The IT industry proposes the theory of application grid technology to solve the game resources, that is, the latest Optimal Grid technology is open to the online game market, but its high growth potential is to be inspected, especially in competitors. In situ. The focus of online game operators currently pays attention to investing in the capital expansion equipment, or in existing resources ... "Grid" temptation The development of the domestic online game industry is in full swing, new game, game operator as if Looking at the new car, it is endless. At the same time as the development of online game market, as a media, it is better to accompany and accompanied, and there are more peace of mind. At the beginning of the year, the Ministry of Culture has undergone the following, intended to regulate online game programs, Internet cafes. At the same time as the government strictly controls the development of online games, the game operators have not revealed enough patience, more and more new games, the equipment is purchased, but the player's eye is always limited, and there are also signs of reduction More and more people who eat cakes here, we don't want to be alarms, recall the 2003 car market bubble, who will oppose the expansion? Who will oppose the money to spend money in the online game, the big choice is not the best choice, because no one can predict how high growth in the game market will last. However, some of the problems in front of them can not be ignored. After all, the lack of response is slow, and it is necessary to combat the player's patience, and even lead to the player to invest in the game, in the IT industry, the IT industry proposes the theory of application mesh technology to solve the game resources. Theory, That is, the latest Optimal Grid technology. The grid itself is not a new thing, first chasing the head by IBM company, the mainstream IT manufacturers participate in it, jointly organize and develop the Open Grid Service Architecture architecture, the purpose is to solve the integrated use of heterogeneous Resources. For example, a user may have a wide variety of servers, including Windows, UNIX, Linux, etc., and these servers may be distributed in different places. At this time, they can use grid to make resource integration utilization. . Although grid technology can resolve heterogeneous environments, in a game's operating environment, most web servers are the same standard environment. Different from the general enterprise application, for a game operator, there may be thousands or more servers, this time the main problem to solve this is how to integrate these servers into a unified resource, not dispersion system. For example, there are two games running, traditionally use 1,500 servers for the first game service, and then use 1,500 servers to serve the second game, and finally use other servers to manage, Call Center, etc. If you apply grid technology, you can integrate all of these servers together. When a player is playing a game, there is no need to care about the resources accessed in the original 1,500 server or on another 1,500 server. So why you have to add an Optimal before the grid, which involves the optimization of the original technology. Because the grid integrates the resources only, it is also required to implement a management system for optimal use of resources. Here, here IBM's autonomous operation manager Automatical Program Manager (APM) is an example, through which Optimal Grid technology implements the network, server access to online game operators, and automatically configures resources.
Take the popular online game of "Stone Age" as an example, its big camp map has been high, then APM can transfer other idle server resources in the grid to implement system configuration and optimization. For example, in the "Stone Age", there are many game groups. Each group includes 8 to 10 servers. When a group of servers in a group have problems, the APM guarantees that the game's operation is not It will be affected, which will transfer other servers in this group, which is called self-healing. Therefore, Optimal Grid is mainly to help online game operators achieve dynamic load balancing. Many of the games are now running, once there is enough resources in a set of resources, the operator first thinks of adding new servers. This approach is not necessarily cost-effective, because the increased cost is likely to be consumed by the peak of the game, and the rest is wasted. With Optimal Grid, you can automatically move online players in the game run, especially when there are multiple games running simultaneously when a player logs in to a game, the game server group may appear to access overheating. At the same time, another game server may be in an idle state. At this time, Optimal Grid can use the idle server resources to use the first game, improve the utilization "a" to "more" revolutionary for online game developers In terms of applying Optimal Grid technology, it is possible to optimize the structure of the game engine from the beginning. For example, "Legend" developers have a large number of game maps, mostly 8 km from 8 km. The traditional approach is to cut the map into a small piece, then store the map on different servers, and communicate in the server and server. If a user logs in to the first server to access the A map, when you want to access the B map, you should register from the internal communication program to the second server, and then access. However, if Optimal Grid technology, developers can freely segment the entire map. For players, this is a map of 8 kilometers throughout 8 km, there is no tradition of stopping a pause, while the game player and All objects in the game world, including characters, scenarios, props, can smooth move between different servers. For game developers, this will not increase any difficulty, because Optimal Grid is equivalent to a middleware platform, with many open APIs and a large number of tools, can quickly achieve the game engine architecture from a single channel, become multi-server games Engine conversion. At the same time, the logic of the original game can be saved without the need to change the logic, but in fact, the basic principle of Optimal Grid technology is based on the original grid unit division model, which uses the Resources, such as maps and servers, cut into small resource units as small as possible. For example, a map of 8 kilometers is 8 km, you can cut into 2 million structural units, a unit is a non-divided cell for game logic, then this cell can represent it on this map. The minimum logo is the minimum characteristic on a map, called Original Program Cell (OPC), which includes its own data information and features, including communication methods between other cells, and on other servers. Communication connection between cells.
These small cells are formed into a subset of an OPC Collection cell, and then form some partitions in this subset, and then put these partitions and subset to the server running. This placement is a dynamic, rather than saying The subset is placed in the first server, which is always in the first server. In the future, it is possible to dynamically regulate the Popularity of the Internet according to the traffic of the user to the server, and the server being accessed. s position. Like IBM's APM can control the boundaries between subsets to achieve resource allocation and integration. When the player walks from a piece of this map, walk to a piece of another map, then this OPC subset is integrated with another OPC subset, so that the map can realize the automatic smooth moving transition to "pool" Online game operators and developers use Optimal Grid technology to make dynamic resource provisioning of games, saving part of operating costs. However, with the breakthrough of game engine technology in recent years, more and more 3D games, FPS games are attractive to player's eye, and the hardware resources of the operator are also crazy. This has produced a bottleneck: when a certain When the RPG game resource is insufficient, Optimal Grid can allocate server resources for it, but this resource may be a set of high-performance servers, and maybe another 3D shooting game has also had enough resources, but Optimal Grid can only take the rest. The resource is dynamically prepared, but it is not guaranteed that the high-performance server assigned to 3D games. This kind of bottleneck has been highlighted for several years, and the model of "resource pool" is proposed for this IT industry, which is Utility Computing Technology. The core of Utility Computing refers to the use of operators' existing resources to maximize costs, integrating different gaming applications, servers, forming a unified resource pool, providing all games. Here, let's take a look at the example of the grand network. In the grand, there are a set of servers for each online game, including a set of Open View Network Management and Game Customer Relationship Management Systems in the background. After the Utility Computing resource is dynamically distributed, the grand network can be realized, no matter how many new games are online, you can unify all server resources of the company through the flexible background infrastructure, including the resources supported by the background customers. For Grand Network, different games are different for server requirements, and Utility Computing supports operators to define different service levels based on different requirements (Services Level Agreement). That is to say, when a general RPG game service, such as "Legend", you can call low-performance server resources; and when the company is on some 3D class games, you can automatically call high-performance resources in the background. Also, if the grand network feels that the existing server is still not enough, it can add new servers to the interior. Different from the past, now the new set of servers is not used for a game, but for all games, services. At present, IBM has developed Tivoli Intelligent Architecture (TIO) software according to Utility Computing Technology, which can be derived from a global range, to control and call all the resource pools Money's way to go to the big city. The city is easy to confuse, for online games Operators, in addition to selling some cards, the transactions between players are extremely likely to become the growth point of future operators' profits.
