Method and apparatus for memory integrated management of cluster system

A method for memory integrated management in an apparatus for memory integrated management of a cluster system including a plurality of physical nodes connected to each other by a network, the method comprising:

• determining one of the plurality of physical nodes as a node to place a new virtual machine;
• allocating the first type of memory allocated to the one physical node to the new virtual machine as much as the memory capacity required by the new virtual machine;
• and
• distributing the second type of memory to a plurality of virtual machines running on the plurality of physical nodes by integrating and managing the second type of memory allocated to each of the plurality of physical nodes,
• wherein the memory of the second type has a faster access speed than that of the first type of memory,
• wherein the distributing the second type of memory includes:
• distributing the second type of memory allocated to each physical node to virtual machines running in the corresponding physical node, respectively;
• and
• additionally distributing the remaining second type of memory in the cluster system to at least one virtual machine among the plurality of virtual machines,
• wherein the distributing the second type of memory allocated to each physical node to virtual machines running in the corresponding physical node, respectively includes:
• distributing the memory of the second type as much as the memory capacity required to guarantee the performance of the virtual machine for each virtual machine running in the corresponding physical node;
• and
• when the second type of memory distributed to the virtual machine running in the corresponding physical node exceeds the capacity of the second type of memory allocated to the corresponding physical node, reducing the second type of memory of the virtual machine of which performance is reduced the least when the second type of memory is reduced by the unit capacity set for each virtual machine, by the unit capacity.

A method for memory integrated management in an apparatus for memory integrated management of a cluster system including a plurality of physical nodes connected to each other by a network, the method comprising:

• determining one of the plurality of physical nodes as a node to place a new virtual machine;
• allocating the first type of memory allocated to the one physical node to the new virtual machine as much as the memory capacity required by the new virtual machine;
• and
• distributing the second type of memory to a plurality of virtual machines running on the plurality of physical nodes by integrating and managing the second type of memory allocated to each of the plurality of physical nodes,
• wherein the memory of the second type has a faster access speed than that of the first type of memory,
• wherein the distributing the second type of memory includes:
• distributing the second type of memory allocated to each physical node to virtual machines running in the corresponding physical node, respectively;
• and
• additionally distributing the remaining second type of memory in the cluster system to at least one virtual machine among the plurality of virtual machines,
• wherein the additionally distributing includes:
• increasing the second type of memory by a unit capacity to a virtual machine of which performance is increased the most when the second type of memory is increased by the unit capacity among the plurality of virtual machines running in the plurality of physical nodes;
• and
• repeating the increasing step until all of the remaining second type of memory in the cluster system is distributed.

An apparatus for memory integrated management in a cluster system including a plurality of physical nodes connected to each other by a network, the apparatus comprising:

• a high-integration memory distributor that integrates and manages a first type of memory allocated to the plurality of physical nodes, and distributes the first type of memory allocated to the corresponding physical node to virtual machines each executed in the plurality of physical nodes;
• and
• a high-performance memory distributor that integrates and manages a second type of memory allocated to the plurality of physical nodes, respectively, and distributes the second type of memory to a plurality of virtual machines in the cluster system in a direction that can maximize the average performance of the plurality of virtual machines,
• wherein the memory of the second type has a faster access speed than that of the first type of memory,
• wherein the high-integration memory distributor distributes the first type of memory of as much as the memory capacity required by a new virtual machine when the new virtual machine is generated,
• wherein the high-integration memory distributor selects a physical node in which the amount of the first type of memory available from among the plurality of physical nodes is greater than the memory capacity required for the new virtual machine, and determines a physical node having the highest average performance of the running virtual machines running on each of the selected physical nodes among the selected physical nodes, as a node on which to place the new virtual machine.

An apparatus for memory integrated management in a cluster system including a plurality of physical nodes connected to each other by a network, the apparatus comprising:

• a high-integration memory distributor that integrates and manages a first type of memory allocated to the plurality of physical nodes, and distributes the first type of memory allocated to the corresponding physical node to virtual machines each executed in the plurality of physical nodes;
• and
• a high-performance memory distributor that integrates and manages a second type of memory allocated to the plurality of physical nodes, respectively, and distributes the second type of memory to a plurality of virtual machines in the cluster system in a direction that can maximize the average performance of the plurality of virtual machines,
• wherein the memory of the second type has a faster access speed than that of the first type of memory,
• wherein the high-performance memory distributor distributes the second type of memory allocated to each physical node to virtual machines running in the corresponding physical node, respectively, and additionally distributes the remaining second type of memory in the cluster system to at least one virtual machine among the plurality of virtual machines,
• wherein the high-performance memory distributor distributes the memory of the second type of as much as the memory capacity required to guarantee the performance of the virtual machine for each virtual machine running in the corresponding physical node, and reduces the second type of memory of the virtual machine of which performance is reduced the least when the second type of memory is reduced by the unit capacity set for each virtual machine, by the unit capacity when the second type of memory distributed to the virtual machine running in the corresponding physical node exceeds the capacity of the second type of memory allocated to the corresponding physical node.

An apparatus for memory integrated management in a cluster system including a plurality of physical nodes connected to each other by a network, the apparatus comprising:

• a high-integration memory distributor that integrates and manages a first type of memory allocated to the plurality of physical nodes, and distributes the first type of memory allocated to the corresponding physical node to virtual machines each executed in the plurality of physical nodes;
• and
• a high-performance memory distributor that integrates and manages a second type of memory allocated to the plurality of physical nodes, respectively, and distributes the second type of memory to a plurality of virtual machines in the cluster system in a direction that can maximize the average performance of the plurality of virtual machines,
• wherein the memory of the second type has a faster access speed than that of the first type of memory,
• wherein the high-performance memory distributor distributes the second type of memory allocated to each physical node to virtual machines running in the corresponding physical node, respectively, and additionally distributes the remaining second type of memory in the cluster system to at least one virtual machine among the plurality of virtual machines,
• wherein when the remaining second type of memory in the cluster system is present, the high-performance memory distributor increases the second type of memory by a unit capacity to a virtual machine of which performance is increased the most when the second type of memory is increased by the unit capacity among the plurality of virtual machines running in the plurality of physical nodes.