EP0301275A2 — Logical resource partitioning of a data processin…

Claim 1

Independent Claim 501 words

Logical resource partitioning method for a data processing system having system resources including one or more real CPUs, a system main storage (MS), an optional system extended storage (ES), and a plurality of channels for connecting the system to a plurality of I/O devices and their device control units and their subchannels (SCHs), comprising: providing a host hypervisor coded entity (host) for controlling the operations of the system, and having access to tables and control blocks containing assignments of system resources, generating a system partition table having a plurality of partition definition sets for defining a plurality of partitions in the system, each set having entries for specifying a subset of the system resources assigned to a partition of the system, a unique partition identifier (AZN) assigned to each definition set, each set assigned a contiguous system main storage (MS) zone located anywhere in a system main storage (MS) and bounded by upper and lower limit MS addresses, each set optionally assigned a contiguous optional system extended storage (ES) zone located anywhere in an optional system extended storage (ES) and bounded by upper and lower limit optional system extended storage (ES) addresses, and each set assigned none or any one or plural subchannels (SCHs) in any order, each assigned SCH containing the unique partition identifier (AZN) of the partition to which the SCH is assigned, each assigned SCH also containing an identifier of each I/O interrupt queue assigned to the subchannel (SCH), also providing a plurality of state description control blocks (SDs) for representing logical CPUs in the system, each state description (SD) containing fields with parameters defining a logical CPU and containing an assigned unique partition identifier (AZN) field to represent the assignment of the logical CPU to a partition, binding logical CPU(s) to selected partitions and to respective control programming systems (guests) in system storage by entering an assigned unique partition identifier (AZN) and program status word (PSW) parameters for a control programming system into the subchannels (SDs), the host controlling the execution of a start interpretive execution (SIE) instruction for accessing the subchannel (SD) of a guest bound to a partition by an assigned unique partition identifier (AZN), dispatching a logical CPU of a guest on a selected real CPU in the data processing system by the host execution of the start interpretive execution (SIE) instruction to invoke a guest to use the resources in the partition bound to the guest by the assigned unique partition identifier (AZN) in the state description control block (SD) of the guest, routing each I/O interrupt signal from a device control unit to an I/O processor that enqueues the subchannel (SCH) onto an I/O interrupt queue assigned to the subchannel (SCH) to provide a pending I/O interruption in response to the I/O interrupt signal, having a pass-through indicator in each subchannel (SCH) to enable a guest to take a pending I/O interruption only for subchannels (SCHs) having the same unique partition identifier (AZN) as the partition to which the guest is assigned.

Claim 2

Dependent Claim 501 words

Logical resource partitioning method for a data processing system having system resources including one or more real CPUs, a system main storage (MS), an optional system extended storage (ES), and a plurality of channels for connecting the system to a plurality of I/O devices and their device control units and their subchannels (SCHs), comprising: providing a host hypervisor coded entity (host) for controlling the operations of the system, and having access to tables and control blocks containing assignments of system resources, generating a system partition table having a plurality of partition definition sets for defining a plurality of partitions in the system, each set having entries for specifying a subset of the system resources assigned to a partition of the system, a unique partition identifier (AZN) assigned to each definition set, each set assigned a contiguous system main storage (MS) zone located anywhere in a system main storage (MS) and bounded by upper and lower limit MS addresses, each set optionally assigned a contiguous optional system extended storage (ES) zone located anywhere in an optional system extended storage (ES) and bounded by upper and lower limit optional system extended storage (ES) addresses, and each set assigned none or any one or plural subchannels (SCHs) in any order, each assigned SCH containing the unique partition identifier (AZN) of the partition to which the SCH is assigned, each assigned SCH also containing an identifier of each I/O interrupt queue assigned to the subchannel (SCH), also providing a plurality of state description control blocks (SDs) for representing logical CPUs in the system, each state description (SD) containing fields with parameters defining a logical CPU and containing an assigned unique partition identifier (AZN) field to represent the assignment of the logical CPU to a partition, binding logical CPU(s) to selected partitions and to respective control programming systems (guests) in system storage by entering an assigned unique partition identifier (AZN) and program status word (PSW) parameters for a control programming system into the subchannels (SDs), the host controlling the execution of a start interpretive execution (SIE) instruction for accessing the subchannel (SD) of a guest bound to a partition by an assigned unique partition identifier (AZN), dispatching a logical CPU of a guest on a selected real CPU in the data processing system by the host execution of the start interpretive execution (SIE) instruction to invoke a guest to use the resources in the partition bound to the guest by the assigned unique partition identifier (AZN) in the state description control block (SD) of the guest, routing each I/O interrupt signal from a device control unit to an I/O processor that enqueues the subchannel (SCH) onto an I/O interrupt queue assigned to the subchannel (SCH) to provide a pending I/O interruption in response to the I/O interrupt signal, having a pass-through indicator in each subchannel (SCH) to enable a guest to take a pending I/O interruption only for subchannels (SCHs) having the same unique partition identifier (AZN) as the partition to which the guest is assigned.

