270 likes | 458 Views
Dynamic Server Selection for Large Scale Interactive Online Games. Bong-Jun Ko, Dan Rubenstein (Columbia U.) Kang-Won Lee, Seraphin Calo (IBM T. J. Watson). Talk Outline. Motivations Synchronization Delay Model Problem Statement Server Selection Algorithms Performance Evaluation
E N D
Dynamic Server Selection for Large Scale Interactive Online Games Bong-Jun Ko, Dan Rubenstein (Columbia U.) Kang-Won Lee, Seraphin Calo (IBM T. J. Watson)
Talk Outline • Motivations • Synchronization Delay Model • Problem Statement • Server Selection Algorithms • Performance Evaluation • Conclusions
Motivation • Large-scale online games (e.g. MMOG) start toincorporate real-time interactive features. • Game service providers must satisfy tight delay requirements and large scale user group at the same time. • To achieve scalability and efficient resource sharing, it is important to reduce game server resources used.
Our Goals • Develop a synchronization delay model for interactive online applications in mirrored-server architecture • Design a distributed server selection algorithm to minimize resource consumption while meeting the overall delay constraints.
Action Action State State Synchronization Delay in Online Games • Two different types of synchronization delay • Upstream (Action Delivery) delay • Downstream (State Update) delay
Server Clients Mirrored-Server Architecture • S (set of available servers) and C (set of clients) • A: CS (server allocationthat maps each client in C to a server in S)
Sync-delay in Mirrored-Servers • DUP(i, k) : Upstream Delay from client i to server k • DDOWN(k, i) : Downstream Delay at server k to client i • DSYNC(A) : Overall sync-delay under allocation A DSYNC(A) = maxk{maxi DUP(i, k) + maxi DDOWN(k, i)}
Problem statement • Find an allocation Amin that minimizes the number of allocated servers,|A(C)|, subject to • DSYNC(A) , • |A(C)| 1, and • C(si) Ci(C(si) : the capacity of server si )
5 3 1 1 1 3 1 5 Sync-delay revisited: Example C1 S1 S2 S3 C2
Sync-delay revisited: Example C1 5 3 1 1 1 S1 S2 S3 3 1 5 C2 DSYNC = 3+3 = 6
Sync-delay revisited: Example C1 5 3 1 1 1 S1 S2 S3 3 1 5 C2 DSYNC = 5+5 = 10
Sync-delay revisited: Example C1 5 3 1 1 1 S1 S2 S3 3 1 5 C2 DSYNC = 3+1 = 4
Observations • Connecting clients to a server further than the ‘center’ doesn’t reduce the overall delay. • Allocating ‘edge’ servers generally decreases the delay, but increases the number of servers.
Server Selection Algorithm:Zoom-In-Zoom-Out(ZIZO) • Find the “core” server that minimizes the maximum distance from the client. • Connect each client to the closest server. • For each client, probe a server “closer” to the core serve. If sync-delay is still within bound, migrate the client to that server. Repeat until no client can move. • For each client, probe an allocated server “further” from the core server, and migrate if the sync-delay is within bound. Repeat until no client can move.
Variations • Selecting next server to probe : • By searching next closest server toward core server (Sr) • By moving up/down along a Core-Based Tree in the server network (Tr) • Migrating clients • Client-based : move each client one at a time (C) • Server-based : move all clients attached to a server simultaneously (S)
Simulation setup • A transit-stub network of 5,000 nodes generated by a topology generator (100 AS x 50 nodes/AS) • |S| servers selected at random and form a low-latency network among them with average degree 5. (|S|=100, 200) • |C| clients selected at random, each connected to the closest server initially. (|C|=50, 100) • Sync-delay bound () : 20 ~ 50 msec.
Greedy Server Selection Algorithm • A centralized algorithm • Keep adding servers until the sync-delay becomes within the bound. • When adding a server, add the one that minimizes the overall sync-delay.
Result : Unbounded server capacity |S|= 100, |C| = 50, C(s) =
ZINS(Zoom-In-’N-Stop) vs ZIZO |S|= 100, |C| = 50, C(s) =
Result : Bounded server capacity |S|= 100, |C| = 50, C(s) = 10
Conclusions • We presented a novel synchronization delay model for interactive online games. • We proposed a set of simple and effective distributed heuristics that efficiently utilize server resource. • Future Work • Complexity analysis of the problem • Efficient core-selection method • Play PlanetSide !