Abstract: Provided is a computer system including I/O devices coupled to PCI switches coupled via interfaces of a plurality of servers, and a management block for managing configurations of the PCI switches. The management block is configured to: set, to the PCI switch, a first access path including a virtual bridge coupling the interface of a first server and a virtual switch, and a virtual bridge coupling the I/O device and the virtual switch; set, to the PCI switch, a second access path including a virtual bridge coupling the interface of a second server of the plurality of servers and a virtual switch, and a virtual bridge coupling the I/O device used by the first server and the virtual switch; disable mapping of the second access path between the I/O device and the virtual bridge; and instruct the second server to make access to the I/O device.

Abstract: When a subject of access of a transaction from an IO device is not any resource allocated to a logical partition to which the device having issued the transaction belongs, a report as an error is sent to a CPU, while the transaction is finished on the IO bus. To prevent a transaction between IO devices from gaining access to any resource in another logical partition, one access permission bit is provided for each combination of all the IO devices, and the access is permitted only when the bit has a predetermined value. A reset signal is provided by IO slot so that only an IO slot allocated to a specific logical partition can be reset without affecting any other logical partition. A transaction issued from an IO device in one logical partition is prevented from gaining access to a resource in another logical partition, while proper error handling can be performed.

Abstract: The physical server includes a hypervisor for managing an association between the virtual server and the I/O device allocated to the virtual server. The I/O switch includes: a setting register for retaining a request to inhibit a transaction from being issued from the I/O device to the virtual server; a Tx inhibition control module for performing an inhibition of the transaction from the I/O device to the virtual server, and guaranteeing a completion of a transaction from the I/O device issued before the inhibition; a virtualization assist module for converting an address of the virtual server into an address within a memory of the physical server; and a switch management module for managing a configuration of the I/O switch.

Abstract: Disclosed herewith is a composite type computer system that can assure that a PCI tree to be allocated to a computer is configured completely before the computer is powered. The composite type computer system includes a PCI switch that connects plural computers through PCI interfaces; plural PCI devices connected to the PCI switch; a system controller that controls the computers; and a PCI manager that controls allocation of the PCI devices to the computers.

Abstract: Provided is a method of controlling a virtual computer system in which a physical computer includes a plurality of physical CPUs that is switchable between a sleep state and a normal state, and a virtualization control unit divides the physical computer into a plurality of logical partitions to run a guest OS in each of the logical partitions and controls allocation of resources of the physical computer to the logical partitions, causes the virtualization control unit to: receive an operation instruction for operating the logical partitions; and if the operation instruction is for deleting a virtual CPU from one of the logical partitions, delete this virtual CPU from a table for managing virtual CPU-physical CPU allocation and put, if the deleting leaves no virtual CPUs allocated to one of the physical CPUs that has been allocated the deleted virtual CPU, this one of the physical CPUs into the sleep state.

Abstract: In order to provide an interface of acquiring physical position information of an I/O device on a virtual machine monitor having an exclusive allocation function of the I/O device and optimize allocation of a resource to a virtual server by using the acquired physical position information, a virtual machine monitor includes an interface of allocating a resource in accordance with a given policy (a parameter of determining to which a priority is given in distributing resources) for an I/O device, a CPU NO., and a memory amount request to guest OS. Further, the virtual machine monitor includes an interface of pertinently converting physical position information of the resource allocated by the virtual machine monitor to notice to guest OS.

Abstract: When a subject of access of a transaction from an IO device is not any resource allocated to a logical partition to which the device having issued the transaction belongs, a report as an error is sent to a CPU, while the transaction is finished on the IO bus. To prevent a transaction between IO devices from gaining access to any resource in another logical partition, one access permission bit is provided for each combination of all the IO devices, and the access is permitted only when the bit has a predetermined value. A reset signal is provided by IO slot so that only an IO slot allocated to a specific logical partition can be reset without affecting any other logical partition. A transaction issued from an IO device in one logical partition is prevented from gaining access to a resource in another logical partition, while proper error handling can be performed.

Abstract: The present invention provides a machine system that enables the arbitration of IO accesses and band control based on the priority of virtual servers while curbing performance overhead during IO sharing among the virtual servers. A virtual machine system including a CPU, a memory, and an IO interface includes a hypervisor that generates plural virtual servers, and an IO controller that controls the IO interface. The IO controller includes: a DMA receiving unit that receives DMA requests from the IO interface; a decoder that decodes received DMA requests and locates the corresponding virtual servers; a DMA monitoring counter that monitors DMA processing status for each of the virtual servers; a threshold register set in advance for each of the virtual servers; and a priority deciding unit that compares the DMA monitoring counter and the value of the threshold register, and based on processing priority obtained as a result of the comparison, decides the priority of processing of the received DMA requests.

Abstract: When the entire system is split into plural partitions on chipsets connecting plural processors, IO hubs, and memory controllers, and an OS is operating on each of the partitions, the present invention prevents a failure in a partition from propagating to other partitions. Based on address information or issuer information included in a packet inputted to a chipset, a partition from which the packet was issued is identified, and an identified partition identifier is added to the packet. Based on the added partition identifier, a partition initializing part selectively deletes the packet issued from the partition in which a failure occurred, thereby preventing the influence of the failure in the failure-causing partition from propagating to other partitions.

