Patent Grant
9875060
The field of the invention is data processing, or, more specifically, methods, apparatus, and products for migrating MMIO.
A logical partition may be configured to utilize a physical, rather than virtual, I/O adapter. Such a configuration may rely on direct memory access operations between the I/O adapter. In current server-class systems there are large I/O fabrics incorporating many layers of bridge chips, switches, and I/O devices. The I/O devices themselves may be further virtualized in technologies such as SR-IOV. A memory mapped I/O (MMIO) access request generated by a logical partition must properly traverse this fabric and be delivered to the I/O device.
The routing and handling of the MMIO accesses involve both hardware configuration and software configuration across many layers of the system including the hypervisor and the logical partition. The logical partition is made directly aware of the MMIO address mappings which are associated with specific I/O devices. Any changes to these relationships require the updated configuration mappings to be communicated across all parties. This limits both configuration flexibility and the ability to reconfigure (for example, due to hardware failure or resource movement).
Methods, apparatus and products for migrating memory mapped input/output (‘MMIO’) from a source input/output (‘I/O’) adapter of a source computing system to a destination I/O adapter of a destination computing system are disclosed in this specification. Such migration may include: collecting, by a source hypervisor of the source computing system, MMIO mapping information, where the source hypervisor supports operation of a logical partition executing on the source computing system and the logical partition is configured for MMIO operations with the source I/O adapter through a memory management unit (‘MMU’) of the source computing system utilizing the MMIO mapping information; placing, by a destination hypervisor of the destination computing system, the destination I/O adapter in an error state; migrating the logical partition from the source computing system to the destination computing system; configuring, by the destination hypervisor of the destination computing system, the destination computing system for MMIO with the LPAR utilizing the MMIO mapping information collected by the source hypervisor; and restarting the logical partition on the destination computing system, including recovering, by the logical partition, from the error state.
The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular descriptions of exemplary embodiments of the invention as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts of exemplary embodiments of the invention.
Embodiments of methods, apparatus, and computer program products for migrating MMIO from a source I/O adapter of a source computing system to a destination I/O adapter of a destination computing system are described with reference to the accompanying drawings, beginning with
The example data center (120) of
The computing system (102) includes at least one computer processor (156) or “CPU” as well as random access memory (168) or “RAM,” which is connected through a high speed memory bus (166) and bus adapter (158) to processor (156) and to other components of the computing system (102).
Stored in RAM (168) is a hypervisor (136) and a management console (138). The management console (138) may provide a user interface through which a user may direct the hypervisor (136) on instantiating and maintaining multiple logical partitions (116, 118), where each logical partition may provide virtualization services to one or more clients. The management console (138) may also administer the migration of MMIO from the source I/o adapter of the source computing system to a destination I/O adapter of the destination computing system.
Also stored in RAM (168) are two instances of an operating system (154), one for each logical partition (116, 118). Operating systems useful in computers configured for MMIO migration according to various embodiments include UNIX™, Linux™, Microsoft Windows™, AIX™, IBM's i™ operating system, and others as will occur to those of skill in the art. The operating systems (154), hypervisor (136), and management console (138) are shown in RAM (168), but many components of such software may typically be stored in non-volatile memory such as, for example, on a data storage (170) device or in firmware.
The computing system (102) may also include a storage device adapter (172) coupled through expansion bus (160) and bus adapter (158) to processor (156) and other components of the computing system (102). Storage device adapter (172) connects non-volatile data storage to the computing system (102) in the form of data storage (170). Storage device adapters useful in computers configured for MMIO migration according to various embodiments include Integrated Drive Electronics (“IDE”) adapters, Small Computing system Interface (“SCSI”) adapters, and others as will occur to those of skill in the art. Non-volatile computer memory also may be implemented as an optical disk drive, electrically erasable programmable read-only memory (so-called “EEPROM” or “Flash” memory), RAM drives, and so on, as will occur to those of skill in the art.
The example computing system (102) may also include one or more input/output (“I/O”) adapters (178). I/O adapters implement user-oriented input/output through, for example, software drivers and computer hardware for controlling output to display devices such as computer display screens, as well as user input from user input devices (181) such as keyboards and mice. The example computing system (104) may also include a video adapter (114), which may be an example of an I/O adapter specially designed for graphic output to a display device (180) such as a display screen or computer monitor. Video adapter (114) may be connected to processor (156) through a high speed video bus (164), bus adapter (158), and the front side bus (162), which may also be a high speed bus.
The example computing system (102) of
The network adapters (124, 126, and 128) may further be configured for data communications with hosts (195) over a network (101) reachable through local area networks (LANs), such as LAN (100). The network adapters (124, 126, and 128) may further be configured for data communications with storage area networks (SANs), such as SAN (112), and for data communications with various storage devices, such as storage devices (106) and storage devices (108).
From time to time and for various reasons, a logical partition may be migrated from one computing system, such as computing system (102), to another computing system, such as one of the hosts (195). In prior art embodiments in which the logical partition is coupled to an I/O adapter through a physical communication channel (rather than an entirely virtualized communications channel), the I/O adapter must be deconfigured from the logical partition completely prior to migration. Once migrated to the destination computing system, an I/O adapter would be added to the logical partition as if a new adapter were added to the system. Further, in some embodiments, the logical partition and I/O adapter are configured for MMIO. A logical partition migration that requires the deconfiguring of an I/O device and an addition of a new I/O device would generally require a complete reconfiguration of MMIO.
When configuring a logical partition (116) for MMIO with a source I/O adapter, the hypervisor, upon instantiation of the logical partition may create or modify are hardware page table (142) in a memory management unit (MMU) (140) with MMIO mapping information, among other address mappings. The logical partition is configured with logical addresses for MMIO. Such logical addresses appear to the logical partition as real physical memory addresses, but do not actually reference a physical memory location directly. Instead, the logical addresses must be translated into physical addresses. The hardware page table (142) may include entries that associates a logical address assigned to a logical partition to a physical address. Readers of skill will recognize that many other data structures and layers of abstraction may exist as well. The MMU receives a logical address as part of an MMIO operation initiated by the logical partition on the memory bus (166), translates the logical address to a physical address and generates an I/O transaction on the expansion bus (160) directed to the I/O adapter coupled to the logical partition for MMIO. In migrating the logical partition from one computing device to another, the mappings from logical addresses to physical addresses will generally not be correct. Completely reconfiguring the logical partition after migration with a new I/O adapter, MMIO, and new logical addresses injects downtime into the migration process.
To that end, the computing system (102) of
Examples of such an I/O adapter may be the network adapters (124, 126, and 128) of
The arrangement of servers and other devices making up the exemplary system illustrated in
For further explanation,
In the example of
The example source computing system (202) of
In the example of
By contrast, the system of
Likewise, the destination hypervisor may be instructed by the management console to prepare for the migration by collecting physical real addresses utilized to route MMIO operations to memory space of the destination I/O adapter (224). Such physical addresses may be collected from the hardware page table (220) of the MMU (222) of the destination computing system (214). Such addresses will be later mapped to the LPAR logical address assigned for MMIO.
The management console (226) may also instruct the destination hypervisor (218) to place the destination I/O adapter in an error state. Such an error state may include the ‘EEH’ (enhanced error handling) error state which is an extension to the PCI standard specification and enabled in systems running IBM's Power™ Processors.
The management console may then migrate (228) the logical partition (204) from the source computing system (202) to the destination computing system (214). Such a migration may include quiescing the logical partition and copying the logical partition from the source computing system memory to the destination computing system memory. At this time, the logical partition has not been restarted on the destination computing system.
The destination hypervisor may then configure the destination computing system for MMIO with the logical partition utilizing the MMIO mapping information collected by the source hypervisor. Such configuration may include: inserting in a hardware page table (220) of an MMU of the destination computing system, mappings of the LPAR logical memory addresses collected by the source hypervisor (206) to the physical real addresses utilized to route MMIO operations to memory space of the destination I/O adapter.
Finally, the logical partition may be restarted on the destination computing system. The term ‘restart’ is not equivalent to a reboot of the logical partition. Instead, ‘restart’ means ‘resume.’ That is, when the logical partition is migrated it is first paused on the source computing system, then unpaused or resumed on the destination computing system. As part of the restarting process, the logical partition and destination I/O adapter may recover from the error state. The logical partition maintains the LPAR logical addresses assigned for MMIO with the source I/O adapter (212) after migration to the destination computing system (214). Further, the logical partition (204) may perform MMIO operations by utilizing the same LPAR logical addresses.
As mentioned above, in some embodiments, specifically in embodiments in which the I/O adapter is implemented as a PCIe adapter or SR-IOV adapter, the MMIO mapping information collected by the source hypervisor and utilized to update mapping information on the destination computing system may be implemented as one or more data structures including a hardware page table. For further explanation, therefore,
The example of
In the example of
The MMU may then generate a PCI transaction (318) through the PCI fabric (314), by identifying the PCI address associated with the physical real address through use of the PCI address table (310). The PCI address table (310) includes entries, each of which in turn includes an association of a physical real address to a PCI address. The PCI address is utilized to address the I/O adapter (316) through the PCI bus. To that end, the MMU generates the PCI transaction (318) with the PCI address and the I/O command from the original MMIO operation (306) and places the PCI transaction (318) on the bus.
Readers will understand that the example of
For further explanation,
The method of
The method of
The method of
The method of
The method of
For further explanation,
The method of
In the method of
For further explanation,
The method of
The method of
For further explanation,
The method of
Administering (704) the migration of the logical partition may include communicating with the source and destination hypervisor to initiate the logical partition migration, and orchestrate the migration after particular steps. For example, migrating (406) the logical partition may include first pausing the logical partition then copying the logical partition from the source to the destination. Although in some embodiments, the source and destination hypervisors may communicate directly to effect such a migration, in other embodiments, the source hypervisor may inform the management console upon the logical partition being paused, and the management console may initiate the copy of the logical partition. Further, the management console may perform various operations to prepare the destination computing system for the migration, prior to the logical partition being copied. Configuring memory space for the logical partition (including creating a hardware page table or updating an existing hardware page table for the logical partition) is one example preparation that may be initiated by the management console.
To that end, the method of
In other embodiments, the migration may be initiated, not by user direction or the management console, but dynamically in response to various conditions. To that end, the method of
The present invention may be a system, a method, and/or a computer program product. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.
The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.
Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.
Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.
Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.
These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.
It will be understood from the foregoing description that modifications and changes may be made in various embodiments of the present invention without departing from its true spirit. The descriptions in this specification are for purposes of illustration only and are not to be construed in a limiting sense. The scope of the present invention is limited only by the language of the following claims.
This application is a continuation application of and claims priority from U.S. patent application Ser. No. 15/299,531, filed Oct. 21, 2016.
| Number | Name | Date | Kind |
|---|---|---|---|
| 7209994 | Klaiber et al. | Apr 2007 | B1 |
| 7240364 | Branscomb | Jul 2007 | B1 |
| 7574537 | Arndt et al. | Aug 2009 | B2 |
| 7613898 | Haertel et al. | Nov 2009 | B2 |
| 7734843 | Bender et al. | Jun 2010 | B2 |
| 7813366 | Freimuth et al. | Oct 2010 | B2 |
| 7882326 | Armstrong et al. | Feb 2011 | B2 |
| 7937518 | Boyd et al. | May 2011 | B2 |
| 7984262 | Battista et al. | Jul 2011 | B2 |
| 8219988 | Armstrong et al. | Jul 2012 | B2 |
| 8321722 | Tanaka et al. | Nov 2012 | B2 |
| 8327086 | Jacobs et al. | Dec 2012 | B2 |
| 8429446 | Hara et al. | Apr 2013 | B2 |
| 8533713 | Dong | Sep 2013 | B2 |
| 8561065 | Cunningham et al. | Oct 2013 | B2 |
| 8561066 | Koch et al. | Oct 2013 | B2 |
| 8607230 | Hatta et al. | Dec 2013 | B2 |
| 8621120 | Bender et al. | Dec 2013 | B2 |
| 8645755 | Brownlow et al. | Feb 2014 | B2 |
| 8677356 | Jacobs et al. | Mar 2014 | B2 |
| 8683109 | Nakayama et al. | Mar 2014 | B2 |
| 8875124 | Kuzmack et al. | Oct 2014 | B2 |
| 8984240 | Aslot et al. | Mar 2015 | B2 |
| 9032122 | Hart et al. | May 2015 | B2 |
| 9047113 | Iwamatsu et al. | Jun 2015 | B2 |
| 9304849 | Arroyo et al. | Apr 2016 | B2 |
| 9317317 | Graham et al. | Apr 2016 | B2 |
| 9501308 | Arroyo et al. | Nov 2016 | B2 |
| 9552233 | Tsirkin et al. | Jan 2017 | B1 |
| 20020083258 | Bauman et al. | Jun 2002 | A1 |
| 20030050990 | Craddock et al. | Mar 2003 | A1 |
| 20030204648 | Arndt | Oct 2003 | A1 |
| 20040064601 | Swanberg | Apr 2004 | A1 |
| 20040205272 | Armstrong et al. | Oct 2004 | A1 |
| 20040243994 | Nasu | Dec 2004 | A1 |
| 20060179177 | Arndt et al. | Aug 2006 | A1 |
| 20060195618 | Arndt et al. | Aug 2006 | A1 |
| 20060195620 | Arndt et al. | Aug 2006 | A1 |
| 20060281630 | Bailey et al. | Dec 2006 | A1 |
| 20070157197 | Neiger et al. | Jul 2007 | A1 |
| 20070260768 | Bender et al. | Nov 2007 | A1 |
| 20080005383 | Bender | Jan 2008 | A1 |
| 20080114916 | Hummel et al. | May 2008 | A1 |
| 20080147887 | Freimuth et al. | Jun 2008 | A1 |
| 20090133016 | Brown et al. | May 2009 | A1 |
| 20090133028 | Brown et al. | May 2009 | A1 |
| 20090249366 | Sen et al. | Oct 2009 | A1 |
| 20090276773 | Brown et al. | Nov 2009 | A1 |
| 20100036995 | Nakayama | Feb 2010 | A1 |
| 20100250824 | Belay | Sep 2010 | A1 |
| 20100262727 | Arndt | Oct 2010 | A1 |
| 20110197003 | Serebrin et al. | Aug 2011 | A1 |
| 20110320860 | Coneski et al. | Dec 2011 | A1 |
| 20120042034 | Goggin et al. | Feb 2012 | A1 |
| 20120131576 | Hatta | May 2012 | A1 |
| 20120137288 | Barrett et al. | May 2012 | A1 |
| 20120137292 | Iwamatsu | May 2012 | A1 |
| 20120151473 | Koch et al. | Jun 2012 | A1 |
| 20120167082 | Kumar | Jun 2012 | A1 |
| 20120179932 | Armstrong et al. | Jul 2012 | A1 |
| 20130086298 | Alanis et al. | Apr 2013 | A1 |
| 20130159572 | Graham et al. | Jun 2013 | A1 |
| 20130191821 | Armstrong et al. | Jul 2013 | A1 |
| 20140149985 | Takeuchi | May 2014 | A1 |
| 20140181801 | Voronkov et al. | Jun 2014 | A1 |
| 20140245296 | Sethuramalingam et al. | Aug 2014 | A1 |
| 20140281263 | Deming | Sep 2014 | A1 |
| 20140351471 | Jebson et al. | Nov 2014 | A1 |
| 20140372739 | Arroyo et al. | Dec 2014 | A1 |
| 20150006846 | Youngworth | Jan 2015 | A1 |
| 20150052282 | Dong | Feb 2015 | A1 |
| 20150120969 | He | Apr 2015 | A1 |
| 20150193248 | Noel et al. | Jul 2015 | A1 |
| 20150193250 | Ito et al. | Jul 2015 | A1 |
| 20150229524 | Engebretsen et al. | Aug 2015 | A1 |
| 20150317274 | Arroyo et al. | Nov 2015 | A1 |
| 20150317275 | Arroyo et al. | Nov 2015 | A1 |
| 20160246540 | Blagodurov | Aug 2016 | A1 |
| 20160350097 | Mahapatra et al. | Dec 2016 | A1 |
| 20170046184 | Tsirkin et al. | Feb 2017 | A1 |
| 20170199768 | Arroyo et al. | Jul 2017 | A1 |
| Number | Date | Country |
|---|---|---|
| 101488092 | Jul 2009 | CN |
| 104737138 | Jun 2015 | CN |
| 2012-113660 | Jun 2012 | JP |
| 5001818 | Aug 2012 | JP |
| Entry |
|---|
| Ajila et al., “Efficient Live Wide Area VM Migration With IP Address Change Using Type II Hypervisor”, 2013 IEEE 14th International Conference on Information Reuse and Integration (IRI2013), Aug. 2013, pp. 372-379, IEEE Xplore Digital Library (online), DOI: 10.1109/IRI.2013.6642495. |
| PCI-SIG, “Single Root I/O Virtualization and Sharing Specification-Revision 1.0”, Sep. 2007, PCI-SIG Specifications Library, pcisig.com (online), URL: pcisig.com/specifications/iov/single—root/. |
| Axnix et al. “IBM z13 firmware innovations for simultaneous multithreading and I/O virtualization”, IBM Journal of Research and Development, Jul./Sep. 2015, vol. 59, No. 4/5, 11-1, International Business Machines Corporation (IBM), Armonk, NY. |
| Salapura et al., “Resilient cloud computing”, IBM Journal of Research and Development, Sep./Oct. 2013, vol. 57, No. 5, 10-1, 12 pages, International Business Machines Corporation (IBM), Armonk, NY. |
| Challa, “Hardware Based I/O Virtualization Technologies for Hypervisors, Configurations and Advantages—A Study”, 2012 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM), Oct. 2012, pp. 99-103, IEEE Xplore Digital Library (online), DOI: 10.1109/CCEM.2012.6354610. |
| Xu et al., “Multi-Root I/O Virtualization Based Redundant Systems”, 2014 Joint 7th International Conference on Soft Computing and Intelligent Systems (SCIS) and 15th International Symposium on Advanced Intelligent Systems (ISIS), Dec. 2014, pp. 1302-1305, IEEE Xplore Digital Library (online), DOI: 10.1109/SCIS-ISIS.2014.7044652. |
| AUS920160393US2, Appendix P; List of IBM Patent or Applications Treated as Related, Mar. 23, 2017, 2 pages. |
| AUS920160393US2, Appendix P; List of IBM Patent or Applications Treated as Related, Jul. 19, 2017, 2 pages. |
| Xu et al., SRVM: Hypervisor Support for Live Migration with Passthrough SR-IOV Network Devices, Proceedings of the 12th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE'16), Apr. 2016, pp. 65-77, ACM New York, NY, USA. |
| Huang et al., Nomad: Migrating OS-bypass Networks in Virtual Machines, Proceedings of the 3rd International Conference on Virtual Execution Environments (VEE'07), Jun. 2007, pp. 158-168, ACM New York, NY, USA. |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 15299531 | Oct 2016 | US |
| Child | 15467052 | US |
