç¤×Í replicated serverµÑÇËÅÑ¡¨Ò¡·Ñé§ËÁ´ÊÒÁÒöãªéä´é¨Ò¡ remote server áµèÅÐÍѹ ¡ÒÃáºè� geographical äÁèÁÕÁҵðҹ TCP/IP

3.Distinguish between servers aimed primarily at end users and servers aimed at network administrators.

�¹Ñ¡§Ò¹·Õè·Ó˹éÒ·Õèãªé¢éÍÁÙÅà¡ÕèÂǡѺ෤â¹âÅÂÕ㹡ÒõÅÒ´, ¡ÒÃà§Ô�, ¡ÒüÅÔ�, ¡ÒâÒ ËÃ×Í˹éÒ·ÕèÍÔè¹æ¢Í§ºÃÔ�� file server, client/server application server áÅ� server Í×è¹�
- ¨Ó¹Ç¹¢Í§ server ·ÕèàÃÒµéͧ¡ÒÃã� network administratior àÃÕ¡ÇèÒ administrative server àÁ×è� network ÁÕ¡ÒÃàµÔºâµ·Õè«Ñº«é͹¢Õé� network administrator ¨Ð¡ÅÒÂà»ç¹ administrative server administrator server ¨Ðãª� network technology ªèÇÂá¡éä� administrator ´éÇÂà·¤â¹âÅÂբͧ network ·Õè«Ñº«é͹ÂÔ觢Öé¹

4.How do directory servers help users? How do directory servers help administrators? What do X.500 and LDAP standardize? How are X.500 and LDAP different?

�
- ªèÇ administrator ¤Çº¤ØÁ end user 㹡ÒÃà�Ò¶Ö§·ÃѾÂÒ¡Ã
- X.500 à»ç¹¹Áҵðҹ¨Ò� OSI à»ç¹ÍØ»¡Ã³ì·ÕèãªéÂÒ¡áÅÐÁÕÃÒ¤Òᾧ
- LADP ä´é¶Ù¡¾Ñ²¹ÒÀÒÂãµéÁҵðҹ¢Í� X.500 ¤×� lightweigth directory access protocol(LDAP) 㹡ÒÃà¾ÔèÁ»ÃÐÊÔ·¸ÔÀÒ¾ãËéà˹×Í¡Çè� X.500, LDAP ¨ÐÃѺº� TCP/IP 㹻Ѩ¨ØºÑ¹ LDAP ä´éà�ÒÁÒ¡ÅÒÂà»ç¹·ÕèÃÙé¨Ñ¡«Ö觶١ÊÃéÒ§Áҵðҹâ´Â IETP
- ¤ÇÒÁᵡµèÒ� X.500 à»ç¹Áҵðҹ·ÑèÇä»ã¹¢Ñé¹ááÊÓËÃѺ directory access protocol Êèǹ LADP ãªé㹡ÒõͺÃѺ¡ÒÃÃѺÃÙé¤ÇÒÁµéͧ¡ÒÃÊÓËÃѺ directory access protocol ÍÂèÒ§ÅÐàÍÕ´

5.In network management, what should the network administrator be able to do from the network management console? Distinguish between network management programs, managed devices, network management agents, and MIBs. How do RMON probes differ from other managed devices?

¥network áÅÐ͹Ø�µÔãË� network administrator áÊ´§â¤Ã§ÊÃéÒ§¢Í§ÍØ»¡Ã³ì·Õèä¡ÅÍÍ¡ä»
- 㹡Ò÷ӧҹàÁ×èÍäáçµÒÁ·ÕèÁջѭËÒà¡ÕèÂǡѺ¡ÒÃä´éÃѺ�ÍÁÙŨҡÍØ»¡Ã³ìº¹ network Áѹ¨Ðà¡çº¢éÍÁÙÅàËÅèÒ¹ÕéäÇé㹡ÒèѴ¡Òðҹ�ÍÁÙÅ àÃÕ¡ÇèÒ management information base (MIB) «Öè§â¤Ã§ÊÃéÒ§°Ò¹�ÍÁÙÅ͹حҵÔãËé network management program ·Ó¡ÒÃÇÔà¤ÃÒÐËìÃٻẺã¹�ÍÁÙÅ ÍØ»¡Ã³ì·Õè¤Çº¤ØÁâ´� network management system ã¹ network network management agent ¨Ð·Ó¡ÒõԴµèÍÊ×èÍÊÒôéÇ network management program º¹¡ÒÃʹѺʹع¢Í§¡ÒèѴ¡ÒÃÍØ»¡Ã³ì
- RMON probe «Ö觷ӡÒÃÃǺÃÇÁ¢éÍÁÙÅà¡ÕèÂǡѺÊèǹ¢Í§network áÅÐÃÇÁ¶Ö§¡ÒÃà´Ô¹·Ò§¢Í§�ÍÁÙÅã¹·Ò§Ê¶ÔµÔ ¨Ò¡àÇÅÒ˹Öè§ä»ÍÕ¡àÇÅÒ˹Öè§ network management program ä´é·Ó¡ÒÃÃǺÃÇÁ�ÍÁÙÅàËÅèÒ¹ÕéÁÒ¨Ò� RMON probe ÂѧÊÒÁÒö·Õè¨ÐÊè§á¼§ network à»ç¹ÊÑ�Ò³�ÍÁÙÅ àÃÕ¡ÇèÒ traps ¶éÒà¡Ô´à˵ءÒóìÍѹµÃÒ�

6.What are the two major standards for network management? What are comprehensive network management programs, and why do we need them?