Claim 3

Dependent Claim 95 words

Logical resource partitioning method as defined in claim 1, further comprising: relocating all channel addresses used by a subchannel (SCH) into a MS zone assigned to the partition indicated by the unique partition identifier (AZN) assigned to the subchannel (SCH), and checking channel accesses to a system main storage (MS) in relation to the boundaries of the system main storage (MS) zone assigned to the partition identified by the unique partition identifier (AZN) assigned to the subchannel (SCH), and indicating any violation by any channel access outside of the assigned system main storage (MS) zone.

Claim 4

Dependent Claim 66 words

Logical resource partitioning method as defined in claim 1, further comprising: providing a partition pointer table for accessing the partition definition sets in any order, and receiving assignments of unique partition identifiers (AZNs) for enabling any unique partition identifier (AZN) value to be assigned to any partition definition table, dispatching on a real CPU a highest priority logical CPU of the logical CPUs ready to execute.

Claim 5

Dependent Claim 84 words

Logical resource partitioning method as defined in claim 2, further comprising: enabling a dispatched guest on a real CPU to detect the existence of I/O interruptions for one or more subchannels (SCHs) assigned to another guest which is assigned to another partition and not currently dispatched on a real CPU, alerting the host after the dispatched guest receives an I/O interruption for a subchannel (SCH) assigned to the other guest, and the host making ready for dispatch the other guest having the I/O interruption.

Claim 6

Dependent Claim 76 words

Logical resource partitioning method as defined in claim 4, further comprising: comparing a dispatching priority of the currently dispatched guest with a dispatching priority of any other guest made ready for dispatch by the host to determine which guest is to be dispatched next on the real CPU of the currently dispatched guest, terminating the dispatch of the currently dispatched guest for a dispatch of another guest having a dispatching priority higher than the current guest.

Claim 7

Dependent Claim 71 words

Logical resource partitioning method as defined in claim 5, further comprising: intercepting the execution of the dispatched guest when an I/O interruption becomes pending for a subchannel (SCH) assigned to an alertable other guest when the dispatched guest is enabled for the I/O interruption, invoking host processing on the CPU of the intercepted guest to detect the I/O interruption of the alertable guest and make the alertable guest ready for dispatching.

Claim 8

Dependent Claim 63 words

Logical resource partitioning method as defined in claim 6, processing by the host further comprising: detecting if there are still other pending I/O interruptions for still other guest(s) enabled for alerting by the intercepted guest and having a pending I/O interruption(s), making each of the still other detected guest(s) ready for dispatching, and dispatching the highest-priority of the guests made ready for dispatching.

Claim 9

Dependent Claim 244 words

Logical resource partitioning arrangement or means for the method defined in claim 1, comprising: having indicators in the subchannels (SCHs) of the assignment of the subchannels (SCHs) to hardware and guest I/O interruption queues, providing a guest I/O interruption enablement mask (GIEM) in the state description control block (SD) of each partition-assigned guest for selectively indicating the guest's I/O interruption enablement state for handling pending interruptions on guest I/O interruption queues, providing an effective I/O interruption enablement mask (EIEM) for use by the host and any dispatched partition-assigned guest to selectively indicate the I/O interruption enablement states for the hardware I/O interruption queue(s) usable by the guest including: at least one hardware queue for receiving the I/O interruptions of the dispatched guest and other hardware queue(s) for receiving the I/O interruption(s) of one or more other guests alertable by the dispatched guest, processing and loading the effective I/O interruption enablement mask (EIEM) into a host control register of a real CPU on which the guest is being dispatched, setting on a disabling field in the state description control block (SD) of the guest when the guest is found selectively disabled for handling an I/O interruption by a subchannel assigned to the guest, including setting on the disabling field after the guest makes a change in its guest I/O interruption enablement mask (GIEM), the on state of the disabling field preventing the guest from handling any I/O interruption for any subchannel assigned to the guest.

