Abstract
Method of information flow control in a hyperconvergent system
Highlights
Methods of distribution of information flows under centralized control are considered in many scientific works [9–12]
Where cu – total information density; hr – number of information flows between a set of nodes of network; hm – number of routes for transfer to j -th flow in distribution ; cmaj – density of the j -th flow on route maj ; hwaj – route length maj, determined by number of the communication channels entering a route; tcomm – average time of switching of a packet; kz – average load factor of a channel; wait – average length of queue to a channel; – the average volume of the packets transferred to networks; pz – average bandwidth of communication channels taking into account their loading
It is visible that with growth of network relative complexity the offered method allows to reduce expenses of a computing resource to two times. It is offered a method of information flow control in a hyperconvergent system
Summary
The computing power, storage, servers, networks are combined. With a hyper-convergent structure, one system administrator is enough to control the system. This significantly reduces the cost of maintaining the system. At the same time, system performance is falling due to centralization of control [6] In such a system, the task of optimizing the distribution of information flows plays a significant role [7–8]. Methods of distribution of information flows under centralized control are considered in many scientific works [9–12]. The method should take into account the features of centralized control and the lack of heterogeneity of software and hardware
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
