GPN-Perf2: Game private networks and game servers performance optimization

GPN-PERF是由奥卡纳根学院及其行业合作伙伴WTFast实施的一个分两个阶段的项目。 第一阶段于2014年开始，作为正在进行的NSERC CCI ARD Level 1项目。它的主要结果是：(A) 用于测试游戏服务器的实验GPN基础设施的设计与实现 软件在“体外”实验室环境中的性能，(B)隔离服务器软件的影响 来自网络延迟和带宽，这在WTFast连接玩家的行业设置中是不可能的 使用互联网上的服务器(C)测量和解释GPN性能的关键统计数据。 奥卡纳根学院的目标是在第一阶段完成后立即安排第二阶段，以便 为了保持学生的兴趣，WTFast的持续参与，从而减少了学术研发之间的滞后 以及未来的高价值WTFast服务。该项目将利用在第一阶段创建的基础设施 项目作为测试和构建可以优化和监控的游戏服务器软件的框架 WTFast实时游戏服务。此基础设施将允许自动部署新服务器并 配置为用作私人游戏服务器，同时还监控其性能和使用统计数据。通过 使用新的创新预测模型(由项目创建)，可以在以下情况下自动添加新服务器 流量级别将需要更多资源来维持最佳性能。 将使用模拟程序对网络基础设施和网络软件进行测试 许多玩家连接到网络中多个虚拟机上的多个游戏服务器。这些项目的目标是 网络测试是为了确定网络软件无法再跟上流量的时间点。 这些测试还将用于识别游戏服务器及其主机虚拟机的容量 他们可以支持的玩家数量。 这一知识和相关的实验功能在行业中是绝无仅有的，将为我们提供 更深入的知识，加上游戏网络领域的产业创新。

GPN-Perf is a two-phase project conducted by Okanagan College and its industrial partner WTFast. The first phase begins in 2014 as an undergoing NSERC CCI ARD Level 1 project. Its main results are (a) the design and implementation of an experimental GPN (Game Private Network) infrastructure to test game server software performance on an "in vitro" laboratory environment, (b) isolation of the effect of server software from network latency and bandwidth, which is not possible in WTFast's industrial setup that connects players with servers over the internet (c) measuring and explaining key statistics of the GPN performance. Okanagan College's goal is to schedule the second phase immediately after completion of the first phase so as to retain student interest, continuous involvement of WTFast and thus reduce the lag between academic R&D and future high-value WTFast services. This project will utilize the infrastructure created in the first phase of the project as the framework on which to test and build game server software that can optimize and monitor WTFast game services in real time. This infrastructure will allow new servers to be automatically deployed and configured for use as private game servers, while also monitoring their performance and usage statistics. By using new innovative predictive models (created by the project), new servers may be automatically added when traffic levels will require more resources to maintain optimal performance. Testing of the network infrastructure and the network software will be done using a program that simulates many players connecting to many game servers on several virtual machines in the network. The goal of these network tests is to identify the point at which the network software can no longer keep up to the flow of traffic. The tests will also serve to identify the capacity of the game servers and their host virtual machines in terms of the number of players they can support. This knowledge and associated experimental functionality exists nowhere else in the industry and will give us deeper knowledge, plus industrial innovation in the game networking area.