In order to improve the data access ability of the current dispatching control system, reduce the data source processing pressure, and meet the increasing demand for high concurrency and large amount of data access, this paper starts with the requirement of high-speed data access, and proposes the key technology of constructing the high speed data cache component of the dispatching system based on the distributed memory cluster, and solves the distributed environment with the circular consistent hashing algorithm. For the balance of business data storage, the gradual hash mechanism is used to improve the scaling efficiency of the mapping table during storage, improve the query and access efficiency of multiple data information, ensure the high reliability of the data through the redundant multi copy cache architecture of the master-slave sentinel cluster, and ensure that the expired cache is cleaned and refreshed in time through the automatic update strategy of multiple cache data, and solve the data consistency requirements of different business scenarios. At the same time, an experimental verification environment is built to verify the concurrent processing performance of cache components.
[1]
Tony Tung,et al.
Scaling Memcache at Facebook
,
2013,
NSDI.
[2]
陈刚,et al.
Design of Scalable Distributed Metadata Management System
,
2015
.
[3]
Li Xiao,et al.
On scalable and locality-aware web document sharing
,
2003,
J. Parallel Distributed Comput..
[4]
Cho-Li Wang,et al.
Building a scalable web server with global object space support on heterogeneous clusters
,
2001,
Proceedings 42nd IEEE Symposium on Foundations of Computer Science.
[5]
Virginie Grandgirard,et al.
Optimization of Fusion Kernels on Accelerators with Indirect or Strided Memory Access Patterns
,
2017,
IEEE Transactions on Parallel and Distributed Systems.
[6]
Qin Xiu.
Progress and Challenges of Distributed Caching Techniques in Cloud Computing
,
2013
.