Title: EdgeBased Cloud Computing as a Feasible Network Paradigm
1Edge-Based Cloud Computing as a Feasible Network
Paradigm
- Joe Elizondo and Sam Palmer
2Introduction
- Edge-based cloud computing new computing
paradigm! -
- Combination of two ideas
- Edge Computing massively distributed grid
computing, public resource computing (e.g.
SETI_at_Home, Folding_at_Home) -
- Cloud Computing virtualized resources,
scalable, dynamically allocated
3Motivation
- Inexpensive computation
- High performance per dollar ratio
-
- Leverage available idle CPU cycles and internet
bandwidth (potentially free to use, no existing
cost model) -
- Existing Infrastructure
- Every host on the Internet could potentially
participate -
- Access an edge cloud from anywhere in the world
4Our Work
Is an edge-based cloud computing paradigm
feasible?
- Find answer through simulation
High level Approach
- Model the Internet
-
- Build a cloud
-
- Simulate MapReduce jobs
-
- Evaluate performance
5Our Work
Is an edge-based cloud computing paradigm
feasible?
- Find answer through simulation
High level Approach
- Model the Internet
-
- Build a cloud
-
- Simulate MapReduce jobs
-
- Evaluate performance
6Model the Internet (1/3)
- Hand-coding thousands of routers and nodes has
obvious disadvantages. - Why not use a topology generator?
-
- GT-ITM - Georgia Tech Internetwork Topology
Models -
- BRITE - Boston university Representative
Internet Topology gEnerator -
- Sacrifice realistic results in simulation.
-
7Model the Internet (2/3)
- Measure link speeds and latency for every
backbone router of every major ISP in the world? -
- Realistic topology with accurate simulation
results -
- Challenging?
8Model the Internet (3/3)
- University of Washington's Rocketfuel Project
-
- Rocketfuel - ISP toplogy mapping engine
-
- Data -
-
-
-
-
-
-
-
-
-
-
-
-
-
-
9Our Work
Is an edge-based cloud computing paradigm
feasible?
- Find answer through simulation
High level Approach
- Model the Internet
-
- Build a cloud
-
- Simulate MapReduce jobs
-
- Evaluate performance
10Build a Cloud (1/3)
- Python script attaches heterogeneous end hosts to
the network topology in our simulation.
11Build a Cloud (2/3)
- Heterogeneity accomplished by assigning end host
resources from the following choices.
End Host Link Speeds
Last hop link speeds are assigned a bandwidth and
latency based on a normal distribution given a
mean and a standard deviation.
12Build a Cloud (3/3)
- Final Step Output files for use in our
simulation - Python script outputs ns-2 readable TCL files
containing our internet topology -
-
-
-
-
-
-
- Python script outputs end host information to an
XML file that we pass into our simulation
TCL code to create two backbone routers
1012914 if info exists n("101Seattle,WA")
0 set n("101Seattle,WA") ns node
if info exists n("2914Seattle,WA") 0
set n("2914Seattle,WA") ns node
TCL code to create link 101Seattle, WA
-gt11608Seattle,WA 0 101Seattle, WA -gt
101Sunnyvale, CA 5.68752395038991ns
duplex-link n("101Seattle,WA")
n("101Sunnyvale,CA") 10.0Gb 5.68752395038991ms
DropTail
ltmachine_typegt ltnamegtEndHost1667lt/namegt
ltdiskgtlttypegtdrive3lt/typegtltcapagt250lt/capagtltnumgt1lt/n
umgtlt/diskgt ltcpugtlttypegt1.6Ghzlt/typegtltnumber_of_
coresgt1lt/number_of_coresgtltnumgt1lt/numgtlt/cpugt
ltmemgtlttypegtECClt/typegtltcapagt1024lt/capagtlt/memgt
ltnicgtlttypegt100Mbpslt/typegtltnumgt1lt/numgtlt/nicgtlt/mach
ine_typegt
13Our Work
Is an edge-based cloud computing paradigm
feasible?
- Find answer through simulation
High level Approach
- Model the Internet
-
- Build a cloud
-
- Simulate MapReduce jobs
-
- Evaluate performance
14Simulate MapReduce Jobs (1/4)
- Why MapReduce?
- MapReduce operations model the high level of
coordination and communication that takes place
between machines in a cloud computing cluster.
- We use MRPerf (Viginia Tech, IBM Almaden)
- MRPerf merges MapReduce and network simulation to
achieve a seamless simulation environment. - Claims to predict simulation performance within
5.22 of actual measurements for map and 12.83
for reduce for a double rack cluster with 16 to
128 cores.
15Simulate MapReduce Jobs (2/4)
- MRPerf - Simulation tool for evaluating MapReduce
performance on large clusters.
MRPerf simulates Hadoop's implementation of
MapReduce using ns-2
MRPerf Original Architecture
16Simulate MapReduce Jobs (3/4)
MRPerf is designed to model performance on a data
center infrastructure.
- Implications
- Chunk size, data replication, node bandwidth,
mappers/reducers per node, scheduling, etc.
17Simulate MapReduce Jobs (4/4)
Our work requires modifications to architecture
and parameters to measure performance of
edge-based cloud.
MRPerf Architecture after modifications (in grey)
18Our Work
Is an edge-based cloud computing paradigm
feasible?
- Find answer through simulation
High level Approach
- Model the Internet
-
- Build a cloud
-
- Simulate MapReduce jobs
-
- Evaluate performance
19Simulation Setup
- Simulations were run over a three week period on
a combination of UT's Condor Cluster and TACC's
Sun Constellation Linux Cluster (Ranger) - All simulations sort 1GB of data
- Variables
- End host link bandwidth
- Chunk size
- Data center
- Mapped Internet
- Total number of hosts
- Data center
- Mapped Internet
- Single AS (United States)
- Map and reduce slots per node
20(No Transcript)
21(No Transcript)
22(No Transcript)
23(No Transcript)
24(No Transcript)
25(No Transcript)
26(No Transcript)
27Future Work
- Verify simulation results
- Investigate effects of node churn
-
- Develop a new MapReduce scheduler optimized for a
WAN -
- Evaluate other cloud-based services in an edge
environment