Cloud-assisted rendering support for mobile online gaming
For cloud gaming services to be successfully utilized in a commercial setting, a certain service level and server responsiveness level needs to be achieved. This issue is explored in the context of modern day cloud services, with easily scalable virtual machines and geo distributed servers. All...
Saved in:
| 主要作者: | |
|---|---|
| 其他作者: | |
| 格式: | Final Year Project |
| 語言: | English |
| 出版: |
2017
|
| 主題: | |
| 在線閱讀: | http://hdl.handle.net/10356/70476 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 機構: | Nanyang Technological University |
| 語言: | English |
| 總結: | For cloud gaming services to be successfully utilized in a commercial setting, a certain
service level and server responsiveness level needs to be achieved. This issue is explored
in the context of modern day cloud services, with easily scalable virtual machines and geo
distributed servers. Allocation of clients to server is done while ensuring latency is low,
allocated server can serve client effectively, and making sure cost is minimized.
This project involves implementing systems to ensure effective allocation of servers to thin
clients. On top of the actual allocation algorithm itself, an automated server management
system, cost calculation algorithm and latency calculation algorithm is implemented
alongside.
Previously developed system images of the rendering server will be used for testing. The
main purpose of the test is to ensure allocation system is able to integrate into the existing
system and be fully functional.
Allocation system is developed with scalability and extensibility in mind. This means the
system should be able to take in newly available datacentre information and still be able to
provide optimum allocations with consideration given to this new set of information. This
means the system is able to both compute new optimum costs and also be able to initialize,
control and connect thin clients to instances within the new datacentre. |
|---|