Abstract: Disclosed herein is a computer system provided with a mechanism for connecting a single port disk to an active server and the disk to a standby server when in a fail-over processing. An “add_pci” command issued from a clustering program is used to let a control program change the allocation of a PCI slot while an interruption signal issued to a standby server permits an ACPI processing routine to hot-add a PCI card that includes the disk unit on the subject guest OS.

Abstract: In a multi-processor system, counting snoop results bottlenecks the broadcast-based snoop protocol. The directory-based protocol delays the latency when remote node caches data. There is a need for shortening the memory access latency using a snoop and cache copy tag information. When the local node's cache copy tag information is available, the memory access latency can be shortened by omitting a process to count snoop results. When memory position information is used to update the cache copy tag during cache replacement, it is possible to increase a ratio to hit a copy tag during reaccess from the local node.

Abstract: Provided is a computer system in which an I/O card is shared among physical servers and logical servers. Servers are set in advance such that one I/O card is used exclusively by one physical or logical server, or shared among a plurality of servers. An I/O hub allocates a virtual MM I/O address unique to each physical or logical server to a physical MM I/O address associated with each I/O card. The I/O hub keeps allocation information indicating the relation between the allocated virtual MM I/O address, the physical MM I/O address, and a server identifier unique to each physical or logical server. When a request to access an I/O card is sent from a physical or logical server, the allocation information is referred to and a server identifier is extracted from the access request. The extracted server identifier is used to identify the physical or logical server that has made the access request.

Abstract: A case occurs in which a preceding transaction to be retried cannot be retried as a result of snooping. If the snoop result is waited for so that a retry decision may be made after the result of snoop has been obtained, latency is prolonged to urge the pipeline to have a variable length, thus complicating the logic. A transaction determined to be retried in the phase of issue of a request is discriminated from a transaction in course of issue and when a transaction representing a starvation protection object sequentially competes twice with the transaction in course of retry decision, the transaction of starvation protection object is issued to thereby eliminate starvation.

Abstract: When a subject of access of a transaction from an IO device is not any resource allocated to a logical partition to which the device having issued the transaction belongs, a report as an error is sent to a CPU, while the transaction is finished on the IO bus. To prevent a transaction between IO devices from gaining access to any resource in another logical partition, one access permission bit is provided for each combination of all the IO devices, and the access is permitted only when the bit has a predetermined value. A reset signal is provided by IO slot so that only an IO slot allocated to a specific logical partition can be reset without affecting any other logical partition. A transaction issued from an IO device in one logical partition is prevented from gaining access to a resource in another logical partition, while proper error handling can be performed.

Abstract: When a subject of access of a transaction from an IO device is not any resource allocated to a logical partition to which the device having issued the transaction belongs, a report as an error is sent to a CPU, while the transaction is finished on the IO bus. To prevent a transaction between IO devices from gaining access to any resource in another logical partition, one access permission bit is provided for each combination of all the IO devices, and the access is permitted only when the bit has a predetermined value. A reset signal is provided by IO slot so that only an IO slot allocated to a specific logical partition can be reset without affecting any other logical partition. A transaction issued from an IO device in one logical partition is prevented from gaining access to a resource in another logical partition, while proper error handling can be performed.

Abstract: The present invention makes coordination of I/O access operations of operating systems independently running in logical partitions. In a data processing system comprising processors, a main memory, I/O slots, and a node controller, wherein the processors, the main memory, and the I/O slots are interconnected via the node controller and divided into a plurality of partitions in which individual operating systems are run simultaneously, the node controller includes a logical partition arbitration unit which stores information as to whether each logical partition is using an I/O slot and controls access from each logical partition to an I/O slot by referring to the information thus stored.

Abstract: Each node includes a node controller for decoding the control information and the address information for the access request issued by a processor or an I/O device, generating, based on the result of decoding, the cache coherence control information indicating whether the cache coherence control is required or not, the node information and the unit information for the transfer destination, and adding these information to the access request. An intra-node connection circuit for connecting the units in the node controller holds the cache coherence control information, the node information and the unit information added to the access request. When the cache coherence control information indicates that the cache coherence control is not required and the node information indicates the local node, then the intra-node connection circuit transfers the access request not to the inter-node connection circuit inter-connecting the node but directly to the unit designated by the unit information.

Abstract: A sever includes an SVP 206 which is a control device capable of executing processing independent of a CPU, and an interrupt 314 is periodically applied to the CPU. An SVP device driver 303, being triggered by the interrupt 314, acquires an amount of server resources used, such as CPU operating time and memory capacity, and user information used by using an API provided by an OS, and delivers such information to an SVP 206. The SVP 206 transmits resource information thus delivered to an accounting server via an NIC 211. The accounting server accounts an amount of server resource used on a user basis and bills the users according to the quantity of server resource used. With such arrangement, it is possible to accurately grasp the usage status of server resource regardless of server loading, and further, it is possible to execute accounting of server resources and billing based on such accounting regardless of types of OS or CPU.

Abstract: The present invention relates to system construction, and operations management, provided by a plurality of servers in an information processing system, and more particularly to a technology for reducing the burden on a system administrator for system construction and operations management.

Abstract: Disclosed herein is a computer system provided with a mechanism for connecting a single port disk to an active server and the disk to a standby server when in a fail-over processing. An “add_pci” command issued from a clustering program is used to let a control program change the allocation of a PCI slot while an interruption signal issued to a standby server permits an ACPI processing routine to hot-add a PCI card that includes the disk unit on the subject guest OS.