2 Network Time Protocol NTP 6518 8 The University of Sydney Page 29 Time 2

# 2 network time protocol ntp 6518 8 the university of

• 19

This preview shows page 7 - 11 out of 19 pages.

2 Network Time Protocol (NTP)
6/5/18 8 The University of Sydney Page 29 Time 2 Physical time : what time is it? 1 2 2 3 3 3 3 stratum 1 stratum 2 stratum 3 6 3 4 1 5 2 7 8 avg Cristian’s algorithm NTP uses the Cristian’s algorithm The Berkeley algorithm synchronize nodes relatively Logical time The University of Sydney Page 31 Time Logical clock Comparison counter Vector clock Comparison coordinates 3 Logical time : is it more recent, less recent or neither of them? a b p 1 VC 1 =(0,0) a b p 1 p 2 VC 1 =(0,0) (1,0) VC 2 =(0,0) (1,1) (1,0) (2,0) a b p i C i =0 1 2 b p i p j C i =0 1 2 C j =0 The University of Sydney Page 32 Multicast 3 Totally ordered broadcast algorithm (con’t) Initial clocks are 3, 2, 4 at resp. processes 1, 2, 3; messages are A, B, C Circles (resp. square) represent msg reception (resp. last ack reception) 7.3 represents clock 7 received from 3, max(local, received) is taken Messages are delivered in the order of their clocks, B, C, A, everywhere 3 2 4 A A A B B B C C C B B B A A A C C C 4 3 5 5 6 4 6 7 6.1->6 7 7.2->7 7.3->7 7.3->7 7.2->7 7.2->7 6.1->7 6.1->7 7.3->7
6/5/18 9 Consensus The University of Sydney Page 34 Consensus Consider a system in which each entity p i receives an input x initially a value from a well-ordered set of inputs outputs y , initially undefined ( ) but that can be assigned only once Solving the consensus problem aims at guaranteeing that: Termination : in every admissible execution, y i is eventually assigned a value for every non-faulty processor p i . Agreement : in every execution, if y i and y j are assigned, then y i = y j , for all non-faulty processors p i and p j . Validity : in every execution, if, for some value v, x i = v for all processes p i , and if y i is assigned for some non-faulty process p i , then y i = v. The value decided has been proposed . 3 Problem definition The University of Sydney Page 35 Consensus Assumptions Synchronous communication using unicast Crash failures A maximum of f processes can fail General result: There is no solution to the consensus problem in an asynchronous system even with a single failure 3 Fault-tolerant asynchronous solution Consistency
6/5/18 10 The University of Sydney Page 37 Serializability Serializability: the result of an execution (of multi-access operations) is serializable if there exists an equivalent sequential execution. 3 Serializability [Papadimitriou, JACM 1979] p 0 time p 1 Operation 1 Operation 2 p 2 Operation 3 R(x)0 R(y)2 W(x)1 R(x)1 W(y)2 The University of Sydney Page 38 Serializability Serializability: the result of an execution (of multi-access operations) is serializable if there exists an equivalent sequential execution. 3 Serializability p 0 time p 1 Operation 1 Operation 2 p 2 Operation 3 R(x)0 R(y)2 W(x)1 R(x)1 W(y)2 Can only occur before in an equivalent sequential execution There is a cycle op1- ->op2- ->op3- ->op1 in the precedence graph non-serializable The University of Sydney Page 39 Serializability Serializability: the result of an execution (of multi-access operations) is serializable if there exists an equivalent sequential execution.

#### You've reached the end of your free preview.

Want to read all 19 pages?

• '19
• routing protocol, Consistency model

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern