WO2016139749A1 - Système informatique et procédé de commande de stockage - Google Patents

Système informatique et procédé de commande de stockage Download PDF

Info

Publication number
WO2016139749A1
WO2016139749A1 PCT/JP2015/056202 JP2015056202W WO2016139749A1 WO 2016139749 A1 WO2016139749 A1 WO 2016139749A1 JP 2015056202 W JP2015056202 W JP 2015056202W WO 2016139749 A1 WO2016139749 A1 WO 2016139749A1
Authority
WO
WIPO (PCT)
Prior art keywords
storage
path
server computer
page
logical volume
Prior art date
Application number
PCT/JP2015/056202
Other languages
English (en)
Japanese (ja)
Inventor
里山　愛
江口　賢哲
Original Assignee
株式会社日立製作所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社日立製作所 filed Critical 株式会社日立製作所
Priority to PCT/JP2015/056202 priority Critical patent/WO2016139749A1/fr
Publication of WO2016139749A1 publication Critical patent/WO2016139749A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/10—Program control for peripheral devices

Definitions

  • the present invention relates to a technology of a computer system and a storage control method.
  • Flash memory has a higher I / O (Input / Output) performance than an HDD (Hard Disk Drive).
  • HDD Hard Disk Drive
  • SCSI Small Computer System Interface
  • NVMe Non-Volatile Memory Express
  • Non-Patent Document 1 discloses a concept of sharing a namespace (hereinafter referred to as “NS”) from a plurality of hosts.
  • a thin provisioning technology that provides a server with a capacity virtualization volume to which a storage area is dynamically allocated from a capacity pool composed of a plurality of storage areas based on a storage medium.
  • Thin provisioning technology that provides a server with a hierarchical capacity virtualization volume that is a capacity virtualization volume to which a storage area is dynamically allocated from a capacity pool composed of a plurality of storage areas based on different types of storage media.
  • NVM Express 1.1a Specification 8.8 (p.158-159), http://www.nvmexpress.org/wp-content/uploads/NVM-Express-1_1a.pdf
  • the server computer can define a plurality of paths (that is, multipaths) for the logical volume of the storage device by using a predetermined multipath I / O function of the SCSI driver.
  • multipaths that is, multipaths
  • the NVMe driver does not have a multipath I / O function according to the specification of NVMe, the multipath cannot be defined.
  • the capacity pool associated with the hierarchical capacity virtualization volume is based on different types of storage media.
  • HDDs and flash memories are used as the different types of storage media. Since flash memory generally has higher access performance than HDDs, a plurality of storage areas based on flash memory belong to Tier 1 (higher hierarchy), and a plurality of storage areas based on HDD belong to Tier 2 (lower hierarchy).
  • the SCSI driver path is defined for the hierarchical capacity virtualization volume
  • the Tier1 storage area configured by the flash memory is also accessed via the SCSI driver path.
  • I / O scheduler control for optimizing the seek time and the like for the HDD and realizing efficient access is performed during access processing.
  • the storage medium is a flash memory
  • I / O scheduler control becomes an overhead
  • the NVMe driver path is defined for the tiered capacity virtualized volume
  • the Tier 2 storage area configured by the HDD is also accessed via the NVMe driver path. In this case, since the I / O scheduler control is not performed, the access efficiency to the HDD deteriorates.
  • an object of the present invention is to provide a computer system and a storage control method for performing access control suitable for the characteristics of a storage medium to be accessed.
  • a computer system is: A server computer, A first storage device having a first storage medium; A second storage device having a second storage medium different from the first storage medium; A storage controller connected to the server computer via PCI-Express and connected to the first and second storage devices.
  • the storage controller Provide logical volumes, Obtaining information of at least the first or second storage medium assigned to the logical volume; Based on the acquired information, at least one of a first path accessed by a command conforming to the first standard, or a second path accessed by a command conforming to the second standard, based on the acquired information Set one.
  • access control adapted to the characteristics of the storage medium to be accessed can be performed.
  • 10 is a flowchart illustrating an example of processing on the storage controller side in heterogeneous path switching control processing. It is a schematic diagram for demonstrating the page of a hierarchical capacity virtualization volume. It is a flowchart which shows an example of the heterogeneous path switching control process in a server computer when attention information is transmitted from a storage controller. It is a flowchart which shows an example of an access process.
  • the information of the present embodiment will be described using the expression “table”.
  • the information may not necessarily be expressed by a data structure using a table.
  • it may be expressed by a data structure such as “list”, “DB (database)”, “queue”, or the like. Therefore, “table”, “list”, “DB”, “queue”, and the like can be simply referred to as “information” in order to show that they do not depend on the data structure.
  • the expressions “identification information”, “identifier”, “name”, “name”, “ID” can be used, and these can be replaced with each other. It is.
  • program will be the subject of the description, but since the program uses a memory and a communication port (communication control device) to perform processing determined by being executed by a CPU (Central Processing Unit), The description may be based on the CPU.
  • the processing disclosed with the program as the subject may be processing performed by a computer such as a server computer, a storage controller, a management computer, or an information processing apparatus. Part or all of the program may be realized by dedicated hardware, or may be modularized.
  • Various programs may be installed in each computer by a program distribution server or a storage medium.
  • PCIe Peripheral Component Interconnect Express
  • PCIe Peripheral Component Interconnect Express
  • FeRAM Feroelectric Random Access Memory
  • MRAM Magnetic Resistant Random Access Memory
  • Phase Change Memory Ovonic Uv
  • RRAM registered trademark, Resistance RAM
  • NVMe is an I / F (Interface) standard for realizing high-speed access to the flash memory SSD.
  • IOPS Input / Output per Second
  • low latency access that is, high-speed access
  • flash memory devices may support the NVMe standard. Therefore, there is a possibility that a vendor other than a device driver (typically an application program) provides a program that directly issues an NVMe command for the purpose of high-speed access to the flash memory device.
  • the flash memory SSD is a kind of flash memory device, and the flash memory device has the following features, for example.
  • the flash memory device has a flash memory chip and a flash memory controller.
  • the flash memory controller receives a read request from an external device, the flash memory controller reads data corresponding to the read request from the flash memory chip and transmits the data to the external device.
  • the flash memory controller receives a write request from an external device, the flash memory controller writes the data received together with the write request to the flash memory chip.
  • the flash memory controller performs an erase process for the flash memory chip.
  • FIG. 1 is a block diagram showing a configuration example of a computer system.
  • the computer system 1 includes one or more server computers, one or more storage controllers, a flash memory device (hereinafter referred to as “flash memory”), and a communication mechanism (communication mechanism). Each of these components of the computer system may be referred to as a computer system component.
  • the computer system 1 may be “Converged Platform”. “Converged Platform” may be referred to as “Converged Infrastructure” or “Converged System”. “Converged” may be expressed as “vertical integration”. “Converged Platform” may be abbreviated as “CPF”.
  • the CPF may have the following characteristics.
  • the CPF may include a server computer, a storage system (including a storage controller and a storage device), and a communication mechanism that connects them.
  • a server computer and a storage system are individually introduced into a company, the company often performs operation verification and connection verification of the server computer and storage system.
  • the vendor side performs operation verification and connection verification of the CPF in advance. Therefore, a company can reduce the troubles such as operation verification and connection verification by introducing CPF.
  • the CPF may include a management subsystem that can collectively set the server computer, the storage system, and the communication mechanism.
  • a management program for performing this batch setting may be executed.
  • the management subsystem can quickly provide an execution environment desired by the administrator (virtual machine, DBMS: Database Management System, Web server, etc.). For example, the management subsystem requests the server computer and the storage system to allocate resources necessary for creating a virtual machine. Then, the management subsystem requests the hypervisor to create a virtual machine using resources allocated from the server computer and the storage system.
  • the administrator virtual machine, DBMS: Database Management System, Web server, etc.
  • the management subsystem requests the server computer and the storage system to allocate resources necessary for creating a virtual machine.
  • the management subsystem requests the hypervisor to create a virtual machine using resources allocated from the server computer and the storage system.
  • the CPF 1 includes a server computer 2, a storage controller 3, a flash memory 5 and an HDD 6 as an example of a storage device, and a management computer 7 as an example of a management subsystem.
  • the server computer 2 executes a program (not shown) that can access the storage controller 3.
  • This program accesses the data area on the flash memory 5 provided by the storage controller 3 by issuing an NVMe command.
  • NVMe NVMe command
  • the element which provides a data area as NS (NameSpace) of NVMe will be described later.
  • the server computer 2 includes a CPU 21, a main memory 22, and a root complex (hereinafter abbreviated as RC) 24.
  • Examples of the server computer 2 include a file server, a blade server system, a PC (Personal Computer) server, and a blade inserted into the blade server system.
  • the server computer 2 may be a virtual machine server or a CPU core unit.
  • the program of the server computer 2 may be a business application program (for example, Web server, DBMS, analysis program, middleware), LPAR (Logical Partitioning), a program capable of creating a virtual machine, an OS, a device driver, and the like.
  • a business application program for example, Web server, DBMS, analysis program, middleware
  • LPAR Logical Partitioning
  • a program capable of creating a virtual machine an OS, a device driver, and the like.
  • the communication mechanism connects the server computer 2 and the storage controller 3 with PCIe.
  • PCIe connection between the server computer 2 and the storage controller 3 is a SAN (Storage Area Network) that uses FC (Fibre Channel) and Ethernet (registered trademark), which are employed in the conventional connection between the server computer and the storage controller. It is not always necessary to go through a network such as Compared with the FC protocol, the PCIe connection can suppress the overhead of the conversion process, so that the I / O process for the data area is accelerated. Also, the PCIe connection can reduce the cost as compared with the devices of Ethernet and SAN (especially switches).
  • NVMe may assume a communication mechanism based on PCIe. Therefore, the element that interprets the NVMe command from the server computer 2 may be an Endpoint (hereinafter referred to as “EP”) 41 in PCIe.
  • EP Endpoint
  • the server-side PCIe I / F device 4 that is an example of a mediating device that mediates the PCIe connection from the server computer 2 and the PCIe connection from the storage controller 3 is an element that interprets the NVMe command.
  • the flash memory 5 or the storage controller 3 may be an element that interprets NVMe.
  • the storage controller 3 uses the storage area of the flash memory 5 to provide high-performance I / O processing.
  • the storage controller 3 may have a function that realizes reliability, redundancy, high functionality, and ease of maintenance and management that an enterprise SAN storage subsystem has.
  • the storage controller 3 may have the following functions.
  • the storage controller 3 may have a function of making the flash memory 5 redundant and providing a shared data area using the redundant storage area.
  • the storage controller 3 may be able to attach / detach the flash memory 5 (that is, device maintenance) while allowing access to the data stored in the shared data area (so-called non-stop).
  • the flash memory 5 has a characteristic that the lifetime of the device is shortened as the number of writing increases. Therefore, the storage controller 3 provides such redundancy and non-stop maintenance, so that the reliability of the computer system 1 can be improved.
  • a PCIe flash memory is directly mounted on the server computer 2, it is necessary to maintain the flash memory for each server computer 2.
  • maintenance of the flash memory 5 can be consolidated on the storage system side. As a result, the maintenance worker can perform maintenance work on the flash memory 5 collectively, which facilitates maintenance.
  • the storage controller 3 may have a copy function such as remote copy or snapshot for the data stored by NVMe.
  • the storage controller 3 may have a function of performing tearing using these storage devices when a storage device other than the flash memory 5 such as the HDD 6 is connected. Note that the storage controller 3 may associate the storage area provided by the HDD 6 with the NVMe NS.
  • the storage controller 3 does not go through the server computer 2 but through a network from a computer system (including server computer and storage system) outside the computer system 1 or a network device (including SAN switch and Ethernet switch). It may have a function of performing access. As a result, for example, the above-described remote copy can be performed, and storage consolidation including a computer system or network device outside the computer system 1 can be provided.
  • PCIe has a shorter communicable distance than FC and Ethernet. This is because PCIe has been used as an internal communication path for server computers and storage systems. Therefore, in the PCIe connection, the server computer 2 and the storage controller 3 are often arranged at physically close positions. However, the following configuration is also possible.
  • a blade server system chassis and a storage controller 3 chassis installed in one rack may be sold as a CPF.
  • Management subsystem can execute the following processing.
  • the management subsystem receives a request from the administrator or the integrated management subsystem, and makes settings corresponding to the request for the computer system component.
  • the management subsystem displays the information acquired from the computer system component to the administrator or sends it to the integrated management subsystem.
  • Examples of information to be acquired include performance information, failure information, setting information, configuration information, and the like.
  • the configuration information may include items unique to the computer system and items that can be changed unless the component is attached or detached.
  • the setting information may be an item that can be changed by setting in the configuration information. These types of information may be collectively referred to as component information.
  • the information displayed to the administrator or transmitted to another computer may be the acquired component information as it is, or may be displayed or transmitted after being converted and processed according to a predetermined rule.
  • the management subsystem may set the computer system component automatically or autonomously based on the component information.
  • the management subsystem may be a set of related functions and computers.
  • the management subsystem may be one or more computers that are separate from the computer system components.
  • a computer such as a computer dedicated to a server computer, a computer dedicated to a storage controller, or a computer dedicated to display processing is included in the management subsystem. May be present.
  • the management subsystem may be part of the computer system component.
  • a BMC Baseboard Management Controller
  • an agent program may be the management subsystem.
  • the integrated management subsystem is a subsystem that integrates and manages management target devices such as servers, storage systems, network devices (including SAN switches and Ethernet switches), and computer systems.
  • the integrated management subsystem is connected to the management subsystem and other managed devices via a network.
  • the integrated management subsystem may include one or a plurality of computers connected to the computer system via a network.
  • PCIe connection between the server computer 2 and the storage controller 3 may be performed via a switch (not shown).
  • the server computer 2 includes a CPU 21, a main memory 22, an RC 24, and a server-side PCIe I / F device 4.
  • the RC 24 and the server side PCIe I / F device 4 are connected by PCIe.
  • the RC 24 and the CPU 21 may be connected via a network that is faster than PCIe.
  • the main memory 22 may be connected to the CPU 21 and the RC 24 via a memory controller (not shown) via a network that is faster than PCIe.
