An evaluation of checkpoint recovery for massively multiplayer online games

Abstract

Massively multiplayer online games (MMOs) have emerged as an exciting new class of applications for database technology. MMOs simulate long-lived, interactive virtual worlds, which proceed by applying updates in frames or ticks, typically at 30 or 60 Hz. In order to sustain the resulting high update rates of such games, game state is kept entirely in main memory by the game servers. Nevertheless, durability in MMOs is usually achieved by a standard DBMS implementing ARIES-style recovery. This architecture limits scalability, forcing MMO developers to either invest in high-end hardware or to over-partition their virtual worlds.In this paper, we evaluate the applicability of existing checkpoint recovery techniques developed for main-memory DBMS to MMO workloads. Our thorough experimental evaluation uses a detailed simulation model fed with update traces generated synthetically and from a prototype game server. Based on our results, we recommend MMO developers to adopt a copy-on-update scheme with a double-backup disk organization to checkpoint game state. This scheme outperforms alternatives in terms of the latency introduced in the game as well the time necessary to recover after a crash.

Achieving Consistent Storage for Scalable MMORPG Environments

“…According to the architecture currently applied in MMORPGs [22,19], modifications of state data are executed by an in-memory database in real-time [5,23], so that these changes can be synchronously propagated to the relevant players within an acceptable period of time [10]. They are then backed up to the disk resident database periodically (checkpointing) [6,23].…”

“…They are then backed up to the disk resident database periodically (checkpointing) [6,23]. The checkpoint is created for two purposes: one is to recover the in-memory database in the case of a system failure; another one is to restore the state data to the in-memory database when a player restarts the game (state data will be removed from the in-memory database when a player logs out of the game in order to reduce memory consumption).…”

A latency and fault-tolerance optimizer for online parallel query plans

“…Main-memory DBMSs [25,34,28] use a variety of checkpointing strategies to preserve the in-memory state of their databases. In contrast, our approach preserves and recovers the state of ongoing computations.…”

Implementing performance competitive logical recovery

“…Recovery performance has rarely been quantified by experiments, and only once to our knowledge [17] in a side-by-side comparison, and that was in a main memory setting for online games. Our third contribution, in Section 5, shows how to control conditions to provide a valid side-by-side experimental comparison of logical recovery with physiological recovery, where both use exactly the same log.…”