Simulation for Grid Computing - PowerPoint PPT Presentation


PPT – Simulation for Grid Computing PowerPoint presentation | free to view - id: 471e29-NDhhY


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation

Simulation for Grid Computing


Simulation for Grid Computing Henri Casanova Univ. of California, San Diego Grid Research Grid researchers often ask questions in the broad area ... – PowerPoint PPT presentation

Number of Views:19
Avg rating:3.0/5.0
Date added: 25 June 2020
Slides: 67
Provided by: HenriCa7


Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Simulation for Grid Computing

Simulation forGrid Computing
  • Henri Casanova
  • Univ. of California, San Diego

Grid Research
  • Grid researchers often ask questions in the broad
    area of distributed computing
  • Which scheduling algorithm is best for this
    application on a given Grid?
  • Which design is best for implementing a
    distributed Grid resource information service?
  • Which caching strategy is best for enabling a
    community of users that to distributed data
  • What are the properties of several resource
    management policies in terms of fairness and
  • What is the scalability of my publish-subscribe
    system under various levels of failures
  • ...

Grid Research
  • Analytical or Experimental?
  • Analytical Grid Computing research
  • Some have developed purely analytical /
    mathematical models for Grid computing
  • makes it possible to prove interesting theorems
  • often too simplistic to convince practitioners
  • but sometimes useful for understanding principles
    in spite of dubious applicability
  • One often uncovers NP-complete problems anyway
  • e.g., routing, partitioning, scheduling problems
  • And one must run experiments
  • Grid computing research based on experiments
  • Most published works

Grid Computing as Science?
  • You can tell you are a scientific discipline if
  • You can read a paper, easily reproduce (at least
    a subset of) its results, and improve
  • You can tell to a grad student Here are the
    standard tools, go learn how to use them and come
    back in one month
  • You can give a 1-hour seminar on widely accepted
    tools that are the basis for doing research in
    the area
  • We are not there today
  • But perhaps I can give a 1-hour seminar on
    emerging tools that could be the basis for doing
    research in the area, provided a few open
    questions are addressed
  • Need for standard ways to run Grid experiments

Grid Experiments
  • Real-world experiments are good
  • Eminently believable
  • Demonstrates that proposed approach can be
    implemented in practice
  • But...
  • Can be time-intensive
  • Execution of applications for hours, days,
    months, ...
  • Can be labor-intensive
  • Entire application needs to be built and
  • including all design / algorithms alternatives
  • include all hooks for deployment
  • Is it a bad engineering practice to build many
    full-fledge solutions to find out which ones work

Grid Experiments (2)
  • What experimental testbed?
  • My own little testbed
  • well-behaved, controlled, stable
  • often not representative of real Grids
  • Real grid platforms
  • (Still) challenging for many grid researchers to
  • Not built as a computer scientists playpen
  • other users may disrupt experiments
  • other users may find experiments disruptive
  • Platform will experience failures that may
    disrupt the experiments
  • Platform configuration may change drastically
    while experiments are being conducted
  • Experiments are uncontrolled and unrepeatable
  • even if disruption from other users is part of
    the experiments, it prevents back-to-back runs of
    competitor designs / algorithms

Grid Experiments (3)
  • Difficult to obtain statistically significant
    results on an appropriate testbed
  • And to make things worse...
  • Experiments are limited to the testbed
  • What part of the results are due to
    idiosyncrasies of the testbed?
  • Extrapolations are possible, but rarely
  • Must use a collection of testbeds...
  • Still limited explorations of what if scenarios
  • what if the network were different?
  • what if we were in 2015?
  • Difficult for others to reproduce results
  • This is the basis for scientific advances!

  • Simulation can solve many (all) of these
  • No need to build a real system
  • Conduct controlled / repeatable experiments
  • In principle, no limits to experimental scenarios
  • Possible for anybody to reproduce results
  • ...
  • Simulation
  • representation of the operation of one system
    (A) through the use of another (B)
  • Computer simulation B ? a computer program
  • Key question Validation
  • correspondence between simulation and real-world

Simulation in Computer Science
  • Microprocessor Design
  • A few standard cycle-accurate simulators are
    used extensively
  • http//
  • Possible to reproduce simulation results
  • Networking
  • A few standard packet-level simulators
  • NS-2, DaSSF, OMNeT
  • Well-known datasets for network topologies
  • Well-known generators of synthetic topologies
  • SSF standard http//
  • Possible to reproduce simulation results
  • Grid Computing?
  • None of the above up until a few years ago
  • Most people built their own ad-hoc solutions
  • Promising recent developments

Simulation of Distributed Computing Platforms?
  • Simulation of parallel platforms used for decades
  • Most simulations have made drastic assumptions
  • Simplistic platform model
  • Processors perform work at some fixed rate
  • Network links send data at some fixed rate
  • Topology is fully connected (no communication
    interference) or a bus (simple communication
  • Communication and computation are perfectly
  • Simplistic application model
  • All computation is CPU intensive
  • Clear-cut communication and computation phases
  • Application is deterministic
  • Straightforward simulation in most cases
  • Just fill in a Gantt chart with a computer rather
    than by hand
  • No need for a simulation standard

Grid Simulations?
  • Simple models
  • perhaps justifiable for a switched dedicated
    cluster running a matrix multiply application
  • Hardly justifiable for grid platforms
  • Complex and wide-reaching network topologies
  • multi-hop networks, heterogeneous bandwidths and
  • non-negligible latencies
  • complex bandwidth sharing behaviors
  • contention with other traffic
  • Overhead of middleware
  • Complex resource access/management policies
  • Interference of communication and computation

Grid Simulations
  • Recognized as a critical area
  • Grid eXplorer (GdX) project (INRIA)
  • Build an actual scientific instrument
  • Databases of experimental conditions
  • 1000-node cluster for simulation/emulation
  • Visualization tools
  • Still in planning
  • What simulation technology???
  • Two main questions
  • What does a representative Grid look like?
  • How does one do simulation on a synthetic
    representative Grid?

Presentation Outline
  • Introduction
  • Simulation for Grid Computing?
  • Generating Synthetic Grids
  • Simulating Applications on Synthetic Grids
  • Current Work and Future Directions

Why Synthetic Platforms?
  • Two goals of simulations
  • Simulate platforms beyond the ones at hand
  • Perform sensitivity analyses
  • Need Synthetic platforms
  • Examine real platforms
  • Discover principles
  • Implement platform generators
  • What Simulation results in my paper?
  • Results for a few real platforms
  • Results for many synthetic platforms

Generation of Synthetic Grids
  • Three main elements
  • Network Topology
  • Graph

network link
Generation of Synthetic Grids
  • Three main elements
  • Network Topology
  • Graph
  • Bandwidth and Latencies

Generation of Synthetic Grids
  • Three main elements
  • Network Topology
  • Graph
  • Bandwidth and Latencies
  • Compute Resources
  • And other resources

Generation of Synthetic Grids
  • Three main elements
  • Network Topology
  • Graph
  • Bandwidth and Latencies
  • Compute Resources
  • And other resources
  • Background Conditions
  • Load and unavailability

data store
Generation of Synthetic Grids
  • Three main elements
  • Network Topology
  • Graph
  • Bandwidth and Latencies
  • Compute Resources
  • And other resources
  • Background Conditions
  • Load and unavailability
  • Failures

data store
Generation of Synthetic Grids
  • Three main elements
  • Network Topology
  • Graph
  • Bandwidth and Latencies
  • Compute Resources
  • And other resources
  • Background Conditions
  • Load and unavailability
  • Failures
  • What is Representative and Tractable?

data store
Synthetic Network Topologies
  • The network community has wondered about the
    properties of the Internet topology for years
  • The Internet grows in a decentralized fashion
    with seemingly complex rules and incentives
  • Could it have a mathematical structure?
  • Could we then have generative models?
  • Three generations of graph generators
  • Random
  • Structural
  • Degree-based