  • Each program executed by the server computer 2 may be loaded into the main memory 22 and executed by the CPU 21.
  • the CPU 21 may be a CPU core.
  • the RC 24, the CPU 21, and the memory controller may be combined into one LSI package.
  • the server-side PCIe I / F device 4 is an example of the aforementioned mediation device.
  • the server-side PCIe I / F device 4 may be arranged outside the server computer 2.
  • the server-side PCIe I / F device 4 may have the following features.
  • the server-side PCIe I / F device 4 interprets an NVMe command issued from a program executed by the CPU 21.
  • the server-side PCIe I / F device 4 provides EP 41 to the RC 24.
  • the server-side PCIe I / F device 4 provides another EP 42 to the RC 33 included in the storage controller 3.
  • the server computer 2 may include a local flash memory 23.
  • the local flash memory 23 may be connected to the RC 24 by PCIe.
  • a plurality of elements may be included in the server computer 2.
  • the local flash memory 23 and the server-side PCIe I / F device 4 are described so as to communicate with each other via the RC 24.
  • the local flash memory 23 and the server-side PCIe I / F device 4 may communicate with each other without using the RC 24. It does not have to be.
  • the storage controller 3 has one or more (two in FIG. 1) control units 36.
  • Each control unit 36 includes a CPU 31, a main memory 32, an RC 33, and a flash I / F 372.
  • the RC 33, the server-side PCIe I / F device 4, and the flash I / F 372 may be connected by PCIe.
  • the RC 33 and the CPU 31 may be connected via a network that is faster than PCIe.
  • the main memory 32 may be connected to the CPU 31 and the RC 33 via a memory controller (not shown) via a network faster than PCIe.
  • Each program executed by the storage controller 3 such as the storage program 320 is loaded into the main memory 32 and executed by the CPU 31.
  • the CPU 31 may be a CPU core.
  • the RC 33, the CPU 31, and the memory controller may be combined into one LSI package.
  • Each control unit 36 may include a disk I / F 34 for connection to the HDD 6.
  • the flash I / F 372 and the disk I / F 34 have the same interface type, these two I / Fs may be shared.
  • Examples of the disk I / F 34 include SAS, SATA, FC, and Ethernet, but other communication mechanisms may be used.
  • the flash I / F 372 (or the disk I / F 34) and the server-side PCIe I / F device 4 are described to communicate via the RC 33, but can communicate without going through the RC 33. Or it may not be able to communicate. This also applies to the flash I / F 372 and the disk I / F 34.
  • the computer system 1 may have a plurality of server computers 2.
  • FIG. 2 is a block diagram illustrating an example of the software configuration of the CPF.
  • the server computer 2 includes a server-side PCIe I / F device 262 that is an I / F device for connecting to the storage controller 3 and a PCIe, and a management I / F 272 that is an I / F device for connecting to the management computer 7. Including.
  • the server computer 2 includes an application program (hereinafter referred to as “application”) 228, an OS 227, a heterogeneous path switching control 223, an NVMe control program 222, a SCSI control program 224, and a server management I / F control.
  • the server computer 2 and the storage controller 3 and the management computer 7 may be connected by Ethernet, or may be connected by other physical or virtual means.
  • the server management I / F control program 229 communicates with the management computer 7 by controlling the management I / F 272.
  • the NVMe control program 222 is a program that issues an NVMe command to the PCIe I / F 262.
  • the NVMe control program 222 may be a part of another program stored in the server computer 2, or may be a program different from other programs stored in the server computer 2.
  • the application program 228 may issue an NVMe command, or a device driver in the OS 227 may issue an NVMe command.
  • the PCIe I / F 262 When the PCIe I / F 262 accepts both the NVMe command and the SCSI command, the PCIe I / F 262 needs to have two of the NVMe function and the SCSI function.
  • the PCIe I / F 262 transmits an NVMe command to the PCIe I / F 362 in accordance with the operation of the NVMe control program 222, receives a response to the NVMe command from the PCIe I / F 362, and returns the response to the NVMe control program 222.
  • the SCSI function transmits a SCSI command to the PCIe I / F 362 in accordance with the operation of the SCSI control program 224, receives a response to the SCSI command from the PCIe I / F 362, and returns the response to the SCSI control program 224.
  • Whether or not the PCIe I / F 362 is to be multifunctional may be determined depending on whether or not the NVMe command is interpreted by the mediation device.
  • the heterogeneous path switching control 223 changes the path used when accessing the LU (Logical Unit) recognized by the server computer 2 itself.
  • the LU may have a multipath configuration and define a plurality of paths. Normally, selecting a path to be actually used from the plurality of paths is performed when an access request is made. In this embodiment, a path to be used is determined and registered. The path to be used is changed from the defined paths when there is a predetermined action. Details will be described later.
  • the storage controller 3 includes a management I / F 382 for connecting to the management computer 7 and a flash I / F 372 for connecting to the flash memory 5.
  • the connection between the flash I / F 372 and the flash memory 5 is preferably a PCIe connection when the flash memory 5 interprets the NVMe command, but otherwise, SAS, SATA (Serial Advanced Technology), FC or Ethernet. Etc.
  • the storage controller 3 executes the storage program 320.
  • the storage program 320 may include a PCIe I / F control program 322, a hierarchy management control program 324, a flash I / F control program 323, and a management I / F control program 324. These programs control communication with each interface.
  • the PCIe I / F control program 322 communicates with the server computer 2 by controlling the PCIe I / F 362.
  • the flash I / F control program 323 communicates with the flash memory device 5 by controlling the flash I / F 372.
  • the management I / F control program 324 communicates with the management computer 7 by controlling the management I / F 382.
  • the entity of the PCIe I / F 262 and the PCIe I / F 362 may be the server-side PCIe I / F device 4 or the storage-side PCIe I / F device.
  • the SCSI control program 224 issues a SCSI request for a LUN (Logical Unit Number) provided from the storage controller 3 to the SCSI function of the PCIe I / F 262 in response to a request from another program.
  • the SCSI control program 224 is, for example, a SCSI device driver.
  • the SCSI control program 224 may be a part of another program stored in the server computer 2 or may be a program different from the other program.
  • a device driver in the OS 227 may issue a SCSI request.
  • the server computer 2 and the storage controller 3 are connected through one PCIe I / F 262, but redundancy may be ensured by another method.
  • Storage program 320 may receive, interpret, and process SCSI requests. For example, when the SCSI request is a read request, the storage program 320 reads data from a storage device such as the flash memory 5 or the HDD 6 and transfers it to the server computer 2. At that time, the storage program 320 may use the main memory 32 of the storage controller 3 as a cache memory. For example, when the SCSI request is a write request, the storage program 320 stores the write data in the cache memory 5 and then writes the write data to the storage device.
  • the storage program 320 may perform RAID processing on the parity group.
  • the storage program 320 may define a storage area provided by the storage controller 3.
  • the storage program 320 may store the defined result as the storage area definition information in the main memory 32 of the storage controller 3 and refer to the request processing described above.
  • the storage program 320 may perform processing for realizing enterprise functions such as thin provisioning.
  • a program that does not support NVMe in the server computer 2 can access a storage area corresponding to the NS of the NVMe. Furthermore, a program that does not support NVMe can access a storage area other than the storage area corresponding to NS of NVMe.
  • the server computer 2 can access the storage area of the HDD by SCSI and can access the storage area of the flash memory by NVMe. Thereby, the access time to the storage system of the server computer 2 can be shortened.
  • the HDD may be used for storing data or as a boot device for the server computer 2.
  • the server computer 2 may switch the access method based on the type of storage medium in which data is stored. For example, the server computer 2 may switch between SCSI and NVMe on a volume basis.
  • the path may be assigned to each LU managed by the server computer 2. Since the storage controller 3 needs to receive a request in the SCSI format, a process for converting the NVMe command into a SCSI request is required at any point of time. Therefore, the server-side PCIe I / F device 4 generates a SCSI command based on the NVMe command received from the server computer 2 and transmits it to the storage controller 3. Details of the conversion process will be described later.
