Determining Optimal Update Period for Minimizing...
Transcript of Determining Optimal Update Period for Minimizing...
![Page 1: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/1.jpg)
Determining Optimal Update Period for Minimizing Inconsistency in Multi-server Distributed Virtual Environments
Li Yusen, Wentong CaiPresented by Stephen John Turner
PDCC, SCENanyang Technological University, Singapore
![Page 2: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/2.jpg)
Overview
• Introduction• System Model• Problem Definition• Optimization Method• Performance Evaluation• Conclusion and Future Work
![Page 3: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/3.jpg)
Introduction to the Research Problem
• Distributed Virtual Environment
virtual world
object/entityavatar
node, client/user, player
To deploy on a group of computers connected via networks
![Page 4: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/4.jpg)
Introduction to the Research Problem
• Fundamental goal– Create a common and consistent representation of the virtual
world among all users– Any state change of an entity in the virtual world should be
disseminated to all users who require it in a timely manner• Challenges
– Network latency– Resource limitations as the number of users increases (e.g.,
MMOG)• Computational power• Network capacity
![Page 5: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/5.jpg)
Introduction to the Research Problem
• Research Objective– Derive state update schedules for improving consistency in
multi-server DVEs with network capacity constraints• Contributions
– Time-space inconsistency is used to evaluate the total inconsistency of an multi-server DVE
– The problem of minimizing total inconsistency is formulated as an Inequality Constrained Problem (ICP)
– Interior point method is used to solve the problem
![Page 6: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/6.jpg)
System Model
Servers are connected to each other in a peer to peer manner
The virtual world is partitioned into several fixed regions
Each region is maintained by one server (e.g., R1 is maintained by S1)
Client connects to the server if its avatar is residing in the region maintained by the server (e.g., C1 is connected to S3)
![Page 7: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/7.jpg)
System Model
The entities in an avatar’s AOI will be replicated at this avatar’s client side
Area of Interest (AOI) is used to define a neighborhood area for avatars
For a replica (e.g., triangle), the target server is the server that maintains the source entity
The contact server is the server that is connected by the client holding the replica
![Page 8: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/8.jpg)
System Model
• State Update Schema– Client first sends the operation on an entity to the server
maintaining this entity– Server executes the operation and disseminates new states to
all interested clients for updating the replicas– For a replica, if its target server and contact server are the same,
target server directly disseminates state update to the replica– If its target server and contact server are different, target server
first sends update to contact server, contact server forwards to replica
![Page 9: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/9.jpg)
Problem Definition
• Time-space inconsistency– ∆(t) : spatial difference between a replica and its source entity– Time-space inconsistency over [ , ]bt et
![Page 10: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/10.jpg)
Problem Definition
• Objective– To minimize total time-space inconsistency over all replicas with
a set of servers with limited network capacity• Assumptions
– For each replica, assume after the replica receives a position update, the difference ∆(t) grows following an increasing function δ(·), ∆(t) = δ(t-(tlast+d))
– Configurations such as world partition, client assignment, server side bandwidth, etc. remain unchanged over a period
![Page 11: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/11.jpg)
Problem Definition
• Theorem – In multi-server DVEs, for any replica, given a fixed number of
updates allowed in a period at the target server, these updates should be disseminated periodically over this period for minimizing time-space inconsistency
– To minimize total time-space inconsistency over all replicas over a period with a set of servers with limited network bandwidth, we just need to determine the optimal update period of each replica
![Page 12: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/12.jpg)
Problem Definition (cont.)
• Notations– the number of servers in the DVE– the ith server in the DVE– the number of replicas in the DVE– the number of replicas whose target server is– the number of replicas whose target server is and contact
server is – the kth replica in the DVE– the entity which is replicating, i.e., source entity of – the target server id of – the contact server id of
NS
iSNR
iNRijNR iS
jS
iS
kr( )ke r kr( )kT r kr( )kC r kr
kr
![Page 13: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/13.jpg)
Problem Definition (cont.)
• Notations– the set of replicas whose target server is– the set of replicas whose contact server is – the transmission delay of position update of replica from
target server to – a bandwidth consumption for disseminating a position update– b bandwidth consumption for receiving and forwarding a
position update– the network capacity of – the update frame length of each server– update period of replica
TiR iS
ic
CiR iS
id irir
iS
rfkp
kr
![Page 14: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/14.jpg)
Problem Definition
Objective function to minimize: total time-space inconsistency over all replicas over period T
Network capacity constraint for each server
Bandwidth consumption on disseminating position updates for the replicas whose target server is this server
Bandwidth consumption on forwarding
![Page 15: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/15.jpg)
Convex Optimization
• Problem Transformation– Let , the problem is converted to minimize
– Convex Property
![Page 16: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/16.jpg)
Convex Optimization
• Interior Point Method– The basic idea is to approximate the original problem to the
following problem
– α is a parameter that sets the accuracy of the approximation• Solution
– Define
– is a convex function and if holds, is a global minimum.
– Gradient Descent Method
![Page 17: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/17.jpg)
Convex Optimization
• Gradient Descent Method– Iterative Algorithm
– t is a constant value, can be different for each iterative step• Values Need to Know
–
–
– and need to be estimated
![Page 18: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/18.jpg)
Performance Evaluation
• Experimental parameters
Parameter Default ValueDVE DimensionNumber of Servers
5000x5000 (distance units)2
25Number of Regions 100Number of Clients/avatars 1500AOI Size 500x500Average Network Latency 100msVariance of LatencyFrame Lengtha, bEntity Moving SpeedNetwork Capacity
0.950.025s1 unit[0.1, 10] distance units/frame[5, 300] units
![Page 19: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/19.jpg)
Simulation Results
• Converge Speed of Iterative Algorithm
Most of variables converge after 3000 iterative steps
![Page 20: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/20.jpg)
Simulation Results
• Impact of α in the Interior Point Approximation
Larger α makes more accurate, but more difficult to converge
parameter valuesNetwork latency 100msNetwork capacity 50T 60sServer Number 25
![Page 21: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/21.jpg)
Simulation Results
• Impact of Network Capacity
parameter valuesNetwork latency 100msT 60sServer number 25
![Page 22: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/22.jpg)
Simulation Results
• Impact of Network Latency
parameter valuesNetwork capacity 50T 60sServer number 25
![Page 23: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/23.jpg)
Simulation Results
• Impact of Inter-server Communication
parameter valuesNetwork latency 100msNetwork capacity 50T 60s
![Page 24: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/24.jpg)
Conclusion and Future Work
• Conclusion – Study the update scheduling issues in multi-server DVEs with
limited network bandwidth– Formulate and solve the problem for an ideal situation where
configurations keep unchanged• Future Work
– Update schedules in practical systems
![Page 25: Determining Optimal Update Period for Minimizing ...c4i.gmu.edu/eventsInfo/conferences/2011/DS-RT... · The entities in an avatar’s AOI will be replicated at this avatar’s client](https://reader034.fdocuments.us/reader034/viewer/2022042401/5f106f777e708231d44918a6/html5/thumbnails/25.jpg)
The End