Random Topology Generators
  • Simplistic
  • Place N nodes randomly in a square
  • Vertices u,v connected with prob. P(u,v) ?
  • Waxman JSAC88
  • Place N nodes randomly in a CxC square
  • P(u,v) ? e-d / (? C v2), 0 lt ?, ? 1
  • d Euclidian distance between u and v
  • First model to be widely used for network
  • Others
  • Exponential, Locality-based, ...
  • Shortcoming Real networks have a non-random
    structure with some hierarchy

Structural Generators
  • ... the primary structural characteristic
    affecting the paths between nodes in the Internet
    is the distribution between stub and transit
    domains... In other words, there is a hierarchy
    imposed on nodes... Zegura et al, 1997
  • Both at the AS level (peering relationships)
    and at the router level (Layer 2)
  • Quickly became accepted wisdom
  • Transit-Stub Calvert et al., 1997
  • Tiers Doar, 1996
  • GT-ITM Zegura et al., 1997

  • In 1999, Faloutsos et al. SIGCOMM99 rocked
    topology generation with power laws
  • Results obtained both for AS-level and
    router-level information from real networks
  • Outdegree (number of edges leaving a node)
  • For each node v, compute its outdegree dv
  • Node rank, rv the index of v in the order
    of decreasing degree
  • Nodes can have the same rank
  • Law dv proportional to rvR

  • Random Generators do not agree with
  • Structural Generators do not agree with Power-laws

GT-ITM, flat
GT-ITM, Transit-Stub
Degree-based Generators
  • New common wisdom A topology that does not agree
    with power-laws cannot be representative
  • Flurry of development of power-law generators
    after the Faloutsos paper
  • CMU power law graph generator Palmer et al.,
  • Inet Jin et al., 2000
  • BRITE Medina et al., 2001
  • PLRG Aiello et al., 2000

Structure vs. Power-law
  • We know network have structure AND power laws
  • Combine both?
  • GridG project Dinda et al., SC03
  • Use a Tiers-generated topology
  • Add random links to satisfy the power-laws
  • How about just using power-laws?
  • Comparison paper Tangmunarunkit et al.,
  • Degree-based generators capture the large-scale
    structure of real networks very well, better than
    structural generators!
  • structural generators impose too strict a
  • Hierarchy arises naturally from the degree-based
  • e.g., backbone links
  • Works both for AS-level and router-level

Short story
  • What generator?
  • To model large networks (e.g., gt 500 routers)
  • use degree-based generators
  • To model small networks (e.g., lt 100 routers)
  • use structural generators
  • degree-based will not create any hierarchy

Bandwidths, latencies, traffic, etc.
  • Topology generators only produce a graph
  • We need link characteristics as well
  • Option 1 Model physical characteristics
  • Some models in topology generators
  • Need to simulate background traffic
  • No accepted model for generating background
  • Simulation can be very costly
  • Option 2 Model end-to-end performance
  • Models (Lee, HCW01) or Measurements (NWS, ...)
  • Go from path modeling to link modeling?
  • Turns out to be a difficult question
  • DARPA workshop on network modeling

Bandwidths, latencies, traffic, etc.
  • Maybe none of this matters?
  • Fiber inside the network mostly unused
  • Communication bottleneck is the local link
  • Appropriate tuning of TCP or better protocols
    should saturate the local link
  • Dont care about topology at all!
  • Or maybe none of this matters for my application
  • No network contention

Compute Resources
  • What resources do we put at the end-points?
  • Option 1 ad-hoc generalization
  • Look at the TeraGrid
  • Generate new sites based on existing sites
  • Option 2 Statistical modeling
  • Examing many production resources
  • Identify key statistical characteristics
  • Come up with a generative/predictive model

Synthetic Clusters?
  • Many Grid resources are clusters
  • What is the typical distribution of clusters?
  • Commodity Cluster synthesizer Kee et al.,
  • Examined 114 production clusters (10K procs)
  • Came up with statistical models
  • Validated model against a set of 191 clusters
    (10K procs)
  • Models allow extrapolation for future
  • Models implemented in a resource generator