  • conventional SCSI control executes I / O scheduling processing and the like.
  • one of the SCSI commands waiting for processing is selected and a path is allocated between the server computer and the storage controller.
  • multipath control is often the default function.
  • multipath control a plurality of paths are provided between the LU recognized on the server side and the storage controller 3 side. This increases availability.
  • multipath control since multipath control is not included in the default function, it is necessary to construct multipath control as a vendor-specific function (see 210 in FIG. 5).
  • FIG. 4 is a conceptual diagram when I / O processing is performed on a hierarchical capacity virtualized volume.
  • the server computer 2 manages the access to the flash memory 5 and generates an NVMe command for the NS space.
  • the server computer 2 When the logical volume is composed of the HDD 6, the server computer 2 generates a SCSI command.
  • the tiered capacity virtualization volume may be managed and controlled by the tier management control program 324.
  • the storage area composed of the flash memory 5 of the hierarchical capacity virtualization volume is accessed via the path on the SCSI driver side.
  • I / O scheduler control included in the control on the SCSI driver side is executed.
  • the flash memory 5 does not have a seek process like the HDD, this I / O scheduler control becomes useless overhead.
  • the storage area constituted by the HDD 6 is accessed via the path on the NVMe driver side, the I / O scheduler control is bypassed, and the access efficiency is lowered.
  • the server computer 2 is provided with the heterogeneous path switching control 223.
  • the heterogeneous path switching control 223 may have the following functions.
  • the first path on the NVMe driver side (sometimes referred to as “NVMe path”) and the second path on the SCSI driver side (referred to as “SCSI path”).
  • the OS 227 can access the same logical volume through paths from different drivers.
  • the heterogeneous path switching control 223 defines multipaths from different drivers. That is, the heterogeneous path switching control 223 constructs a multipath by a path on the SCSI control side and a path on the NVMe control side. Since this is constructed in a lower layer than the OS 227, the OS 227 recognizes that the same volume is accessed even if accessed from a different driver through a path. That is, the OS 227 is not affected.
  • the server computer 2 obtains information from the storage controller 3 that “the volume is a mixture of different storage media”, that is, information that the volume is a hierarchical capacity virtualization volume, and Multiple paths from different types of drivers.
  • Each defined multipath is assigned to each page. For each page requested to be accessed, it is determined which path of a different driver is used, and the determined path is assigned. The path assigned for the first access may be set by default.
  • the heterogeneous path switching control 223 acquires from the storage controller 3 information that the logical volume is a hierarchical capacity virtualization volume and storage medium information for each storage area. Then, the heterogeneous path switching control 223 determines whether a path corresponding to the storage medium information is selected for each storage area, and switches to the corresponding path if not selected.
  • the heterogeneous path switching control 223 selects the NVMe driver side path for the page assigned to Tier 1 (flash memory 5), and selects the SCSI driver side path for the page assigned to Tier 2 (HDD 6). .
  • the heterogeneous path switching control 223 may determine and access any Tier at the first time, receive storage medium information from the storage controller 3, and select an appropriate path from the next time based on the storage medium information.
  • the heterogeneous path switching control 223 may acquire information on the rearrangement destination from the storage controller 3 and change the path assignment.
  • the heterogeneous path switching control 223 may notify the server computer 2 of the change after the execution of the access process, or may be notified when the change is made. For example, when the data stored in the storage area of Tier 1 is moved to the storage area of Tier 2, or when the data stored in the storage area of Tier 2 is moved to the storage area of Tier 1, the storage controller 3 The server computer 2 may be notified of information relating to the movement.
  • Access control can be reduced by performing access control suitable for the characteristics of the storage medium. Thereby, the CPU load can be reduced.
  • NVMe devices such as SSD and flash memory do not generate delays such as HDD seek time.
  • storage areas are allocated from both the NVMe device and the FC device.
  • I / O schedule control is not executed.
  • the I / O schedule control is a control that reduces the seek time and shortens the access time by continuously accessing data having a close physical position on the disk. By shortening the access time, the occupation time of the processor can be reduced. As a result, processor resources can be allocated to other processes such as application processes. That is, I / O schedule control is an important process for improving system performance.
  • FIG. 5 is a schematic diagram showing an outline of processing according to the present embodiment.
  • the server computer 2 has a group of elements 210 that can be independently developed by the vendor according to the NVMe specifications.
  • the element group 210 may include NVM control 222, heterogeneous path switching control 223, and multipath control 225.
  • the NVMe control 222 and the heterogeneous path switching control 223 are as described above.
  • the multipath control 225 may have the same function as the multipath control in the conventional SCSI control.
  • a multi-path may be defined for one logical volume, and the logical volume may be accessible from any path.
  • the tiered capacity virtualization volume 384 for the tiered capacity virtualization volume 384, a path passing through the NVMe control side and a path passing through the SCSI control side are defined. The same hierarchical capacity virtualization volume 384 is accessed through any path.
  • the path passing through the NVMe control side and the path passing through the SCSI control side may each be composed of a plurality of paths for redundancy. Further, when accessing the hierarchical capacity virtualization volume 384, it may be set to use a predetermined path.
  • a path passing through the NVMe control side is used. Which of a plurality of paths for redundancy is used may be appropriately determined at the time of use. At this time, the path passing through the SCSI control side is not used in principle. However, it may be used when specified by the user, the management computer 7 or the system.
  • the server computer 2 may determine that all accesses are performed using the NVMe command.
  • FIG. 6 is a schematic diagram showing the relationship between the NVMe NS and the storage area.
  • the storage controller 3 manages a parity group, a logical volume, a pool, a virtual volume, and a logical unit regarding the storage area.
  • the parity group may be defined using a plurality of storage devices (flash memory 5 and HDD 6). Thereby, high reliability, high speed, large capacity, etc. can be achieved by RAID (Redundant Arrays of Inexpensive Disks).
  • the logical volume is an area obtained by dividing the storage area of the parity group.
  • the capacity of the parity group storage area may be too large to be provided to the server computer as it is.
  • the parity group storage area is divided into a plurality of logical volumes.
  • the pool is composed of a plurality of logical volumes and is used for thin provisioning or tearing.
  • the pool may be configured by parity groups or storage devices.
  • the virtual volume is a virtual storage area to which thin provisioning or tiering is applied, and the virtual storage area is configured by a pool.
  • the logical volume and the virtual volume may be referred to as “volume” without being distinguished.
  • a volume to which tearing is applied may be referred to as a hierarchical capacity virtualized volume.
  • LU is a storage area that is allowed to be accessed from the server computer 2 among virtual volumes or logical volumes.
  • a SCSI LUN is assigned to the LU.
  • the storage controller 3 does not necessarily provide all these types of storage areas.
  • NS may be associated with any of these types of storage areas.
  • NS is preferably associated with each logical unit. This is because the storage program 320 is easily compatible with the storage program 320 of the SAN storage system, and the definition of the storage area is also highly compatible with the SAN storage system.
  • the storage controller 3 may dynamically allocate storage areas to the volumes.
