parallel simulation. past, present and future c.d. pham laboratoire resam universit ₫ claude...

Parallel Simulation.Past, Present and Future

C.D. PhamLaboratoire RESAM

Universit₫ Claude Bernard Lyon 1cpham@resam.univ-lyon1.fr

Past

Introduction Discrete Event Simulation (DES) Parallel DES and the synchronization

problems Chandy-Misra-Bryant rules

Architecture of a conservative LP The « Safe is better » approach The lookahead ability

Jeffersons point of view Architecture of an optimistic LP Time Warp

Mixed/adaptive approaches,

Present

Ongoing projects SSF, TeD/GTW, GloMoSim, CSAM.

Future

Challenges & Perspectives Ultra-large scale simulations, Wide-area federation-based simulations, WEB-based simulations.

PAST:the algorithms, only the algorithms!

Simulation To simulate is to reproduce the

behavior of a physical system with a model

Practically, computers are used to numerically simulate a logical model

Simulations are used for performance evaluation and prediction of complex systems fluids dynamic, chemistry reactions (continous) communication network models: routing,

congestion avoidance, mobile (discrete) Simulation is more flexible than

analytical methods

Discrete Event Simulation (DES)

assumption that a system changes its state at discrete points in simulation time

a1 a2 a3 a4d1 d2 d3

S1 S3

S2

0 Dt 2Dt 3Dt 4Dt 5Dt 6Dt

time step

DES concepts

fundamental concepts: system state (variables) state transitions (events) simulation time: totally ordered set of values

representing time in the system being modeled

the system state can only be modified upon reception of an event

modeling can be event-oriented process-oriented

Life cycle of a DES

a DES system can be viewed as a collec-tion of simulated objects and a sequence of event computations

each event computation contains a time stamp indicating when that event occurs in the physical system

each event computation may: modify state variables schedule new events into the simulated future

events are stored in a local event list events are processed in time stamped order usually, no more event = termination

A simple DES model

local event list

A B

5

link model delay = 5send processing time = 5

receive processing time = 1packet arrival

P1 at 5, P2 at 12, P3 at 22

<e4,15> B receive P1 from Ae4<e5,16> B sends ACK(P1) to Ae5

e8 <e8,23> B receive P2 from A

<e2,10> A sends P1 to B e2<e1,5> A receive packet P1 e1

<e6,17> A sends P2 to B e6

<e3,12> A receive packet P2 e3

<e9,22> A receive packet P3 e9

e7<e7,21> A receive ACK(P1)

Why it works?

events are processed in time stamp order

an event at time t can only generate future events with timestamp greater or equal to t (no event in the past)

generated events are put and sorted in the event list, according to their timestamp

the event with the smallest timestamp is always processed first,

causality constraints are implicitly maintained.

Why change? It s so simple!

models becomes larger and larger the simulation time is overwhelming

or the simulation is just untractable example:

parallel programs with millions of lines of codes, mobile networks with millions of mobile hosts, ATM networks with hundreds of complex

switches, multicast model with thousands of sources, ever-growing Internet, and much more...

Some figures to convince...

ATM network models Simulation at the cell-level, 200 switches 1000 traffic sources, 50Mbits/s 155Mbits/s links, 1 simulation event per cell arrival.

simulation time increases as link speed increases,

usually more than 1 event per cell arrival, how scalable is traditional simulation?

More than 26 billions events to simulate 1 second!30 hours if 1 event is processed in 1us

Parallel simulation - principles

execution of a discrete event simulation on a parallel or distributed system with several physical processors.

the simulation model is decomposed into several sub-models that can be executed in parallel spacial partitioning, temporel partitioning,

radically different from simple simulation replications.

Parallel simulation - pros & cons

pros reduction of the simulation time, increase of the model size,

cons causality constraints are difficult to maintain, need of special mechanisms to synchronize

the different processors, increase both the model and the simulation

kernel complexity. challenges

ease of use, transparency.

Parallel simulation - examplelogical process (LP)

packetheventt

parallel

A simple PDES model

local event list

A B

5

link model delay = 5send processing time = 5

receive processing time = 1packet arrival

P1 at 5, P2 at 12, P3 at 22

<e5,16> B sends ACK(P1)e5

<e2,10> A sends P1 to B e2

e6<e6,17> A sends P2 to B

<e1,5> A rec. packet P1 e1

<e3,12> A rec. packet P2 e3<e4,15> B rec. P1 from Ae4

<e8,23> B rec. P2 from Ae8e7<e3,21> A rec. ACK(P1)

t

e9<e9,22> A rec. packet P3

causality error, violation

Synchronization problems

fundamental concepts each Logical Process (LP) can be at a

different simulation time local causality constraints: events in each LP

must be executed in time stamp order synchronization algorithms

Conservative: avoids local causality violations by waiting until it s safe

Optimistic: allows local causality violations but provisions are done to recover from them at runtime

Chandy-Misra-Bryant rules

Architecture of a conservative LPThe « Safe is better » approachThe lookahead ability

Architecture of a conservative LP

LPs communicate by sending non-decreasing timestamped messages

each LP keeps a static FIFO channel for each LP with incoming communication

each FIFO channel (input channel, IC) has a clock ci that ticks according to the timestamp of the topmost message, if any, otherwise it keeps the timestamp of the last message

LPB LPA

LPC LPD

c1=tB1

tB1tB

2

tC3tC

4tC5

tD4

c2=tC3

c3=tD3

A simple conservative algorithm

each LP has to process event in time-stamp order to avoids local causality violations

The Chandy-Misra-Bryant algorithm

while (simulation is not over) { determine the ICi with the smallest Ci

if (ICi empty) wait for a message else { remove topmost event from ICi

process event }}

Safe but has to block

LPB LPA

LPC LPD

36

147

10

5

IC1

IC2

IC3

min IC event

12

31

42

53

BLOCK3

61

729

Blocks and even deadlocks!

S

A

B

M

merge point

BLOCKED

cycle

S sends allmessages to B

444 446

How to solve deadlock: null-messages

SA

B

M

null-messages for artificial propagation of simulation time

10 10

4410 445

67

12

10

UNBLOCKED

What frequency?

How to solve deadlock: null-messages

a null-message indicates a Lower Bound Time Stampminimum delay between links is 4LP C initially at simulation time 0

11 910 7A B C

4

LP C sends a null-message with time stamp 4

LP A sends a null-message with time stamp 8

8

LP B sends a null-message with time stamp 12

12

LP C can process event with time stamp 7

12

The lookahead ability

null-messages are sent by an LP to indicate a lower bound time stamp on the future messages that will be sent

null-messages rely on the « lookahead » ability communication link delays server processing time (FIFO)

lookahead is very application model dependant and need to be explicitly identified

Lookahead for concurrent processing

LPB

LPA

LPC

LPD

s

TA TA+LA

s s

s s

s s

s safe event

unsafe event

What if lookahead is small?a null-message indicates a Lower Bound Time Stamp

minimum delay between links is 4LP C initially at simulation time 0

11 910 7A B C

1

LP C sends a null-message with time stamp 1

LP A sends a null-message with time stamp 2

2

LP B sends a null-message with time stamp 3

3

LP C can process event with time stamp 7

7

1

then 5

5

then 6

6

then 7

7

Conservative: pros & cons

pros simple, easy to implement good performance when lookahead is large

(communication networks, FIFO queue) cons

pessimistic in many cases large lookahead is essential for performance no transparent exploitation of parallelism performances may drop even with small

changes in the model (adding preemption, adding one small lookahead link)

Jeffersons point of view

Architecture of an optimistic LPTime Warp

Architecture of an optimistic LP

LPs send timestamped messages, not necessarily in non-decreasing time stamp order

no static communication channels between LPs, dynamic creation of LPs is easy

each LP processes events as they are received, no need to wait for safe events

local causality violations are detected and corrected at runtime

LPB LPA

LPC LPD

tB1tB

2 tC3tC

4 tC5 tD

4

Processing events as they arrive

11

LPB

13

LPD

18

LPB

22

LPC

25

LPD

28

LPC

36

LPB

32

LPD

LPB

LPA

LPC

LPD

LPA

processed!

what to do with late messages?

Time Warp. Rollback? How?

Late messages are handled with a rollback mechanism undo false/uncorrect local computations,

state saving: save the state variables of an LP reverse computation

undo false/uncorrect remote computations, anti-messages: anti-messages and (real) messages

annihilate each other

process late messages re-process previous messages: processed

events are NOT discarded!

A pictured-view of a rollback

11131822252836

32

4345

25 13state points

anti-msg 13152024273038

1118222832

3438 30

36

The real rollback distance depends on the state saving period: short period reduces rollback overhead but increases state saving overhead

11131822252832364345 283236

unprocessed

processed

Reception of an anti-message

may initiate a rollback if the corresponding positive message has already been processed,

may annihilate the corresponding positive message if it is still unprocessed,

may wait in the input queue if the corresponding positive message has not been received yet.

222528364345

43

22252836434548

222528364345

25 rollback

48

Need for a Global Virtual Time

Motivations an indicator that the simulation time advances reclaim memory (fossil collection)

Basically, GVT is the minimum of all LPs  logical simulation time timestamp of all messages in transit

GVT garantees that events below GVT are definitive events (I/O) no rollback can occur before the GVT state points before GVT can be reclaimed anti-messages before GVT can be reclaimed

A pictured-view of the GVT

LPB

LPA

LPC

LPD

c

c

old GVT

c c c

cccc

c c c c

new GVT

c c cc

c

D

conditional event

definitive event

c c

c

c c

c

c

c

c

c

c c

D D

D

D

D

D

D

D

DWAN

TED

Optimistic overheads

Periodic state savings states may be large, very large! copies are very costly

Periodic GVT computations difficult in a distributed architecture, may block computations,

Rollback thrashing cascaded rollback, no simulation progress!

Memory! memory is THE limitation

Optimistic: pros & cons

pros exploits all the parallelism in the model,

lookahead is less important, transparent to the end-user, interactive simulations possible, can be general-purpose.

cons very complex, needs lots of memory, large overheads (state saving, GVT,

rollbacks)

Mixed/adaptive approaches

General framework that (automatically) switches to conservative or optimistic

Adaptive approaches may determine at runtime the amount of conservatism or optimism

conservative optimistic

mixed

messages

performance

optimistic

conservative

PRESENT:how to survive?

and how to get money?

Parallel simulation today

Lots of algorithms have been proposed variations on conservative and optimistic adaptives approaches

Paradoxically few end-users impossible to compete with sequential simulators

in terms of user interface, generability, ease of use...

Ongoing research mainly focus on ultra-large scale simulations of networks, tools and execution environments composability and interoperability issues

Ongoing projects

DOMAINS/GloMoSim SSF TeD/GTW CSAM

DOMAINS/GloMoSim project

Design of Mobile Adaptive Networks, DARPA/DAAB07-97-C-D321

Provides a library for simulating millions of mobile nodes

Proves the efficiency of parallel simulation for scalability issues

Based on the PARSEC simulation language

Conservative or optimistic execution

Glomo objectives

Glomo librairies

SSF-Scalable Simulation Framework

DARPA/ITO (Next Generation Internet Program) and NSF/ANIR (Special Projects in Networking)

SSF proposes discrete event simulations of large complex systems, with serial and scalable parallel implementations

SSFNet is a collection of SSF-based models for simulating Internet protocols and networks

Based on YAWNS, a conservative kernel

SSFNet, modeling the Internet

TeD/GTW

TeD (Telecommunications Description Language) is a language for modeling telecommunicating network elements and protocols (PNNI).

GTW is a general purpose parallel discrete event simulation executive using optimistic synchronization techniques.

The TeD compiler translates TeD models into C++ code which uses GTW for parallel simulation

Modeling with TeD

CSAM

CSAM: Conservative Simulator for ATM network Model

Simulation at the cell-level C++ programming-style, predefined

generic model of sources, switches, links

Test-bed for parallel simulations on high-performance clusters.

Test case: 78-switch ATM network

Distance-Vector Routing with dynamic link cost functionsConnection setup, admission control protocols

CSAM - Some results...

Routing protocols reconfiguration time

CSAM - visualization tool

FUTURE:the great challenges!

Ultra-Large scale simulations,Wide-area federation-based simulations, WEB-based simulations.

Ultra-large scale simulations

Millions of mobile nodes, Thousands of multicast connections, Full Internet simulation, Ultra-large scale simulations require

lots of memory! lots of CPU! new modeling techniques: reuse of model

description, decoupling state from model description;

advanced memory management schemes: shared events, application memory regulation.

« Out of core » simulations.

Federation-based simulations

Cost of model developping is increasing at a high rate.

Reuse and interoperability is a key issue in the development phase of new models.

Need for a unified framework so that independent simulators can run together and achieve a given goal.

The DoD has proposed the High-Level Architectureframework for federation-based simulations

The HLA framework

Runtime Infrastructure (RTI)Federation Management Declaration ManagementObject Management Ownership ManagementTime Management Data Distribution Management

The High Level Architecture calls for a federation of simulations to achieve interoperability and reuse of software.

Federation

Without HLA, simulations are mostly independents and interoperability is not easy.

Logical simulations

Hardware, human-in-the loopreal-time simulators

Display, statistics...

10 Rules for the federationand the federates behavior

An Object Model Template to describe the simulationobjects

An Interface Specification

simulator

real-timeplayers

tools

Wide-area interactive simulations

INTERNET

human in the loopflight simulator

battle field simulation

displaycomputer-basedsub-marine simulator

WEB-Based simulation

Users build models and submit them on the web (meta-computing) Hides the complexity of parallel simulation

techniques Provides computing resources for the users

ASCII RED1st ranktop 500 list

Cplant

?

JTeD project, the Java-based TeD

Summary

Parallel simulation is a mature field Applications, especially

communication network models, are the centre of interest

The challenges are for very-large scale simulations and re-usability of models

Real-time interactive simulations are desirable on a wide-area interconnection

As-fast-as possible simulations will likely remain « indoor » (cluster, SMP)

Requirements put on networking

In wide-area simulation, data distribution relies mainly on multicast and broadcast operations

Near real-time behaviors are desirable for interactive simulation

References

Parallel simulation K. M. Chandy and J. Misra, Distributed Simulation: A

Case Study in Design and Verification of Distributed Programs, IEEE Trans. on Soft. Eng., 1979, pp440-452

R. Fujimoto, Parallel Discrete Event Simulation, Comm. of the ACM, Vol. 33(10), Oct. 90, pp31-53

HLA http://hla.dmso.mil

Projects GlomoSim - http://pcl.cs.ucla.edu/projects/glomosim SSF - http://www.ssfnet.org/homePage.html TeD/GTW - CSAM - http://resam.univ-lyon1.fr/CSAM