Google Apps Engine 与 Amazon EC2 中的 Flex 扑克游戏扩展

发布于 2024-09-28 12:35:06 字数 450 浏览 3 评论 0原文

我有一个用java编写的服务器端代码和用flex编写的客户端代码。由于客户端大约每 3 秒向服务器发出一次请求,所以我很确定我在某个时候遇到了可扩展性问题。

据我所知,我可以在 Google Apps Engine 或 Amazon Cloud 中托管该应用程序。据我了解,亚马逊提供了一种自己动手的方法,我必须自己设置和维护集群,我不太喜欢这种方式,所以我倾向于使用 GAE,如果可以的话,它承诺几乎无限的可扩展性当然有限制和限制。

其中一个限制是每秒 500 个请求(可以通过填写特殊表格临时请求更多),这大致相当于 1500 个并发用户,每 3 秒对服务器进行一次 ping 操作。

让我困扰的是,我真的不知道这个东西能发展到多大,而且 1500 名玩家看起来绝对不是很多,所以除了使用 Google Apps 和 AWS(除非它可以更好地扩展我想要做的事情) ,我还有什么其他选择,而且不会花费我一大笔钱?

谢谢。

I have a server side code written in java and client side code written in flex. As client makes a request roughly every 3 seconds to the server, then I'm pretty sure I'm running into scalability problems at some point.

As far as I can see, I can host the application either in Google Apps Engine or Amazon Cloud. As I understand Amazon offers a do-it-yourself approach and I have to set up and maintain the cluster myself, which I'm not really fond of so I'm inclined to use GAE, which promises virtually infinite scalability if one can conform to the limitations and restrictions of course.

One of the limitations is the number of requests at 500 per second (can request temporarily more by filling in a special form), which roughly translates to 1500 simultaneous users, each pinging after every 3 seconds to the server.

What bothers me is that I really don't know how big this thing can grow and 1500 players definitely doesn't seem like a lot so besides using Google Apps and perhaps AWS (unless it scales better with what I'm trying to do), what other options do I have, which wouldn't cost me a fortune?

Thank you.

如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

扫码二维码加入Web技术交流群

发布评论

需要 登录 才能够评论, 你可以免费 注册 一个本站的账号。

评论(1

南城追梦 2024-10-05 12:35:06

您的 App Engine 应用程序可以维持的 QPS 数量不再有硬性限制(并且无论如何,增加都不是暂时的)。只要您的平均请求延迟低于 1000 毫秒,我们就会根据需要配置更多实例。

就轮询而言,是的,在资源消耗方面,这不会很好地扩展。即将推出的 Channels API 将提供一种异步通知用户浏览器事件的方法。

There's no longer a hard limit on the number of QPS your App Engine app can sustain (and in any case, increases were never temporary). As long as your average request latency is below 1000 milliseconds, we'll provision more instances as needed.

As far as polling goes, yes, that won't scale well when it comes to resource consumption. The upcoming Channels API will provide a way to asynchronously notify a user's browser of events.

~没有更多了~
我们使用 Cookies 和其他技术来定制您的体验包括您的登录状态等。通过阅读我们的 隐私政策 了解更多相关信息。 单击 接受 或继续使用网站,即表示您同意使用 Cookies 和您的相关数据。
原文