  • the storage controller 3 may configure a plurality of physical storage devices (flash memory 5 and HDD 6) at the same RAID level to form a RAID group.
  • a virtual storage device may be configured from the RAID group, and the virtual storage device may be divided into a plurality of logical storage devices.
  • a page is a unit of a storage area composed of a predetermined capacity and a unit of read / write access from the host.
  • Write data may be stored in one or more pages.
  • one page may be allocated at the time of write access, a plurality of write data may be stored in the one page, and a new page may be allocated when it cannot be stored in the one page.
  • a virtual page of a virtual volume is a unit of virtual storage capacity without an actual storage area, unlike a page such as a pool.
  • Read / write access from the host is processed in units of virtual pages of the virtual volume.
  • the real page of the pool volume is allocated to the virtual page of the virtual volume.
  • the storage controller 3 manages the correspondence between pools and virtual volumes. A page is allocated to a virtual volume from a pool corresponding to the virtual volume. When the server computer 2 recognizes that the storage medium is only the flash memory, the server computer 2 may access all with the NVMe command.
  • FIG. 7 shows a configuration example of the path management table.
  • the bus management table 400 may be stored in the main memory 22 of the server computer 2.
  • the path management table 400 manages the volume type 420 and the SCSI path numbers 430 and 431 for each logical volume number 410.
  • the volume type 420 is information indicating which type of storage medium the physical storage area constituting the logical volume in the storage controller 3 is configured from.
  • the logical volume number in the storage controller 3 may have a correspondence relationship with the logical volume number managed by the server computer 2.
  • the type 420 of the volume configured by the HDD (FC disk) managed by the storage controller 3 may be “FC”.
  • the volume type 420 may be determined by a storage medium that constitutes a pool in which a storage area is allocated to a logical volume. According to the path management table 400 of FIG. 7, it can be seen that the storage medium with the logical volume number “1” is “HDD (FC disk)” and the storage medium with the logical volume number “2” is flash memory.
  • the storage medium allocated for each page of the logical volume may be different.
  • the volume type 420 may be “mixed”.
  • SCSI path numbers 430 and 431 indicate the path number used by the logical volume and the control type. According to the path management table 400 of FIG. 7, since the logical volume number “1” is an FC disk, the server computer 2 executes SCSI control through the SCSI side driver, and then performs SCSI path number “8” or “10”. The storage controller 3 may be accessed through this path.
  • a plurality of SCSI path numbers 430 and 431 may be set as shown in the path management table 400 of FIG. This is for managing redundant paths. Path redundancy may be automated with multipath control.
  • the volume type 420 is “mixed”, that is, a hierarchical capacity virtualization volume
  • the server computer 2 allocates a SCSI path and an NVMe path.
  • the path management table 400 includes a sub-table 440 and may manage which storage medium storage area is allocated for each page. As a result, a path suitable for the storage medium can be selected.
  • FIG. 8 shows information managed by the storage controller 3.
  • the LU management table 500 manages logical volume information provided to the server computer 2.
  • the LU management table 500 has an entry for each LUN in the storage controller 3.
  • the LU management table 500 includes, as items, a path number 510, a port number 520, a LUN 530, an NVMe / SCSI 540, and a logical volume number 550.
  • the port number 520 is an identification number of the port on the server computer 2 side associated with the LUN 530 on the storage controller 3 side.
  • the LUN 530 is an LU identification number on the storage controller 3 side, and is a LUN on the storage controller 3 side associated with the port number 520.
  • the logical volume number 550 is the number of the logical volume on the storage controller 3 side that constitutes the LU of the LUN 530.
  • the logical volume with the logical volume number 550 may be a normal volume, a capacity virtualization volume, a hierarchical capacity virtualization volume, or the like.
  • the logical volume management table 600 is a table for managing which storage medium is allocated to each logical volume.
  • a storage area in the HDD is allocated to the logical volume.
  • one logical volume is assigned a storage area of one type of storage medium. Storage areas are allocated to normal volumes in units of volumes, and capacity virtualized volumes in units of pages.
  • a free storage area is selected from the pool associated with the capacity virtualized volume and assigned to the capacity virtualized volume page.
  • This pool may be composed of one type of storage medium.
  • a free storage area is selected and assigned to the hierarchical capacity virtualized volume from the pool associated with the hierarchical capacity virtualized volume.
  • This pool may be composed of a plurality of types of storage media. Therefore, in the hierarchical capacity virtualization volume, the storage medium allocated for each page can be different.
  • the hierarchical capacity virtualization volume management entry 640 manages these pieces of information.
  • the volume type 620 of the logical volume management table 600 may be linked from “mixed (hierarchical capacity virtualized volume)”.
  • the hierarchical capacity virtualization volume management entry 640 manages addresses to storage areas allocated to all pages of the link source logical volume. Typically, for each page, a storage device and an address corresponding to a storage area allocated to the page are managed. In FIG. 8, the logical volume management table 600 indicates that the logical volume with the logical volume number “3” is “mixed (hierarchical capacity virtualization volume)”. Then, the hierarchical capacity virtualization volume management entry 640 linked from there indicates that the address “100” of the flash memory is assigned to the page of the page number “0” of the logical volume. The hierarchical capacity virtualization volume management entry 640 may be further hierarchized.
  • the logical volume number 550 may be the same on the server computer 2 side and the storage controller 3 side.
  • the server computer 2 and the storage controller 3 may be an integrated device or may be separate devices.
  • the storage controller 3 side identifies the LUN based on the information on the server computer 2 side included in the read command transmitted from the server computer 2 side, and identifies the logical volume number on the storage controller 3 side. May be.
  • the same logical volume number on the server computer 2 side and the storage controller 3 side will be described as having a correspondence relationship.
  • the server computer 2 receives the attention information from the storage controller 3 and recognizes the information on the storage medium of the page. Details of the attention information will be described later.
  • FIG. 9 is a flowchart showing an example of the first path allocation process.
  • the storage area is defined in order from the physical hierarchy such as the flash memory 5 or the HDD 6 toward the logical hierarchy.
  • a pool used for the hierarchical capacity virtualization volume may also be defined.
  • a path to be used when accessing the logical volume is determined and registered in the logical volume management table 600.
  • the storage controller 3 defines a storage area from the flash memory 5 and the HDD 6 which are examples of physical devices.
  • the storage controller 3 divides this storage area, associates one of the storage areas with the logical volume, and registers the correspondence relationship in the logical volume management table 600. Details of the definition of the storage area will be described later.
  • the storage controller 3 defines the pool after defining the logical volume. For example, the administrator specifies a pool ID, the number and number of pool volumes that are logical volumes to be registered in the pool, and creates a pool through the management computer 7.
  • the server computer 2 transmits a discovery command to the storage controller 3.
  • the discovery command is a command for obtaining a list of LUs that can be provided by the storage controller 3. For example, when the administrator instructs volume allocation through the management computer 7, the server computer 2 transmits a discovery command to the storage controller 3.
  • the storage controller 3 When the storage controller 3 receives the discovery command, it sends a list of LUs to the server computer 2. The server computer 2 allocates a volume using the received LU list.
  • the server computer 2 can set a LUN path after assigning a volume.