Claim 10

Dependent Claim 97 words

Logical resource partitioning arrangement or means as defined in claim 8, further comprising: alerting the host whenever the dispatched guest receives an I/O interruption for a subchannel assigned to another partition enabled by the effective I/O interruption enablement mask (EIEM) in the CPU of the dispatched guest, even though the state description control block (SD) disabling field of the dispatched guest is set off or the effective I/O interruption enablement mask (EIEM) in the CPU of the dispatched guest is disabling the dispatched guest in order to prevent the dispatched guest from handling its own I/O interruptions.

Claim 11

Dependent Claim 119 words

Logical resource partitioning arrangement or means as defined in claim 8, executing a test pending instruction (TPI) comprising: testing the state of an state description control block (SD) queuing indicator field for the dispatched guest to determine if the dispatched guest has any I/O pending interruptions for any subchannels (SCHs) assigned to the dispatched guest in any guest queue, intercepting the processing of the test pending instruction (TPI) if the SD queuing indicator field of the dispatched guest is on, but invoking the host to execute a test pending instruction (TPI) instruction for the guest to determine if there is any pending I/O interruption for the guest on a hardware I/O interruption queue used for receiving guest I/O interruptions.

Claim 12

Dependent Claim 144 words

Logical resource partitioning arrangement or means as defined in claim 10, further processing the test pending instruction (TPI) comprising: providing a partition mode field in each state description control block (SD) to indicate if the respective guest is assigned to a partition-assigned guest, providing an I/O interrupt pending register (IPR) in each CPU having an I/O interrupt pending register (IPR) bit positions that respectively indicate if any I/O interruption is pending in an associated hardware I/O interrupt queue of which at least one queue is assigned for the use of each partition-assigned guest, setting on a condition code (CC) in a program status word (PSW) of the guest to a first predetermined state when the guest's respective I/O interrupt pending register (IPR) bit position indicates no I/O interruption is pending for the guest, and ending the operation for the test pending instruction (TPI) instruction.

Claim 13

Dependent Claim 107 words

Logical resource partitioning arrangement or means as defined in claim 11, further processing the test pending instruction (TPI) comprising: dequeuing an I/O interruption if the guest is in partitioned mode and the guest's I/O interrupt pending register (IPR) bit position indicates an I/O interruption is pending for a subchannel assigned to the guest, testing if a pass-through field is set on in the interrupting subchannel, intercepting the processing of the test pending instruction (TPI) instruction if the pass-through field is set off in the interrupting subchannel, and invoking processing by the host to set on the pass-through field to enable the guest to handle the I/O interruption.

Claim 14

Dependent Claim 126 words

Logical resource partitioning arrangement or means as defined in claim 12, the processing for a test pending instruction (TPI) further comprising: testing the guest I/O interruption enablement mask (GIEM) of the dispatched guest to determine whether the guest is selectively enabled for a pending I/O interruption when the state of the pass-through field indicates the guest may handle the I/O interruption, storing an I/O interruption code in the state description control block (SD) of the dispatched guest when the testing operation finds the guest is selectively enabled, setting on a condition code (CC) in the program status word (PSW) of the guest to a second predetermined state when the storing operation has stored the I/O interruption, and ending the operation for the test pending instruction (TPI).

Claim 15

Dependent Claim 80 words

Logical resource partitioning arrangement or means as defined in claim 13, the processing for a test pending instruction (TPI) further comprising: intercepting the guest processing of the TPI instruction by storing an intercept code if the testing operation finds the dispatched guest is disabled by its guest I/O interruption enablement mask (GIEM) from handling the I/O interruption, and invoking processing by the guest to enable the guest I/O interruption enablement mask (GIEM) for the guest to handle the I/O interruption.

