2022 Fiscal Year Annual Research Report
Reliable Resource Allocation Models and Management System with considering computing workload
| Project/Area Number |
22J20748
|
|---|
| Allocation Type | Single-year Grants |
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
ZHU MENGFEI 京都大学, 情報学研究科, 特別研究員(DC1)
|
|---|
| Project Period (FY) |
2022-04-22 – 2025-03-31
|
| Keywords | function virtualization / function deployment / backup / protection / failure / recovery / workload / reliability |
|---|
| Outline of Annual Research Achievements |
Fault tolerance and load balancing are two key roles in resource allocation against failures. In the first year of my Ph.D. research, I conducted the resource-sharing model design and description. I proposed a primary and backup resource allocation model with preventive recovery priority setting to minimize a weighted value of unavailable probability against multiple failures. In addition, I design and implements a Real-time Function Deployment system with Resource Migration in Kubernetes to manage the primary and backup resources of network functions in the dynamic scenario for prompt function deployment and management, which is a key role in network function virtualization to improve the continuity and reliability of network services.
|
| Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
In past studies, function deployment models considering backup protection, load balance, and recovery/migration have been studied. The related implementation methods in real networks have also been completed. I am moving forward with considering the robustness of recovery against failures and fast failover algorithm against failures.
|
| Strategy for Future Research Activity |
In the second year of my Ph.D. research, I begin by working with Prof. Klaus-Tycho Foerster at TU Dortmund in collaborative research. Our research topic focuses on the after-failure recovery strategy. While my previous research has extensively investigated backup resource allocation for protection, it has not delved into transfer after a failure has occurred in the shared protection scenario. As a result, we will be putting forth a local algorithm considering the pre-processing. I will present a theoretical analysis of its performance.
In addition, local after-failure recovery strategies can also involve the use of fault-tolerant algorithms. These algorithms are designed to continue functioning even in the presence of failures or errors.
|