  • the volume for which the LUN path is set can be accessed from the server computer 2. That is, the server computer 2 can transmit an access request for the volume.
  • the server computer 2 can recognize one LU from a plurality of paths by setting a multipath.
  • the storage controller 3 associates the port number 520 and the LUN 530 with the path number 500 and registers them in the LU management table (S4010).
  • the storage controller 3 transmits path management information and volume attribute information to the server computer 2 (S4012).
  • the volume attribute information includes, for example, information indicating whether it is HDD, flash memory, or a mixture of FC (HDD) and flash memory by hierarchical capacity virtualization.
  • the server computer 2 receives the path information and the volume attribute information from the storage controller 3, and registers the path information and the volume attribute information in the path management table 400 (S4014).
  • the server computer 2 assigns multipaths to the volumes by the multipath controls 230 and 231. For example, the server computer 2 refers to the attribute of the volume and assigns an NVMe path having the attribute of the flash memory to the volume configured with the flash memory 5.
  • the server computer 2 allocates a SCSI path having the SCSI attribute to the volume configured by FC or the tiered capacity virtualization volume (S4016).
  • the attribute information of the recording medium on the storage side is acquired, the type of protocol used for access is determined according to the type of the medium, and the path corresponding to the protocol is set. . This enables access using a protocol according to the access characteristics of the medium.
  • the server computer 2 registers the assigned path information in the path management table 400 (S4018). Then, the server computer 2 transmits the assigned path information to the storage controller 3 (S4020).
  • the storage controller 3 registers the received path information in the logical volume management table 600 (S4022).
  • the storage controller 3 may assign a multipath to the volume in the same manner as described above.
  • the storage controller 3 transmits the assigned path information to the server computer 2, and the server computer 2 registers the received path information in the path management table 400.
  • the management computer 7 may manage the path information, and the management computer 7 may transmit the path information to the storage controller 3 and the server computer 2.
  • the storage controller 3 acquires information of at least a first storage medium (for example, flash memory) or a second storage medium (for example, HDD) assigned to the logical volume, and stores the information in the logical volume based on the acquired information.
  • At least one of the NVMe path and the SCSI path is set.
  • FIG. 10 is a flowchart showing an example of processing on the server computer 2 side in the heterogeneous path switching control processing.
  • the server computer 2 generates an access request (S5010).
  • the server computer 2 refers to the path management table 400 and determines whether or not the LU having data corresponding to the access request is a hierarchical capacity virtualization volume (S5012).
  • the server computer 2 performs the following processing. That is, the server computer 2 refers to the sub-table 440, calculates a page number based on the access target address, and determines whether a path is assigned to the calculated page number (S5014).
  • the server computer 2 selects one of the paths assigned to the logical volume and assigns the selected path to the page. For example, in the path management table 400 shown in FIG. 7, one of the SCSI path numbers “9” and “1” corresponding to the logical volume number “3” is selected (S5016). The selection method will be described later.
  • the server computer 2 registers the selected pass number in the sub-table 440 (S5018).
  • the server computer 2 determines whether or not the selected path is a path via NVMe control (S5020).
  • the server computer 2 includes an attention request in the access request transmitted to the storage controller 3 (S5022).
  • This attention request may include information (for example, a predetermined bit value) indicating that the path selection for this page is made in the initial unconfirmed state. Details of the attention request will be described later.
  • the server computer 2 executes an NVMe control process (S5024). Details of the NVMe control process will be described later (see S8220). Then, the server computer 2 converts the access request into a SCSI command format that can be recognized by the storage controller 3, and transmits it to the storage controller 3 (S5050).
  • the server computer 2 includes the same attention request as described above in the access request (S5022). Then, the server computer 2 executes the SCSI control process (S5026). Details of the SCSI control processing will be described later (see S8320). Then, the server computer 2 converts the access request into a SCSI command format that can be recognized by the storage controller 3, and transmits it to the storage controller 3 (S5050).
  • the server computer 2 refers to the sub-table 440, and determines whether or not the previous path for the page number was a path through NVMe control (S5030).
  • the server computer 2 executes the NVMe control process (S5024). Then, the server computer 2 converts the access request into a SCSI format that can be recognized by the storage controller 3, and transmits it to the storage controller 3 (S5050).
  • the server computer 2 executes SCSI control processing (S5026). Then, the server computer 2 converts the access request into a SCSI format that can be recognized by the storage controller 3, and transmits it to the storage controller 3 (S5050).
  • Attention information is for transmitting the type of storage medium to the server computer 2.
  • the server computer 2 may select the path without considering the type of storage medium in the storage area allocated to the page.
  • the server computer 2 can recognize the type of the storage medium by receiving the attention information from the storage controller 3. Since the type of the storage medium only needs to be recognized once, the server computer 2 sends an attention request to the storage controller 3 at the timing when the path first associated with the page is selected, and the storage controller 3 sends attention information. That's fine. If the storage medium in the storage area allocated to the page is changed, the storage controller 3 may spontaneously transmit the changed attention information to the server computer 2.
  • the attention information may include address information and storage medium information of a storage area managed on the storage controller 3 side including the page requested to be accessed.
  • the storage medium information may be Tier information.
  • the path first associated with the page may be selected by any of the following methods.
  • a path that passes through the NVMe side control is selected.
  • a path is selected at random.
  • a path is selected based on the monitored information.
  • FIG. 11 is a flowchart showing an example of processing on the storage controller 3 side in the heterogeneous path switching control processing.
  • the storage controller 3 receives an access request from the server computer 2 (S5110).
  • the storage controller 3 determines whether or not an attention request is included in the received access request (S5120).
  • the storage controller 3 executes an access process (read / write process, etc.) corresponding to the access request (S5130).
  • the storage controller 3 transmits read data (in the case of read processing) and an execution result to the server computer 2 after completion of the access processing (S5140).
  • S5120 If an attention request is included in S5120 (S5120: YES), the storage controller 3 executes an access process corresponding to the access request in the same manner as in S5130 above (S5150). Then, after the completion of the access processing, the storage controller 3 transmits read data (in the case of read processing), an execution report, and attention information relating to the page to be accessed to the server computer 2 (S5160).
  • the storage medium for storing data may be changed on the storage controller 3 side.
  • the storage controller 3 may migrate the data stored in the flash memory 5 to the HDD 6 (or vice versa) based on the result of monitoring the access frequency or the like or based on an instruction from the user.
  • the storage controller 3 executes such a migration between Tiers (S5180)
  • the storage controller 3 transmits attention information having the changed contents to the server computer 2 (S5200).
  • the storage controller 3 When the storage controller 3 detects that a request has been received from a different path such as an NVMe path at the timing of the access request, the storage controller 3 executes and completes the access process.
  • the server computer 2 may be notified of attention information to change the path.