Claim 16

Dependent Claim 194 words

Logical resource partitioning arrangement or means as defined in Claim 8, executing a test subchannel (TSCH) instruction issued by a guest dispatched on a CPU to determine if the subchannel has any I/O interruption pending for the guest, comprising: testing the state of the partition mode field in the SD of the dispatched guest to determine if the partition mode field indicates the guest is a partition-assigned guest, testing if the I/O interrupting subchannel is partition-assigned to the dispatched guest, and testing a pass-through field in the subchannel to determine if I/O interruptions of the subchannel can be handled by partition-assigned guests, intercepting the processing of the test subchannel (TSCH) instruction if any of the testing operations find the state description control block (SD) partition-mode field of the dispatched guest is set off, or the guest I/O interruption queue assignment in the subchannel being tested is not for the partition assigned to the dispatched guest, or if the pass-through field is set off in the subchannel being tested, invoking the host to execute a test subchannel (TSCH) instruction for the guest to determine if the subchannel has any pending I/O interruption for the guest.

Claim 17

Dependent Claim 168 words

Logical resource partitioning arrangement or means as defined in Claim 15, executing a test subchannel (TSCH) instruction issued by a guest dispatched on a CPU to determine if the subchannel has any I/O interruption pending for the guest, comprising: dequeuing any I/O interruption found for the subchannel being tested when the testing operations find the SD partition-mode field of the dispatched guest is set on, and the guest I/O interruption queue assignment in the subchannel being tested is for the partition assigned to the dispatched guest, and the pass-through field is set on in the subchannel being tested, testing if any I/O exception condition exists to the execution of the test subchannel (TSCH) instruction and if an exception condition is found to exist, setting off the pass-through field and setting a predetermined intercept code in the state description control block (SD) of the dispatched guest, intercepting the processing of the test subchannel (TSCH) instruction, and invoking the host to execute a test subchannel (TSCH) instruction for the guest.

Claim 18

Dependent Claim 195 words

Logical resource partitioning arrangement or means as defined in claim 16, executing a test subchannel (TSCH) instruction issued by a guest dispatched on a CPU to obtain subchannel status for the guest, comprising: obtaining and storing an information response block (IRB) containing subchannel status in guest storage if no exception condition is found by the testing operation, providing an I/O status pending field in each subchannel to indicate if any I/O status is pending for the subchannel, and copying the state of the field into the information response block (IRB), testing the I/O status pending field in the information response block (IRB) to determine if any I/O status is pending for the subchannel, setting on a condition code (CC) in the program status word (PSW) of the guest to a first predetermined state when the I/O status pending field indicates an I/O status is pending for the subchannel being tested, setting on a condition code (CC) program status word (PSW) of the guest to a second predetermined state when the I/O status pending field indicates no I/O status is pending for the subchannel being tested, and ending the operation for the test subchannel (TSCH) instruction.

Claim 19

Dependent Claim 304 words

Logical resource partitioning method as defined in claim 1, characterized by providing in the state description control block (SD) of each partition-assigned guest an alerting I/O interruption enablement mask an unique partition identifier (AZM) for selectively indicating which of plural hardware I/O interrupt queues can be monitored by the guest for alerting other guests, and a guest I/O interruption enablement mask (GIEM) for selectively indicating the enablement state for the guest's I/O interrupt queues, providing an effective I/O interruption enablement mask (EIEM) for use by a dispatched partition-assigned quest to selectively indicate the guest's enablement states: for hardware I/O interruption queue(s) on which I/O interruptions can be handled by the dispatched guest and for hardware I/O interruption queues(s) on which I/O interruption(s) for other guest(s) can cause alerting by the dispatched guest, processing and loading the effective I/O interruption enablement mask (EIEM) into a control register of the real CPU of the guest, selecting a pending I/O interruption on any hardware queue that is enabled by the effective I/O interruption enablement mask (EIEM) for examination by the dispatched guest, testing if the I/O interruption is enabled by the unique partition identifier (AZM) for another guest, identifying the unique partition identifier (AZN) in the subchannel (SCH) providing the I/O interruption if the testing operation finds the I/O interruption is for another guest being alerted, storing the unique partition identifier (AZN) of the alerting guest in the state description control block (SD) of the dispatched guest, setting an intercept code into the state description control block (SD) of the dispatched guest to indicate an alerting intercept condition, and intercepting the guest's dispatch to invoke the host to examine the unique partition identifier (AZN) of the alerting guest in the state description control block (SD) of the dispatched guest to make the alerting guest ready for dispatching.