1.SNMP (Simple Network Management Protocol) was created by the IETF as past of the overall TCP/IP standarda. The most popular network management protocol. The newest vertion, SNMP v3, add a number of security enhancements.
2.CMIP/CMIS (Common Management Information Prpotocol/Common Management Information Services) OSI has its own network management protocol. Even the abbreviation is complex, reflecting the complexity of CMIP/CMIS. These OSI protocol also possess a high degree of sophistication. In the future if SNMP does not deliver the functionality that network administrators need in a timely manner.
- Comprehensive network management programs that can collect data using multiple network protocols and integrate this information

7.What are the three layers of functionality for network management? Briefly characterize each layer.

- 3 Layer ã¹ network management
1.Transmission management :»ÃСͺ´éÇ hubs, switches, routers, modems, and carrier transmission lines.
2.Systems management : moves us up to the management of individual servers and client PCs.
3.Application management : is concerned with the management of distribute application, such as database application. »ÃСͺ´éÇ application access control and the collection of application use statistics across servers.

8.Characterize the following: authentication, encryption, privacy, confidentiality, message integrity, denial-of-service attack, intrusion detection, and security selfanalysis. Why is intrusion detection, which is often ignored, critical for security?

¥forms of authentication.
- Encryption : If someone does intercept messages, you would like to prevent that person from benefiting from this action.
- Privacy : is transform your messages into a form that will be unreadable by anyone intercepting the message. That person will see only an unreadable sequence of oncs and zeros.
- Confidentiality : Encryption gives privacy
- Message Integrity : controls ensure that if a message is altered, the receiving party will be able to detect the alteration.
- Denial-Of-Service Attack : to deluge a server with hundrens of thousands of computer-generated message. The sistem will be so overloaded that it will not be able to server its normal users. Other attacks use fewer message,each of which requires the system to take a long period of time to handle.
- Intrusion Detection : If a network lacks the ability to delect intruders, attackers will be able to take their time, trying many different thing until something works.
- Security self-analysis : essentially attack the system themselves, in order to detect security holes before hackers can find them.

9.Why is understanding the business the first step in developing a security policy? What are policy servers, and why are they important?

¥policies, one size does not fit all.
- Other security devices check with the policy server to implement their security protocols For instance, to prevent a hacker from finding a weak server to attack, every server might check with the policy server when it implements its access protocols.

10.Characterize the following: plaintext, ciphertext, encryption methodology, and encryption key. Why is it important for keys to be as large as possible?

�
- Ciphertext : Encryption converts this plaintext into ciphertext, which an interceptor will not be able to read. The receiver converts the ciphertext back to the original plaintext.
- Encryption Methodology : is to subtract N positions in the alphaber for each letter. So if we subtract two positions, we would encrypt the letter C by changing it to A.
- Encryption key : the key length is made quite long. A key of 56 bits provides minimal commercial security, whereas a key of 128 bits provides good security.
- Because there are only four possible keys, 00, 01, 10, and 11.

11.Distinguish between single key and public key encryption. Which reduces problems of key distribution?

�
- Public Key Encryption : each person has a public key. Instead of keeping. It secret, everyone makes his or her public key available to everyone
- This reduces the problem of key distribution. Athought private keys must still be distributed securely, there is only one secure key delivery per user, rather than two far each of a user's business partners.

12.In public key encryption, when A sends to B, what key does A use? When A authenticates himself or herself to B, what key does A use? If A and B send messages to each other, what keys do A and B use? If A encrypts a message to B, can A read it afterward?

�
- Sender's Public Key
- Although anyone can encrypt a message with a receiver's public key,the public key cannot decrypt the message afterward.

13.How are public key encryption and single key encryption usually combined? Why are they combined?

¥encryption. The key creator sends this single key encrypted with the other party's public key. The receiver decrypts the single key using the receiver's private key and the rest of the session uses single key encryption.

14.In public key authentication, how do you prove you are who you say you are? What are the steps in creating a message containing a digital signature? How does a digital signature differ from a digital certificate?

�
- Digital signature that the next step is to encrypt the message digest with the sender's private key.
- That a company attaches the digital certificate to every message it sends. The receiver, seeing the digital certificate authority has checked the company'identity and has assigned it a digital certificate. If the receiver is suspicious,it can verify the company's identity and other characteristics by checking with the certificate authority.

15.What is the function of a firewall? Distinguish between IP firewalls and application firewalls. What happens between the time that you send an HTTP request message and the time that you get the HTTP response message when you have an IP firewall? When you have an application layer firewall?

�
- IP firewalls provide some protection, a skilled hacker can easily defeat them.
- Application firewalls are also called proxy firewalls. To be a proxy means to act in place of someone else.

16.What are the levels of depth in filtering? What are the purposes of filtering? Explain what an integrated transmission network of the future is likely to look like.

�
- Purposes of Filtering »ÃСͺ´éÇ 4 ª¹Ô´
1.Effective, Efficient, and Reliable Delivery
2.Access Control
3.Quality of Service (QoS)
4.Translation
- Although companies want to have the "best of breed" in each type of device, such as firewalls and routers, an integrated transmission network would be far easier to manage.