  • the storage controller 3 may perform heterogeneous path switching control. For example, the storage controller 3 accesses the logical volume via the NVMe path when accessing the storage area of the flash memory 5, and accesses the logical volume via the SCSI path when accessing the storage area of the HDD 6. It's okay.
  • FIG. 12 is a schematic diagram for explaining a case where the management unit (for example, page) size of the storage area of the hierarchical capacity virtualization volume is different between the server computer 2 and the storage controller 3.
  • the management unit for example, page
  • the unit size (page size) of the storage area managed by the storage controller 3 is larger than the unit size (page size) of the storage area managed by the server computer 2.
  • the server computer 2 receives the attention information from the storage controller 3, the server computer 2 acquires storage medium information corresponding to one page managed by the storage controller 3, and associates the page with a plurality of pages managed by the server computer 2.
  • pages “1” and “2” managed by the storage controller 3 correspond to pages “1” to “4” and “5” to “8” managed by the server computer 2, respectively.
  • page “1” in the storage controller 3 is configured from the flash memory 5 and page “2” is configured from the HDD 6.
  • the server computer 2 receives the attention information indicating that from the storage controller 3, the storage media of the pages “1” to “4” managed by the server computer 2 are “flash memory”, and the page “5” The information indicating that the storage media “8” to “8” are “HDD” is registered.
  • storage medium information for each page of the hierarchical capacity virtualization volume is not shown in the sub-table 440, but an entry relating to the storage medium information may be provided in the sub-table 440.
  • the path information is managed as 450_1, 450_2,... Therefore, when the server computer 2 receives that the page “1” managed by the storage controller 3 is configured from the flash memory 5, the server computer 2 associates with the NVMe control among a plurality of paths associated with the volume. The obtained MVNe bus may be associated with the page “1”. In this way, the server computer 2 can determine which type of path is associated with each of a plurality of pages with one attention information.
  • FIG. 13 is a flowchart showing a processing example of heterogeneous path switching control in the server computer 2 when attention information is transmitted from the storage controller.
  • the attention notice may be a signal output from the storage controller 3 to the server computer 2.
  • the attention information transmitted together with the attention notification includes information related to the page specified by the access request transmitted from the server computer 2 to the storage controller 3.
  • the information may include storage medium information, page start address, and page number information.
  • the storage medium information is information indicating the type of storage medium constituting the storage area allocated to the page.
  • the page start address is information indicating the start address in the storage controller 3 corresponding to the page number in the server computer.
  • the management unit size of the server computer 2 is smaller than the management unit size in the storage controller 3. Therefore, the type of the storage medium corresponding to one page number of the server computer 2 is the same as the type of storage medium constituting the page in the storage controller 3 corresponding to the page number. Therefore, the server computer 2 can acquire the types of storage media of all pages of the storage controller 3 using the page number managed by the server computer 2.
  • the server computer 2 receives the attention information from the storage controller 3 (S5310). There are, for example, the following two patterns for receiving attention information. In any case, the processing after receiving attention information is the same.
  • the server computer 2 may have a mechanism for detecting attention information.
  • the server computer 2 may periodically check the attention detection queue and start the following processing if the queue is registered there.
  • the server computer 2 When the server computer 2 receives the attention information, the server computer 2 retrieves the logical volume number and page number managed by the server computer 2 from the attention information (S5312). Based on the attention information, the server computer 2 updates the pass number information associated with each retrieved page number.
  • the server computer 2 determines whether the attention information storage medium type is Tier 1 (any storage medium type can be identified) for one page number (S5314).
  • the server computer 2 associates the SCSI path with the page number (S5324). Then, the process proceeds to S5322.
  • the server computer determines whether or not the page number to be updated still remains, returns to S5314 if it remains, and ends the process if it does not remain.
  • FIG. 14 is a flowchart illustrating an example of access processing according to the present embodiment.
  • the NVMe command is issued from a program that issues an I / O command in the server computer 2.
  • the NVMe command may be issued by an application or a driver in the OS. The same applies to the SCSI command.
  • the NVMe command is issued to the logical volume 710 (FIG. 6) configured from the flash memory.
  • a SCSI command is issued to the logical volume 720 configured from the HDD 6.
  • a SCSI command or an NVMe command may be set as a command to be issued first. This setting may be set and changed by the user. In this embodiment, it is assumed that the NVMe command is set for the first command to be issued.
  • FIG. 10 when the server computer 2 sets the path correspondence relationship and becomes ready to accept an access request, for example, when the OS 227 issues an I / O command, the processing of FIG. 10 is executed. Details of the NVMe control process of S5024 in FIG. 10 are shown in S8220 of FIG. 14, and details of the SCSI control process of S5026 in FIG. 10 are shown in S8320 to S8340 of FIG. Details of S5050 in FIG. 10 are shown in S8420 to S8460 in FIG.
  • the NVMe control 222 performs various processes such as a process for interpreting information included in the NVMe command and a necessary file open process, but only a difference from the SCSI control will be described here.
  • the SCSI control 224 performs I / O scheduling processing.
  • the NVMe command processing related to the NVMe command performed between the server computer 2, the server-side PCIe I / F device 4, and the control unit 36 is shown below. The following processing is applied when the NVMe command is read or / and write, but may be applied to other NVMe commands.
  • the detailed processing procedure of S8420 to S8460 may be as described in SS1 to SS7 below.
  • the following processing is performed when the storage controller 3 includes a plurality of control units 36, each control unit 36 includes a plurality of CPUs 31, and the logical unit corresponds to NS.
  • the server computer 2 transmits the NVMe command by the processing of the program described above. It should be noted that the server computer 2 can designate the target NS by including the NS ID in the NVMe command.
  • the NVMe command includes the access range in the NS ID and the memory range of the server computer 2.
  • the server-side PCIe I / F device 4 receives the NVMe command.
  • the server-side PCIe I / F device 4 interprets the received NVMe command and converts the NSID included in the command into a corresponding LUN.
  • the server-side PCIe I / F device 4 generates a SCSI command including the converted LUN.
  • the server-side PCIe I / F device 4 determines the control unit 36 and the CPU 31 that are transmission destinations of the generated SCSI command.
  • the server-side PCIe I / F device 4 transmits the generated SCSI command to the determined transmission destination.
  • the CPU 31 of the destination control unit 36 receives the SCSI command and processes the received SCSI command.
  • the server-side PCIe I / F device 4 transfers the write data stored in the main memory 22 of the server computer 2 to the main memory 32 of the storage controller 3. You can do it.
  • the server-side PCIe I / F device 4 may transfer the read data stored in the main memory 32 of the storage controller 3 to the main memory 22 of the server computer 2.
  • the conversion from NSID to LUN in SS3 may be performed by, for example, either (B1) or (B2) below or a combination thereof.
  • the server-side PCIe I / F device 4 converts NSID to LUN by a predetermined conversion formula (which may include bit operation).
  • the server-side PCIe I / F device 4 can convert from LUN to NSID by an inverse conversion formula that is paired with a predetermined conversion formula.
  • the server-side PCIe I / F device 4 stores a conversion table for converting the NSID into LUN in the memory of the server-side PCIe I / F device 4, and uses the conversion table to convert the NSID to LUN. Convert.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