Claim 20

Dependent Claim 118 words

Logical resource partitioning method as defined in claim 18, the host invocation for the alerting guest further comprising: determining the unique partition identifier (AZN) for any other guest enabled in the unique partition identifier (AZM) in the state description control block (SD) of the dispatched guest, executing a test pending zone interrupt (TPZI) instruction for each other guest's unique partition identifier (AZN) identified by the determining operation to determine if any pending I/O interruption exists for any other guest enabled by the unique partition identifier (AZM) of the dispatched guest, making ready for dispatch each other guest found by the executing operation to have a pending I/O interruption, and dispatching the ready guest having a highest dispatching priority.

Claim 21

Dependent Claim 410 words

Logical resource partitioning method as defined in claim 1, executing a test pending zone interrupt (TPZI) instruction for determining if any pending I/O interruption exists for a guest identified by a specified unique partition identifier (AZN), characterized by finding that no exception conditions exist at the beginning of execution of the test pending zone interrupt (TPZI) that may interfere with its execution, providing in a first general register associated with the test pending zone interrupt (TPZI): a control field to indicate whether a first or second type of response is to be provided by the test pending zone interrupt (TPZI), an unique partition identifier (AZN) field specifying the unique partition identifier (AZN) of a guest being examined for I/O interruptions, and a guest I/O interruption enablement mask (GIEM) field containing a guest I/O interruption enablement mask (GIEM) for the guest, providing in a second general register associated with the test pending zone interrupt (TPZI) an address of a storage location where a response to the test pending zone interrupt (TPZI) is to be stored, having an I/O interrupt pending register (IPR) in each CPU with respective I/O interrupt pending register (IPR) bit positions that respectively indicate whether any I/O interruption is pending in each of respectively associated hardware I/O interrupt queues, at least one assigned I/O interrupt pending register (IPR) bit position being assigned for use by a guest regarding the guest's use of a corresponding hardware queue that receives the guest's I/O interruptions, examining the state of the assigned I/O interrupt pending register (IPR) bit position for the guest to find if any pending I/O interrupt exists for the guest, and setting a condition code in the host PSW to a first value to indicate when no pending I/O interruption exists for the guest being examined, storing I/O information about any pending I/O interruption when the first general register indicates a first type response is to be provided, and setting the condition code in the host PSW to a first value to indicate the second type response has been provided, dequeuing any pending I/O interruption when the first general register indicates a second type response is to be provided, saving the state of a pass-through field in the interrupting subchannel, storing an I/O interrupt code for the pending interrupt at the address in the second general register, and setting a condition code in the host PSW to a third value to indicate the second type response has been provided.

Claim 22

Dependent Claim 208 words

Logical resource partitioning arrangement or means as defined in claim 8, characterized by a dispatched guest generating an intercept code and putting it into the guest's state description control block (SD) to indicate the guest cannot process its I/O interruption because conditions exist preventing the guest from currently handling the interrupt, intercepting the guest processing by passing control of the CPU to the host in response to the guest indicating an intercept code, and the host processing including: testing a pass-through field in the interrupting subchannel to find if it is in a state allowing partition-assigned guests to handle the subchannel's interruptions, moving the pending I/O interruption from the hardware queue to a guest queue assigned in the subchannel to receive interruptions of the subchannel for the guest, setting on the dispatched guest's state description control block (SD) disabling field to indicate the guest is currently disabled from handling a pending I/O interruption for a subchannel assigned to the guest and being stored in a guest's I/O interruption queue while the guest is disabled, the host passing control of the CPU back to the guest to continue guest processing when the pass-through field is set to a state allowing partition-assigned guest's to handle the interruptions of the subchannel.

Claim 23

Dependent Claim 253 words

Logical resource partitioning arrangement or means as defined in claim 8, host processing for a diagnose compare and swap instruction (DCSI) for setting a pass-through field in a subchannel, characterized by finding that no exception conditions exist at the beginning of execution of the diagnose compare and swap instruction (DCSI) that may interfere with its execution to continue the execution of the diagnose compare and swap instruction (DCSI), setting on a lock field in the subchannel, testing if the pass-through field in a third operand is set on, and if set on testing if the subchannel is on an I/O interruption queue, and dequeuing the subchannel if found on a queue, comparing the status pending field and the pass-through field in the first operand with corresponding status pending and pass-through fields in the subchannel, setting the pass-through field in the subchannel to the new value contained in the third operand of the diagnose compare and swap instruction (DCSI) if equality is found by the comparing operation, but when inequality is found by the comparing operation, setting the first operand from the current status pending and pass-through fields in the subchannel, generating a condition code in the program status word (PSW) of a guest for which the diagnose compare and swap instruction (DCSI) is being executed by setting the condition code to a first value when equality is found, and setting the condition code to a second value when inequality is found, and setting off the lock field after the condition code is generated.

Claim 24

Dependent Claim 114 words

Logical resource partitioning method as defined in claim 1, channel address relocation for channel accesses in an assigned zone in a system mains storage (MS), characterized by providing in each subchannel any unique partition identifier (AZN) assigned to the subchannel and the upper and lower limit system mains storage (MS) addresses assigned to the unique partition identifier (AZN), adding the lower limit address to each channel address to be accessed in the system main storage (MS) to obtain a resulting relocation address, and comparing the resulting relocation address to the upper limit address, suppressing the system main storage (MS) access by the resulting relocation address if it is greater than the upper limit address.

Claim 25

Dependent Claim 368 words

Logical resource partitioning method as defined in claim 1, CPU address relocation for CPU accesses in assigned zones in the system main storage (MS), the assignment of the system main storage (MS) to a partition characterized by generating a host segment table and a plurality of guest segment tables, a segment table being provided for each operational guest in the data processing system and a host segment table being provided for the host, setting a guest control register to an system main storage (MS) address of the guest's segment table, and setting a host control register to an system main storage (MS) address of the host segment table, allocating a contiguous set of page frames in the system main storage (MS) between the upper and lower limit system main storage (MS) addresses assigned to the respective unique partition identifier (AZN) of each guest, generating at least one host page table for each guest, the host page table(s) for the guest containing a contiguous set of entries corresponding to the page frames allocated to the respective guest, setting the page table entries in a guest's page table(s) to the respective absolute addresses of the page frames in a system main storage (MS) allocated to the guest, and setting to a valid state each of the guest's page frame entries, also setting a segment table entry in the host's segment table to the absolute address of each host page table for the guest in a system main storage (MS) while maintaining the order of segment table entries for the guest in the host segment table consistent with the order of contiguous page frames in the guest's page tables if there is more than one host segment table entry for the guest, on the dispatch of a guest, loading the host segment table address into the host control register in the CPU and loading the guest segment table address from the guest's state description control block (SD) into a guest control register in a CPU on which the guest is being dispatched, performing address translations for the guest absolute addresses using the host segment table entry(s) and page table(s) for the guest accessed through the host control register in the CPU.

Claim 26

Dependent Claim 105 words

Logical resource partitioning method as defined in claim 24, address relocation for a page in an assigned zone in expanded storage (ES) characterized by adding a guest-provided page number to be access in an optional system extended storage (ES) to the lower limit page number obtained from the partition definition set for the respective guest to obtain a resulting relocation page number, comparing the resulting relocation page number to the upper limit page number obtained from the partition definition set, suppressing the an optional system extended storage (ES) access by the resulting relocation page number if it is greater than the upper limit page number.

Claim 27

Dependent Claim 362 words

Logical resource partitioning method as defined in claim 1, enabling the sharing of an I/O device and its control unit between a plurality of partitions, characterized by having an I/O processor connected to a plurality of I/O subchannels and a plurality of I/O channels, in which the plurality of subchannels need not be equal to the plurality of channels, each channel being capable of controlling a transmission of a stream of data between channel-program designated addresses in a system main storage (MS) and the I/O device and its control unit, and each subchannel being capable of controlling the operation of an I/O device, each channel also being capable of providing a separate transmission path between associated I/O devices and system (MS), connecting one side of an N-way switch to the control unit for the I/O device, and connecting another side of the N-way switch to N number of I/O channels for providing N-way switching of the device and its control unit to any selected one of the N number of channels, the device switch operating under manual or program control to connect the I/O device and its control unit to any channel connected to the device switch, providing up to C number of copies of a subchannel that controls the operation of the I/O device, in which all C number of the subchannel copies designate the same I/O device, selecting C number of different partitions, and assigning each of the C copies of the subchannel to a different one of the partitions, the selected partitions respectively having C number of different zones in a system main storage (MS), and further assigning each of the C number of subchannel copies to one or more of the N number of channels connected to the device switch to enable the device switch to operate under manual or program control to connect the I/O device to any of the C number of different zones in a system main storage (MS) through a device switch selected channel in order to allow data transmission between the I/O device and any one of the C number of zones as selected by the device switch under manual or program control.

Logical resource partitioning of a data processing system

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag