# US 10,871,986 B2 · Virtual server migration after performance deterioration > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,871,986 B2 - Document ID: US10871986B2 - Status: Granted - Kind: B2 - Application number: 16/427,815 - Filed: 2019-05-31 - Published: 2020-01-09 - Granted: 2020-12-22 - Earliest claimed: 2018-07-03 - Assignees: Fujitsu Limited - Applicants: Fujitsu Limited - Inventors: Hiroyoshi Kodama; Shigeto Suzuki; Hiroyuki Fukuda; Kazumi Kojima - Web page: https://patentgrep.com/patent/US10871986B2 ## CPC - G06F 9/4856: {resumption being on a different machine, e.g. task migration, virtual machine migration (G06F9/5088 takes precedence)} - G06F 9/5077: {Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455)} - G06F 9/5088: {involving task migration} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A migration management method includes referring to a performance deterioration rate of a specific virtual server when utilization of a virtual server other than the specific virtual server in virtual servers that work on a physical server is changed in a stepwise manner, and calculating a first index value relating to a load state of the physical server before the performance deterioration rate exceeds a threshold based on a number of virtual CPUs allocated to each of the virtual servers and utilization of the virtual CPUs; calculating a second index value relating to the load state based on the number of virtual CPUs and the utilization while the specific virtual server is activated on the physical server; and conducting migration to another physical server for a virtual server other than the specific virtual server in the virtual servers when the calculated second index value exceeds the calculated first index value. ## Claims ### Claim 1 A migration management method executed by a computer, the migration management method comprising: referring to a performance deterioration rate of a specific virtual server when utilization of a virtual server other than the specific virtual server in virtual servers that work on a physical server is changed in a stepwise manner, and calculating a first index value relating to a load state of the physical server before the performance deterioration rate of the specific virtual server exceeds a threshold based on a number of virtual CPUs allocated to each of the virtual servers that work on the physical server and utilization of the virtual CPUs; calculating a second index value relating to the load state of the physical server based on the number of virtual CPUs allocated to each of the virtual servers that work on the physical server and the utilization of the virtual CPUs while the specific virtual server is activated on the physical server; and conducting migration to another physical server for a virtual server other than the specific virtual server in the virtual servers that work on the physical server when the calculated second index value exceeds the calculated first index value. ### Claim 2 (depends on 1) The migration management method according to claim 1, wherein the other physical server is a dedicated physical server prepared as a migration destination in advance. ### Claim 3 (depends on 2) The migration management method according to claim 2, further comprising: conducting migration to the physical server for any virtual server on the other physical server when utilization of the physical server has become lower than a threshold. ### Claim 4 (depends on 1) The migration management method according to claim 1, wherein the other physical server is a physical server in which utilization is lower than a threshold in a plurality of physical servers as candidates for a migration destination. ### Claim 5 (depends on 1) The migration management method according to claim 1, further comprising: referring to a past running status of each of the virtual servers that work on the physical server and predicting utilization of each of the virtual servers after elapse of a certain time; and calculating a third index value relating to the load state of the physical server after elapse of the certain time based on the predicted utilization of each of the virtual servers after elapse of the certain time, wherein the conducting the migration includes conducting the migration when the calculated second index value exceeds the first index value and the calculated third index value exceeds a threshold. ### Claim 6 (depends on 1) The migration management method according to claim 1, wherein the physical server is in a state in which a total number of virtual CPUs allocated to the virtual servers that work on the physical server exceeds the number of physical cores which the physical server has. ### Claim 7 (depends on 1) The migration management method according to claim 1, wherein the load state of the physical server before the performance deterioration rate of the specific virtual server exceeds the threshold is the load state of the physical server in a state immediately before the performance deterioration rate of the specific virtual server exceeds the threshold in a plurality of states obtained by raising the utilization of the virtual server other than the specific virtual server in a stepwise manner. ### Claim 8 (depends on 1) The migration management method according to claim 1, wherein the calculating the second index value includes occasionally calculating the second index value relating to the load state of the physical server while the specific virtual server is activated on the physical server. ### Claim 9 A non-transitory computer-readable storage medium storing a program that causes a processor included in a computer to execute a process, the process comprising: referring to a performance deterioration rate of a specific virtual server when utilization of a virtual server other than the specific virtual server in virtual servers that work on a physical server is changed in a stepwise manner, and calculating a first index value relating to a load state of the physical server before the performance deterioration rate of the specific virtual server exceeds a threshold based on a number of virtual CPUs allocated to each of the virtual servers that work on the physical server and utilization of the virtual CPUs; calculating a second index value relating to the load state of the physical server based on the number of virtual CPUs allocated to each of the virtual servers that work on the physical server and the utilization of the virtual CPUs while the specific virtual server is activated on the physical server; and conducting migration to another physical server for a virtual server other than the specific virtual server in the virtual servers that work on the physical server when the calculated second index value exceeds the calculated first index value. ### Claim 10 A migration system, comprising: an information processing apparatus including a processor; a plurality of physical servers coupled to the information processing apparatus; and a standby server that is coupled to the information processing apparatus and is a dedicated physical server prepared as a migration destination in advance, wherein the processor is configured to: refer to a performance deterioration rate of a specific virtual server when utilization of a virtual server other than the specific virtual server in virtual servers that work on a physical server is changed in a stepwise manner, and calculate a first index value relating to a load state of the physical server before the performance deterioration rate of the specific virtual server exceeds a threshold based on a number of virtual CPUs allocated to each of the virtual servers that work on the physical server and utilization of the virtual CPUs; calculate a second index value relating to the load state of the physical server based on the number of virtual CPUs allocated to each of the virtual servers that work on the physical server and the utilization of the virtual CPUs while the specific virtual server is activated on the physical server; and conduct migration to the standby server for a virtual server other than the specific virtual server in the virtual servers that work on the physical server when the calculated second index value exceeds the calculated first index value.