Architecture/Clock Models
Processor Fraction ()
Pentium2 1.4
Celeron 4.1
Pentium3 40.3
Pentium4 34.6
Itanium 3.9
Athlon 0.0
AthlonMP 12.4
AthlonXP 1.3
Opteron 2.0
  • Current distribution of proc families
  • Linear fit between clock-rate and release year
    within a processor family
  • Quadratic fraction of processors released on a
    given year
  • Model future distributions and speeds

Other models?
  • Other models
  • Cache size grows logarithmically
  • Number of processors per node log2-normal
  • Memory size log2-normal
  • Number of nodes per cluster log2-normal
  • Models were validated against a snapshot of ROCKS
  • These clusters have been added to the training
  • More clusters are added every month
  • GridG
  • Provide a generic framework in which such laws
    and other correlations can be encoded

Resource Availability / Workload
  • Probabilistic models
  • Naive exp. distributed availability and
    unavailability intervals
  • Sophisticated Weibull distributions Wolski et
  • Traces
  • NWS, etc.
  • Desktop Grid resources Kondo, SC04
  • Workload models
  • e.g., Batch schedulers
  • Traces
  • Models Feitelson, JPDC03
  • job inter-arrival times Gamma
  • amount of work requested Hyper-Gamma
  • number of processors requested Compounded (2,
    1, ...)

A Sample Synthetic Grid
  • Generate 5,000 routers with BRITE
  • Annotate latency according to BRITEs Euclidian
    distance method (scaling to obtain the desired
    network diameter)
  • Annotate bandwidth based on a set of end-to-end
    NWS measurements
  • Pick 30 of the end-points
  • Generate a cluster at each end-point according to
    Kees synthesizer for Year 2006
  • Model cluster load with Feitelsons model with a
    range of parameters for the random distributions
  • Model resource failures based on Inca
    measurements on TeraGrid

Synthetic Grid Generation
  • Still far from widely accepted standards
  • Many ongoing, promising efforts
  • Researchers have recognized this as an issue
  • Tools from networking can be reused
  • A few Grid tools are available
  • What is really needed
  • Repository of Grid Measurements
  • Repository of Synthetic/Generated Grids

Presentation Outline
  • Introduction
  • Simulation for Grid Computing?
  • Generating Synthetic Grids
  • Simulating Applications on Synthetic Grids
  • Current Work and Future Directions

Simulation Levels
  • Simulating applications on a synthetic platform?
  • Spectrum of simulation levels

Mathematical Simulation Discrete-event Simulation
more abstract
Based solely on equations
Abstraction of system as a set of dependent
actions and events (fine- or coarse-grain)
less abstract
Trapping and virtualization of low-level applicati
on/system actions
  • Boundaries above are blurred (d.e. simulation
  • A simulation can combine all paradigms at
    different levels

Simulation Options
  • Network
  • Macroscopic Flows in pipes
  • coarse-grain d.e. simulation mathematical
  • Microscopic Packet-level simulation
  • fine-grain d.e. simulation
  • Actual flows go through some network
  • emulation
  • CPU
  • Macroscopic Flows in a pipe
  • coarse-grain d.e. simulation mathematical
  • Microscopic Cycle-accurate simulation
  • fine-grain d.e. simulation
  • Virtualization via another CPU / Virtual Machine
  • emulation

more abstract
less abstract
Simulation Options
  • Application
  • Macroscopic Application analytical flow
  • Less Macroscopic sets of abstract tasks
    with resource needs and dependencies
  • coarse-grain d.e. simulation
  • Application specification or pseudo-code API
  • Virtualization
  • emulation of actual code with trapping of
    application-generated events
  • Two projects
  • MicroGrid (UCSD)
  • SimGrid (UCSD IMAG Univ. Nancy)

  • Set of simulation tools for evaluating
    middleware, applications, and network services
    for Grid systems
  • Applications are supported by emulation and
  • Actual application code is executed on
    virtualized resources
  • Microgrid accounts for
  • CPU
  • network
  • Virtualization

Virtual Resources
Physical Resources
MicroGrid Virtualization
  • Resource virtualization
  • resource names are virtualized
  • gethostname, sockets, Globus GIS, MDS, NWs, etc.
  • Time virtualization
  • Simulating the TeraGrid on a 4 node cluster
  • Simulating a 4 node cluster on the TeraGrid
  • CPU virtualization
  • Direct execution on (a fraction of) a physical
  • No application modification
  • Main challenge
  • Synchronization between real time and virtual

MicroGrid Network
  • Packet-level simulation
  • Network calls are intercepted
  • Are sent to a network simulator that has been
    configured with the virtual network topology
  • Implemented with MaSSF
  • An MPI version of DaSSF Liu et al., 2001
  • Configured via SSF standard (MDL, etc.)
  • Protocol stacks implemented on top of MaSSF
  • TCP, UDP, BGP, etc.
  • Socket calls are trapped without application
  • real data is sent
  • delays are scaled to match the virtual time
  • Main challenges
  • scaling (20,000 routers on a 128-node cluster)
  • synchronization with computation

MicroGrid in a NutShell
  • Virtualization via trapping of
  • application events
  • (emulation)
  • Can have high overhead
  • But captures the overhead!
  • Virtualization via another CPU
  • (emulation)
  • Can be really slow
  • But hopefully accurate
  • Microscopic Packet-level simulation
  • (fine-grain discrete event simulation)
  • Can be really slow for long transfers
  • But hopefully accurate

Emulation Discrete-event Mathematical
Simulation Simulation
more abstract
less abstract
  • Originally developed for scheduling research
  • Must be fast to allow for thousands of simulation
  • Application
  • No real application code is executed
  • Consists of tasks that have
  • dependencies
  • resource consumptions
  • Resources
  • No virtualization
  • A resource is defined by
  • a rate a which it does work
  • a fixed overhead that must be paid by each task
  • traces of the above if needed failures
  • A task can use multiple resources
  • data transfer over multiple links, computation
    that uses a disk and a CPU

  • Uses a combination of mathematical simulation and
    coarse-grain discrete event simulation
  • Simple API to specify an application rather
    than having it already implemented
  • Fast simulation
  • Key issue Resource sharing
  • In MicroGrid resource sharing emerges out of
    the low-level emulation and simulation
  • Packets of different connections interleaved by
  • CPU cycles of different processes get slices of
    the CPU
  • Drawback slow simulation
  • How can one do something faster that is still
  • Come up with macroscopic models of resource

Resource Sharing in SimGrid
  • Macroscopic resource sharing can be easy
  • A CPU CPU-bound processes/threads get a fair
    share of the CPU in steady state
  • Why go through the trouble of emulating CPU-bound
  • Just say how many cycles they need, and compute
    how many cycles they get per second
  • Macroscopic resource sharing can be not so easy
  • The Network
  • Many end-points, routers, and links
  • Many end-to-end TCP flows?
  • How much bandwidth does each flow receive?

Bandwidth Sharing
  • Macroscopic TCP modeling is a field
  • Fluid in Pipe analogy
  • Rule of Thumb Share of what a flow gets on its
    bottleneck link is inversely proportional to its
    Round-Trip Time
  • Turns out TCP in steady-state implements a type
    of resource sharing called Max-Min Fairness

Max-Min Fairness
  • Principle
  • Consider the set of all network links, L
  • cl is the capacity of link l
  • Considers the set of all flows, R
  • a flow a subset of L
  • xr is the bandwidth allocated to flow r
  • Bandwidth capacities are respected
  • ? l ? L, ? r ? R l ? r xr cl
  • TCP in steady-state is such that
  • minr ? R xr is maximized
  • The above can be solved efficiently (with
    appropriate data structures)

  • Uses the Max-Min fairness principle for all
    resource sharing
  • fast
  • validated in the real-world for CPUs
  • validated with NS-2 for networks
  • Limitation
  • Max-Min fairness is for steady-state
  • e.g., no TCP slow-start
  • e.g., no process priority boosts
  • Unclear when it starts breaking down
  • Is justified for long enough transfers and
  • reasonable for scientific applications
  • not so much for applications such as a Grid
    information service

SimGrid in a NutShell
  • Macroscopic Flows in a pipe
  • (mathematical simulation
  • coarse-grain d.e. simulation)
  • Very fast
  • Not accurate for short transfers
  • Macroscopic Flows in a pipe
  • (mathematical simulation
  • coarse-grain d.e. simulation)
  • Very fast
  • Abstract application model
  • Macroscopic abstract tasks with resource needs
    and dependencies
  • (coarse-grain d.e. simulation)
  • Very fast
  • Abstract application model

Emulation Discrete-event Mathematical
Simulation Simulation
more abstract
less abstract
Other Projects
  • ModelNet
  • Network emulation
  • unmodified application
  • packets routed through a core cluster
  • GigaBit-switched nodes running a modified kernel
  • Emulates router queues
  • More emulation than MicroGrid
  • Only for networking, but plans to add support for
  • Still many in-house simulators that may aspire
    to become widely-used tools
  • EmuLab/DummyNet
  • ChicagoSim
  • OptorSim
  • EDGSim
  • GridSim
  • ...

So what should I use?
  • It really depends on your goal / resources
  • SimGrids network model has clear limitations,
    e.g. for short transfers
  • SimGrid simulations are easy to set up
  • MicroGrid simulations take a lot of time
    (although they can be parallelized)
  • ModelNet requires some hardware setup
  • SimGrid does not require for a full application
    to be written
  • MicroGrid models overhead of system calls
  • ...
  • Key trade-off accuracy and speed
  • The more abstract the simulation the fastest
  • The less abstract the simulation the most
  • Does this trade-off really hold?

Simulation Validation
  • The crux of most simulation work in most domains
    of computer science
  • Validation is difficult and almost never done
  • Provide justification that the model is plausible
  • Convince people that the simulator implements the
    model (verification)
  • Provide a few graphs that show that its
  • validation in a few special cases, at best
  • validation against another validated simulator
  • Argue that although absolute values are off, the
    trends are respected
  • Conclude that the simulator is useful to compare
  • Obtain scientific results?????

  • FLASH vs. (Simulated) FLASH Closing the
    Simulation Loop Gibson et al., ASPLOS00
  • FLASH project at Stanford
  • building large-scale shared-memory
  • Went from conception, to design, to actual
    hardware (32-node)
  • Used simulation heavily over 6 years
  • The authors went back and compared simulation to
    the real world!
  • Simulation error is unavoidable
  • 30 error in their case was not rare
  • Negating the impact of we got 1.5 improvement
  • One should focus on simulating the important
  • A more complex simulator does not ensure better
  • simple simulators worked better than
    sophisticated ones, which were unstable
  • simple simulators predicted trends as well as
    slower, sophisticated ones
  • It is key to use the real-world to tune/calibrate
    the simulator
  • Conclusion for FLASH, the simple simulator was
    all that was needed

Presentation Outline
  • Introduction
  • Simulation for Grid Computing?
  • Generating Synthetic Grids
  • Simulating Applications on Synthetic Grids
  • Current Work and Future Directions

Grid Simulation Accuracy vs. Speed
  • Comparing simulators and validating them is a
    gigantic amount of work
  • It will never been clear-cut
  • identify simulation regimes
  • It doesnt lead to many papers
  • Its eminently politically incorrect
  • Results depend on what the simulation is used for
  • Therefore nobody does it
  • The story one would like to tell is, e.g.
  • Start with SimGrid simulations at first to
    identify promising approaches
  • Move to MicroGrid emulations to precisely
    quantify the trade-offs
  • How can we substantiate this story?

Current Work in SimGrid
  • SimGrid uses a simulation engine called SURF
  • currently SURF performs a blend of mathematical
    simulation and discrete event simulation
  • Current work
  • Adding a MaSSF back-end (i.e., MicroGrid)
  • Adding a ModelNet back-end
  • Goal evaluate the speed-accuracy trade-off for
    simulation of the network
  • Expected result
  • SimGrid sufficient and fast for large-enough
  • e.g., Good for scientific applications
  • MaSSF/ModelNet required for small/frequent
  • e.g., Good for middleware applications (e.g.,
    NWS), p2p applications
  • But beware FLASH vs. FLASH!

What about Validation?
  • The GRAS project Quinson et al.
  • Idea provide a way to compile code into
    real-world code and into simulation code
  • Write the application using the GRAS API
  • Compile it into a SimGrid simulation
  • Compile it into a real-world code
  • currently provides its own back-end and
  • could use Globus as a back-end
  • Run and compare
  • Goals
  • smooth transition from design to prototyping to
  • easy validation and simulation calibration
  • Plan
  • Use GRAS to easily compare SimGrid, MaSSF,
    ModelNet, and real world networks
  • Move on to full applications

  • Simulation is difficult
  • Eternal question What does really matters?
  • Grid researchers are actively working on it
  • Usable tools exist
  • Grid simulation today should not re-invent the
  • Two crucial next steps
  • Repository of synthetic Grids, Grid measurement
    datasets, and Grid simulation software
  • Scientifically sound validation experiments
  • Validate simulators
  • Understand what matters and what does not
  • Only then will we have a scientific discipline
    with a standard way to conduct experiments and a
    way for researcher to reproduce each others

(No Transcript)
A Simple Experiment
  • Sent out files of 500MB up to 1.5GB with TTCP
  • Using from 1 up to 16 simultaneous connections
  • Recorded the data transfer rate per connection

Experimental Results
Normalized data rate per connection
Number of concurrent TCP connections
Bandwidths, latencies, traffic, etc.
  • Option 2 Model end-to-end performance
  • Sub-option A Just scale physical link
  • Sub-option B Model end-to-end performance
  • Use a few simple laws to generate perceived
    network performance at the application level
  • e.g., Distribution of end-to-end transfer rates
    in Grid platforms is Normal Lee, HCW01
  • Use application-level measurements collected on
  • e.g., NWS measurements, Inca measurements
  • Pick real values randomly and assign them to
    end-to-end paths

Bandwidths, latencies, traffic, etc.
  • Problem with Option 2 Mapping between
    end-to-end network paths characteristics and
    individual links
  • We need link-level information for the simulation
  • e.g., for bandwidth-sharing between flows
  • We often only have path-level information
  • Heavy-duty tools exist for obtaining link-level
  • So far only ad-hoc approaches for this mapping
  • e.g., assume bottleneck is inside the network
  • Subversive thought Maybe none of this matters?
  • Fiber inside the network mostly unused
  • Each communication bottleneck is the local link
  • Appropriate tuning of TCP or better protocols can
    saturate the local link?
  • Bottom-line Still a lot of open questions
  • Upcoming DARPA workshop on large-scale network

Bandwidths, latencies, traffic, etc.
  • It gets worse
  • Network performance fluctuations?
  • No accepted model
  • One option
  • Use traces from measurement tools (e.g., NWS)
  • Conduct trace-driven simulations (which may be
  • Network failures?
  • No accepted model
  • Use arbitrary probabilistic models
  • Use failure data from measurement tools (e.g.,
  • For now, a combination of the above approaches
    with reasonable assumptions is the

Banwidth Sharing
  • Naïve Assumption Fair Sharing

20 Mb/sec
10 Mb/sec
10 Mb/sec
  • Good approximation for LANs
  • But what about WANs?
  • Different characteristics
  • massive bandwidth, massive traffic inside the
  • Multi-hop network paths

Max-Min Fairness
  • Captures other resource sharing beyond networks!

Interference of Communication and Computation
kreaseck et al., IJHPCA05
CPU sharing