Date post: | 21-Dec-2015 |

Category: |
## Documents |

View: | 223 times |

Download: | 0 times |

Share this document with a friend

Popular Tags:

49

Concurrency in Distributed Systems: Mutual exclusion

Transcript

Concurrency in Distributed Systems:

Mutual exclusion

Shared Memory ModelChanges to the model from the MPS:

– Processors communicate via a set of shared variables, instead of passing messages

– Each shared variable has a type, defining a set of operations that can be performed atomically (i.e., instantaneously, without interferences)

– No inbuf and outbuf state components– Configuration includes a value for each

shared variable– The only event type is a computation step

by a processor

Shared Memory

1p

3p

2p

4p

processes

Types of Shared Variables1. Read/Write2. Read-Modify-Write3. Test & Set4. Fetch-and-add5. Compare-and-swap

.

.

.

We will focus on the Read/Write type (the simplest one to be realized)

Read/Write VariablesRead(v) Write(v,a)

return(v); v = a;

In one atomic step a processor can:– read the variable, or– write the variable… but not both!

1p write 10

1p write 1010

1p write 1010

2pread

1p write 1010

2pread

10

1p write 10 2pwrite 20

Simultaneous writes

1p write 10 2pwrite 20

Possibility 1

10

Simultaneous writes are scheduled:

1p write 10 2pwrite 20

Possibility 2

20

Simultaneous writes are scheduled:

1p write 2pwrite

In general:

1a2a

3pwrite 3a

kpwrite ka

xє{a1,…,ak}

Simultaneous writes are scheduled:

1p read 2pread

Simultaneous Reads: no problem!

aa a

All read the same value

A more powerful type:Read-Modify-Write Variables

RMW(v, f)temp = v;v = f(v);return (temp);

function on v

Atomic operation

Remark: a R/W type cannot simulate a RMW type, since reads and writes might interleave!

Computation Step in the Shared Memory Model

• When processor pi takes a step:– pi 's state in old configuration specifies

which shared variable is to be accessed and with which operation

– operation is done: shared variable's value in the new configuration changes according to the operation's semantics

– pi 's state in new configuration changes according to its old state and the result of the operation

Assumptions on the execution

–Asynchronous–Fault-free

• Each processor's code is divided into four sections:

– entry: synchronize with others to ensure mutually exclusive access to the …

– critical: use some resource; when done, enter the…

– exit: clean up; when done, enter the…– remainder: not interested in using the resource

Mutual Exclusion (Mutex) Problem

entry

critical

exit

remainder

Mutex Algorithms

• A mutual exclusion algorithm specifies code for entry and exit sections to ensure:– mutual exclusion: at most one

processor is in its critical section at any time, and

– some kind of liveness condition. There are three commonly considered ones…

Mutex Liveness Conditions

• no deadlock: if a processor is in its entry section at some time, then later some processor is in its critical section

• no lockout: if a processor is in its entry section at some time, then later the same processor is in its critical section

• bounded waiting: no lockout + while a processor is in its entry section, any other processor can enter the critical section no more than a certain number of times.

These conditions are increasingly strong.

Mutex Algorithms: assumptions

• The code for the entry and exit sections is allowed to assume that– no processor stays in its critical section

forever– shared variables used in the entry and

exit sections are not accessed during the critical and remainder sections

Complexity Measure for Mutex

• Main complexity measure of interest for shared memory mutex algorithms is amount of shared space needed.

• Space complexity is affected by:– how powerful is the type of the shared

variables– how strong is the progress property to

be satisfied (no deadlock vs. no lockout vs. bounded waiting)

Mutex Results Using R/W

number of distinct vars.

upper bound lower bound

no deadlock n

no lockout 3n booleans(tournament alg.)

bounded waiting

2n unbounded(bakery alg.)

Bakery Algorithm

• Guaranteeing:– Mutual exclusion– Bounded waiting

• Using 2n shared read/write variables– booleans Choosing[i]: initially false,

written by pi and read by others

– integers Number[i]: initially 0, written by pi and read by others

Bakery AlgorithmCode for entry section:

Choosing[i] := trueNumber[i] := max{Number[0],...,

Number[n-1]} + 1Choosing[i] := falsefor j := 0 to n-1 (except i) do

wait until Choosing[j] = falsewait until Number[j] = 0 or (Number[j],j) > (Number[i],i)

endfor

Code for exit section:Number[i] := 0

BA Provides Mutual Exclusion

Lemma 1: If pi is in the critical section and Number[k] ≠ 0 (k ≠ i), then (Number[k],k) > (Number[i],i).

Proof: Since pi is in the CS, it passed the second wait statement for j=k. There are two cases:

pi in CS andNumber[k] ≠ 0

pi 's most recentread of Number[k];Case 1: returns 0Case 2: returns (Number[k],k) > (Number[i],i)

Case 1

pi in CS andNumber[k] ≠ 0

pi's most recentread of Number[k],returns 0.So pk is in remainderor choosing number.

pi's most recentread of Choosing[k],returns false. So pk is not in the middleof choosing number.

pi's most recentwrite to Number[i]

So pk chooses number in this interval,sees pi's number, and chooses a larger one.

Case 2

pi in CS andNumber[k] ≠ 0

pi's most recentread of Number[k] returns (Number[k],k)>(Number[i],i).So pk has already taken its number.

So pk chooses chooses a number larger than pi in this interval, and does not change it until pi exits from the CSEND of PROOF

Mutual Exclusion for BA

• Lemma 2: If pi is in the critical section, then Number[i] > 0.– Proof is a straightforward induction.

• Mutual Exclusion: Suppose pi and pk are simultaneously in CS.– By Lemma 2, both have number > 0.– By Lemma 1,

•(Number[k],k) > (Number[i],i) and •(Number[i],i) > (Number[k],k)

No Lockout for BA• Assume in contradiction there is a starved

processor.• Starved processors are stuck at the wait

statements, not while choosing a number.• Let pi be a starved processor with smallest

(Number[i],i).• Any processor entering entry section after

pi has chosen its number, chooses a larger number, and therefore cannot overtake pi

• Every processor with a smaller number eventually enters CS (not starved) and exits.

• Thus pi cannot be stuck at the wait statements.

What about bounded waiting?

YES: It’s easy to see that any processor in the entry section can be overtaken at most once by any other processor (and so in total it can be overtaken at most n-1 times).

Space Complexity of BA

• Number of shared variables is 2n

· Choosing variables are booleans

· Number variables are unbounded: as long as the CS is occupied and some processor enters the entry section, the ticket number increases

• Is it possible for an algorithm to use less shared space?

Bounded 2-Processor ME Algorithm with ND

• Start with a bounded algorithm for 2 processors with ND, then extend to NL, then extend to n processors.

• Some ideas used in 2-processor algorithm:

– each processor has a shared boolean W[i] indicating if it wants the CS

– p0 always has priority over p1 ; asymmetric code

Bounded 2-Processor ME Algorithm with ND

Code for p0 's entry section:1 .2 .3 W[0] := 14 .5 .6 wait until W[1] = 0

Code for p0 's exit section:7 .8 W[0] := 0

Bounded 2-Processor ME Algorithm with ND

Code for p1 's entry section:1 W[1] := 02 wait until W[0] = 03 W[1] := 14 .5 if (W[0] = 1) then goto Line 16 .

Code for p1 's exit section:7 .8 W[1] := 0

Discussion of 2-Processor ND Algorithm

• Satisfies mutual exclusion: processors use W variables to make sure of this

• Satisfies no deadlock

• But unfair w.r.t. p1 (lockout)

• Fix by having the processors alternate in having the priority

Bounded 2-Processor ME Algorithm with NL

Uses 3 binary shared read/write variables:

• W[0]: initially 0, written by p0 and read by p1

• W[1]: initially 0, written by p1 and read by p0

• Priority: initially 0, written and read by both

Bounded 2-Processor ME Algorithm with NL

Code for pi’s entry section:1 W[i] := 02 wait until W[1-i] = 0 or Priority = i3 W[i] := 14 if (Priority = 1-i) then5 if (W[1-i] = 1) then goto Line 16 else wait until (W[1-i] = 0)

Code for pi’s exit section:7 Priority := 1-i8 W[i] := 0

Analysis: ME

Mutual Exclusion: Suppose in contradiction p0 and p1 are simultaneously in CS.

W[0]=W[1]=1,both procs in CS

p1 's mostrecent writeof 1 to W[1](Line 3)

p0 's mostrecent writeof 1 to W[0](Line 3)

p0 's most recentread of W[1] (Line 6):must return 1, not 0!

Analysis: No-Deadlock

• Useful for showing no-lockout.• If one proc. ever enters remainder

forever, other one cannot be starved.– Ex: If p1 enters remainder forever, then

p0 will keep seeing W[1] = 0.

• So any deadlock would starve both procs. In the entry section

Analysis: No-Deadlock

• Suppose in contradiction there is deadlock.• W.l.o.g., suppose Priority gets stuck at 0

after both processors are stuck in their entry sections.

p0 and p1

stuck in entryPriority=0

p0 not stuckin Line 2, skipsline 5, stuck in Line 6 with W[0]=1waiting forW[1] to be 0

p0 seesW[1] = 0,enters CS

p1 skipsLine 6, stuckat Line 2 withW[1]=0, waitingfor W[0] to be 0

Analysis: No-Lockout• Suppose in contradiction p0 is starved.• Since there is no deadlock, p1 enters CS infinitely

often.• The first time p1 executes Line 7 in exit section

after p0 is stuck in entry, Priority gets stuck at 0.

p1 at Line 7;Priority=0forever after

p0 stuckin entry

p0 stuck atLine 6 withW[0]=1, waitingfor W[1] to be 0

p1 entersentry, getsstuck atLine 2, waitingfor W[0] to be 0

Bounded Waiting?

• NO: A processor, even if having priority, might be overtaken repeatedly (in principle, an unbounded number of times) when it is in between Line 2 and 3.

Bounded n-Processor ME Alg.

• Can we get a bounded space NL mutex algorithm for n>2 processors?

• Yes!• Based on the notion of a tournament tree:

complete binary tree with n-1 nodes– tree is conceptual only! does not represent

message passing channels

• A copy of the 2-proc. algorithm is associated with each tree node– includes separate copies of the 3 shared

variables

Tournament Tree

1

2 3

4 5 6 7

p0, p1 p2, p3 p4, p5 p6, p7

Tournament Tree ME Algorithm

• Each proc. begins entry section at a specific leaf (two procs per leaf)

• A proc. proceeds to next level in tree by winning the 2-proc. competition for current tree node:– on left side, play role of p0

– on right side, play role of p1

• When a proc. wins the 2-proc. algorithm associated with the tree root, it enters CS.

The code

More on Tournament Tree Alg.

• Code is recursive• pi begins at tree node 2k + i/2, playing

role of pi mod 2, where k = log n -1.• After winning node v, "critical section"

for node v is– entry code for all nodes on path from v/2

to root– real critical section– exit code for all nodes on path from root to

v/2

Tournament Tree Analysis• Correctness: based on correctness of 2-

processor algorithm and tournament structure:– ME for tournament alg. follows from ME for 2-

proc. alg. at tree root.– NL for tournament alg. follows from NL for the

2-proc. algs. at all nodes of tree

• Space Complexity: 3n boolean read/write shared variables.

• Bounded Waiting? No, as for the 2-processor algorithm.

Recommended