La présente invention concerne un système informatique comprenant: un ordinateur serveur; un premier dispositif de stockage doté d'un premier support de stockage; un deuxième dispositif de stockage doté d'un deuxième support de stockage qui diffère du premier support de stockage; et un contrôleur de stockage qui est relié à l'ordinateur serveur via PCI-Express et qui est relié aux premier et deuxième dispositifs de stockage. Le contrôleur de stockage met en place un volume logique; obtient des informations du premier ou du deuxième support de stockage attribué au volume logique; et, d'après les informations obtenues, affecte au volume logique soit un premier chemin qui fait l'objet d'un accès par une consigne qui se conforme à une première norme, soit un deuxième chemin qui fait l'objet d'un accès par une consigne qui se conforme à une deuxième norme.
PCT/JP2015/056202 2015-03-03 2015-03-03 Système informatique et procédé de commande de stockage WO2016139749A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2015/056202 WO2016139749A1 (fr) 2015-03-03 2015-03-03 Système informatique et procédé de commande de stockage

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2015/056202 WO2016139749A1 (fr) 2015-03-03 2015-03-03 Système informatique et procédé de commande de stockage

Publications (1)

Publication Number Publication Date
WO2016139749A1 true WO2016139749A1 (fr) 2016-09-09

Family

ID=56848831

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2015/056202 WO2016139749A1 (fr) 2015-03-03 2015-03-03 Système informatique et procédé de commande de stockage

Country Status (1)

Country Link
WO (1) WO2016139749A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018073950A1 (fr) * 2016-10-21 2018-04-26 株式会社日立製作所 Système informatique et procédé de contrôle d'instructions
CN108632338A (zh) * 2017-03-16 2018-10-09 三星电子株式会社 存储系统和用于提供自动存储发现的方法

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003345658A (ja) * 2002-05-29 2003-12-05 Hitachi Ltd ストレージ集中管理方法
JP2006011932A (ja) * 2004-06-28 2006-01-12 Hitachi Ltd ストレージ装置及びストレージ装置の排他制御方法
JP2007226596A (ja) * 2006-02-24 2007-09-06 Hitachi Ltd 記憶制御装置及び記憶制御装置を用いたデータマイグレーション方法
JP2012208916A (ja) * 2011-03-28 2012-10-25 Hitachi Ltd 仮想ボリュームに領域を割り当てるための方法及び装置
JP2014506688A (ja) * 2011-06-07 2014-03-17 株式会社日立製作所 フラッシュメモリを含むストレージシステム、及び記憶制御方法
JP2014514622A (ja) * 2011-09-13 2014-06-19 株式会社日立製作所 フラッシュメモリを含むストレージシステム、及び記憶制御方法
JP2014232543A (ja) * 2012-06-19 2014-12-11 株式会社東芝 制御プログラム、ホスト装置の制御方法、情報処理装置およびホスト装置
JP2015036982A (ja) * 2013-08-16 2015-02-23 エルエスアイ コーポレーション ホストとコントローラとの間でパーティション化された変換レイヤ

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003345658A (ja) * 2002-05-29 2003-12-05 Hitachi Ltd ストレージ集中管理方法
JP2006011932A (ja) * 2004-06-28 2006-01-12 Hitachi Ltd ストレージ装置及びストレージ装置の排他制御方法
JP2007226596A (ja) * 2006-02-24 2007-09-06 Hitachi Ltd 記憶制御装置及び記憶制御装置を用いたデータマイグレーション方法
JP2012208916A (ja) * 2011-03-28 2012-10-25 Hitachi Ltd 仮想ボリュームに領域を割り当てるための方法及び装置
JP2014506688A (ja) * 2011-06-07 2014-03-17 株式会社日立製作所 フラッシュメモリを含むストレージシステム、及び記憶制御方法
JP2014514622A (ja) * 2011-09-13 2014-06-19 株式会社日立製作所 フラッシュメモリを含むストレージシステム、及び記憶制御方法
JP2014232543A (ja) * 2012-06-19 2014-12-11 株式会社東芝 制御プログラム、ホスト装置の制御方法、情報処理装置およびホスト装置
JP2015036982A (ja) * 2013-08-16 2015-02-23 エルエスアイ コーポレーション ホストとコントローラとの間でパーティション化された変換レイヤ

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018073950A1 (fr) * 2016-10-21 2018-04-26 株式会社日立製作所 Système informatique et procédé de contrôle d'instructions
CN108632338A (zh) * 2017-03-16 2018-10-09 三星电子株式会社 存储系统和用于提供自动存储发现的方法
CN108632338B (zh) * 2017-03-16 2022-05-24 三星电子株式会社 存储系统和用于提供自动存储发现的方法

Similar Documents

Publication Publication Date Title
JP6286542B2 (ja) 計算機システム
JP5502232B2 (ja) ストレージシステム、及びその制御方法
JP6273353B2 (ja) 計算機システム
US8639899B2 (en) Storage apparatus and control method for redundant data management within tiers
US9124613B2 (en) Information storage system including a plurality of storage systems that is managed using system and volume identification information and storage system management method for same
EP1837751B1 (fr) Système de stockage, procédé de libération d'étendue de stockage et appareil de stockage
JP5981563B2 (ja) 情報記憶システム及び情報記憶システムの制御方法
US8984221B2 (en) Method for assigning storage area and computer system using the same
JP4814119B2 (ja) 計算機システム、ストレージ管理サーバ、及びデータ移行方法
JP5512833B2 (ja) ストレージの仮想化機能と容量の仮想化機能との両方を有する複数のストレージ装置を含んだストレージシステム
US8527699B2 (en) Method and system for distributed RAID implementation
US8904121B2 (en) Computer system and storage management method
US9003087B2 (en) Compound storage system and storage control method
JP2007148812A (ja) ストレージシステム及びその負荷分散方法
JP5996098B2 (ja) ストレージデバイスの高速アクセス及びデータ保護を実現する計算機、計算機システム、及びｉ／ｏ要求処理方法
US8572347B2 (en) Storage apparatus and method of controlling storage apparatus
US9239681B2 (en) Storage subsystem and method for controlling the storage subsystem
WO2016139749A1 (fr) Système informatique et procédé de commande de stockage
US20210243082A1 (en) Distributed computing system and resource allocation method
JP6019169B2 (ja) 情報記憶システム
JP2018101440A (ja) 計算機システム
JP2012155583A (ja) ディスクアレイ装置及び論理ボリュームアクセス方法
JP6231700B2 (ja) サーバストレージシステムと管理システムを有する計算機システム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15883917

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15883917

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP