This invention relates generally to processing within a computing environment, and more particularly to switch failover control in a multi-processor computer system.
Today's computing platforms and processing system are moving toward an I/O interconnect topology that provides a single communication path between each peripheral device and the host. These computing platforms and processing systems may use packetized communications within the tree structure. Examples of such computing platforms and processing systems include what is referred to as, for example, peripheral component interconnect (PCI) systems and PCI Express (PCIe) systems. Peripheral devices are discovered by such platforms and systems through an enumeration process performed by a host system element.
I/O adapters or peripheral devices are connected to a primary processor node and to a secondary failover processor node through its I/O expansion network consisting of pairs of interconnected proprietary PCIe fan-output chips (i.e., switches), which can be costly.
An exemplary embodiment is a computer implemented method for failover control within a computer system. The method includes maintaining a primary device table entry (DTE) in a first device table activated for a first adapter in communication with a first host processor node having a first root complex via a first switch assembly and maintaining a secondary DTE in standby for a second adapter in communication with a second host processor node having a second root complex via a second switch assembly; maintaining a primary DTE in a second device table activated for the second adapter and maintaining a secondary DTE in standby for the first adapter; and upon a failover condition, updating the secondary DTE in the first device table as an active entry for the second adapter and forming a failover path to enable traffic to route from the second adapter through the second switch assembly over to the first switch assembly and up to the first root complex of the first host processor node.
Another exemplary embodiment is a computing system for failover control. The system includes a first host processor node with a first root complex in communication with a first adapter via a first switch assembly; a second host processor node with a second root complex in communication with a second adapter via a second switch assembly; a first device table in the first host processor node, the first device table includes a primary device table entry (DTE) activated for the first adapter and a secondary DTE in standby for the second adapter; and a second device table in the second host processor node, the second device table includes a primary (DTE) activated for the second adapter and a secondary DTE in standby for the first adapter; wherein the computing system has a failover condition, in which the secondary DTE in the first device table is updated as an active entry for the second adapter, and in which a failover path is formed and traffic is routed from the second adapter through the second switch assembly over to the first switch assembly and up to the first root complex of the first host processor node.
A further exemplary embodiment is a computer program product for failover control in a computer system. The computer program product includes a tangible storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method comprising: maintaining a primary device table entry (DTE) in a first device table activated for a first adapter in communication with a first host processor node having a first root complex via a first switch assembly and maintaining a secondary DTE in standby for a second adapter in communication with a second host processor node having a second root complex via a second switch assembly; maintaining a primary DTE in a second device table activated for the second adapter and maintaining a secondary DTE in standby for the first adapter; and upon a failover condition, updating the secondary DTE in the first device table as an active entry for the second adapter and forming a failover path to enable traffic to route from the second adapter through the second switch assembly over to the first switch assembly and up to the first root complex of the first host processor node.
Additional features and advantages are realized through the techniques of the present invention. Other embodiments and aspects of the invention are described in detail herein and are considered a part of the claimed invention. For a better understanding of the invention with advantages and features, refer to the description and to the drawings.
Referring now to the drawings wherein like elements are numbered alike in the several FIGURES:
Exemplary embodiments of the present invention provide systems and methods for failover control in a multi-processor computing system. In an exemplary embodiment, the computing system supports controlled and surprise failover conditions as well as control failback conditions by enabling traffic from a failed processor node to flow over an operational or surviving processor node. One way of establishing if an adapter is operational is described in commonly assigned U.S. patent application Ser. No. 12/821,181, entitled “CONTROLLING THE SELECTIVELY SETTING OF OPERATIONAL PARAMETERS FOR AN ADAPTER” Jun. 23, 2010, which is incorporated by reference herein in its entirety. This can be accomplished using off-the-shelf Peripheral Component Interconnect express (PCIe) fan-out chips (i.e., switches) according to an exemplary embodiment. In accordance with one exemplary embodiment, the computing system has device table entries for each peripheral device at each processor node and assigns unique PCI address routing ranges and unique PCI bus numbers to all the components within the input/output (I/O) expansion network of the computing system. In accordance with an exemplary embodiment, the computing system includes a master function table with function table entries (FTEs) that each represents an end point (peripheral devices) and describes or identifies the active/operational and failover routes between the host (processor node) and the endpoint and indicates to each node which path to use. These exemplary embodiments minimize traffic disruption and failover latencies in a PCIe environment in which the system is implemented in accordance with an exemplary embodiment.
Briefly stated, current large computer systems or servers are often composed of several processor nodes, each having processors, memory, a memory controller and one or more I/O hubs providing one or more PCI ports. I/O devices or adapters are attached to the PCI ports through PCIe switches to provide connectivity as shown in
In large, multi-node servers, it is desirable to allow one or more processor nodes to be taken off-line for maintenance or reconfiguration (e.g., upgrades) while the other nodes remain operational and on-line. The processor node to be taken off-line is first quiesed by moving its work to the other processor nodes. When a processor node is taken off-line, it can no longer access its attached I/O. To make this I/O operational when its processor node is taken off-line, a secondary path or alternate route is provided according to an exemplary embodiment. Using the example above, if processor node 2 is taken off-line, connectivity is lost to I/O adapters 2A and 2B. However, a failover path between a PCIe switch connected to processor node 1 and a PCIe switch connected to processor node 2 may be activated by system firmware giving processor node 1 access to I/O adapters 2A and 2B in accordance with an exemplary embodiment. As such, one of the processor nodes can be evacuated in a controlled manner, which means that all memory in one processor node can be moved into another processor node on the fly. This process is referred to as a ‘controlled failover’, which is one type of failover condition or event.
Another case where the failover path is useful is in the case of a node or hub failure. This process is referred to as ‘surprise failover’, which is another type of failover condition. Although often disruptive to processing operations, the failover path can be activated to give access to the I/O adapters owned by the failed processor node, referred to herein as the ‘affected’ adapters or ‘failover’ adapters, to a surviving processor node according to an exemplary embodiment.
When the processor node is ready to be put back on-line, the ‘affected’ I/O is attached to the repaired or reconfigured processor node according to an exemplary embodiment. This process is referred to as ‘failback’.
Turning now to
In an exemplary embodiment, system 100 is divided into one or more host processor nodes 102A, 102B in communication with one another via a symmetric multi-processor (SMP) bus 104. Although only two host processor nodes are depicted, the system 100 may include additional processors nodes. Each of the host processor nodes 102A, 102B includes a collection of processors, memory and I/O interfaces, which are connected together by the SMP bus 104, which supports all the coherency traffic and all the memory traffic within the system. Specifically, the host processor nodes 102A, 102B include memory device, generally depicted as memory 106A and 106B respectively, a processor 108A, 108B, a memory controller 110A, 110B, and a I/O hub 112A, 112B. The I/O hubs 112A, 112B respectively include conventional or standard root complexes 114A, 114B, which may be PCIe defined according to an exemplary embodiment. The SMP bus enables the processor nodes 102A, 102B to be connected to one another so that, for example, if processor node 102A wants data from the memory 106B of processor node 102B it does so through the SMP bus 104. Or, if processor node 102A modifies a piece of memory, it can tell processor node 102B that has a copy that it modified that copy and that the local copy in processor node 102B is no longer valid.
When the system 100 is in operation, each processor 108A, 108B is configured to execute instructions stored within system memory, to communicate data to and from system memory, and generally to control operations of the system 100 pursuant to the instructions. The processors 108A, 108B described herein can be any custom made or commercially available processor, a central processing unit (CPU), an auxiliary processor among several processors associated with the computer system, a semiconductor based micro-processor (the form of a microchip or chip set), a macro-processors, or generally any device for executing instructions. In an exemplary embodiment, each processor comprises a combination of hardware and/or software/firmware with a computer program that, when loaded and executed, permits the processor to operate such that it carries out the methods described herein. Although only one processor is depicted for each host processor node, each node may include additional processors. The processors execute the failover control systems and methods as described herein.
The system 100 may also comprise a switching environment, in particular switching fabric or switch assembly 120A, 120B to route packetized communications between the root complexes 112A, 112B and I/O peripheral devices or adapters, which are generally depicted as adapters 122A, 122B. Of course, one or more adapters can be located at each processor node; however, for ease of discussion only one adapter is shown at each processor node. Each switching assembly 120A, 120B may include a number of bridges 124, generally labeled as B1-BA, to provide the switching functionality. The bridges 124 may provide switching functionality in accordance with PCIe systems. Although only eight bridges for each switch assembly are shown in
Communication paths 130A, 130B couple the ports of I/O hubs 112A and 112B to switch assemblies 120A and 120B respectively, while communication paths 132 couple the ports of the bridges with the I/O adapters. The communication paths can be PCIe defined interfaces according to an exemplary embodiment. The communication paths 130A and 130B can each be PCIe defined interfaces that each allows one or more adapters to communicate with their respective root complex via their respective switch assembly during a normal mode operation. In general, all traffic from adapters coupled to switch assembly 120A goes through switch assembly 120A and up to processor node 102A while all traffic from adapters coupled to switch assembly 120B goes through switch assembly 120B and up to processor node 102B. For example, traffic from adapter 122A goes through communication path 132 up to switch assembly 120A (via bridge B8 and B1) to I/O hub 112A via communication path 130A. In this example, traffic from adapter 122B goes through communication path 132 up to switch assembly 120B (via bridge B6 and B1) to I/O hub 112B via communication path 130B during normal mode operation.
As used herein, the term “downstream” may be used to refer to communications in the direction from hub to adapter (e.g., from hub 112A to adapter 122A), while the term “upstream” may be used to refer to communications in the direction from adapter to I/O hub (e.g., from adapter 122A to I/O hub 112A). Although system 100 is illustrated as having several separate elements, one or more of the elements may be combined or more tightly integrated and may be implemented by combinations of software-configured elements, such as processors including digital signal processors (DSPs), and/or other hardware elements.
In an exemplary embodiment, a failover path 140 is defined between switch assembly 120A and 120B. During normal mode operation, the failover path 140 is disabled providing no functional traffic over the failover path 140 as shown in
In accordance with one exemplary embodiment, each I/O hub includes and maintains a device table with a device table entry (DTE) for each adapter within the system enabling each I/O hub to communicate to each adapter including ‘affected’ adapters. As such, each I/O hub has its own device table. The device table of each I/O hub keeps track of everything the I/O hub needs to communicate with each adapter. Specifically, I/O hub 112A includes a device table 150 having a primary and secondary DTE, generally depicted as DTE 154 and 156, for adapter 122A and 122B respectively. I/O Hub 112B includes a device table 152 having a primary and secondary DTE, generally depicted as DTE 158 and 160, for adapter 122B and 122A respectively. Of course, each device table may include additional primary DTEs for each adapter it owns or is located on its side of the hub and additional secondary DTEs for each adapter located opposite its side of the hub. However, only a primary and second entry for each device (adapters 122A, 122B) are discussed to illustrate exemplary embodiments of the systems and methods of the present invention.
The primary DTE at each I/O hub is for the adapter and switch assembly at the side of that I/O hub. In this example, the primary DTE 154 at I/O hub 112A is for adapter 122A and the primary DTE 158 at I/O hub 112B is for adapter 122B. The secondary DTE at each I/O hub is used for the ‘failover’ adapter and switch assembly. In this example, the secondary DTE 156 at I/O hub 112A is for adapter 122B and the secondary DTE 160 at I/O hub 112B is for adapter 122A. Secondary DTEs at each I/O hub are preloaded to support ‘failover’ adapters and the corresponding switching assembly owned by another I/O hub. In other words, device table 150 has secondary DTE 156 preloaded to support adapter 122B and switch assembly 120B in the event processor node 102B fails (surprise) or is taken offline (controlled). Similarly, device table 152 has secondary DTE 160 preloaded to support adapter 122A and switching assembly 120A in the event processor node 102A fails or is taken offline. The secondary DTE in a device table at a surviving processor node for an adapter at a failed processor node is activated as an active entry by firmware running on the surviving processor node in the event of a failover condition. Otherwise, the secondary DTE in each device table is in standby until a failover condition occurs.
According to an exemplary embodiment, each adapter 122A, 122B has an entry in each device table 150, 152 to enable the processor nodes 102A, 102B to keep track of error state information, address translation protection (ATP) information, and interrupt information of each adapter 122A, 122B. Each DTE, whether activated or in standby, points to a particular adapter as shown in
In an exemplary embodiment, the system 100 further includes a master device table or function table 170 with function table entries (FTEs) 172 each individually representing an end point or adapter within the I/O expansion network. The master device table 170 is accessible by each of the processor nodes 102A, 102B according to an exemplary embodiment. The master device table 170 includes routing information to identify to the processor in each of the first and second host processor nodes (102A, 102B) routing paths for communicating to each of the first and second adapters (122A, 122B). In the event of a failover, the surviving processor node can look to the master device table 170 for route information on how to communicate to an ‘affected’ adapter since the surviving processor node can no longer communicate to the ‘affected’ adapter at the ‘failed’ processor node through the SMP bus 104. As such, the master device table 170 is updated upon a failover event to identify new or alternate routings paths for each surviving processor node to take to communicate to each ‘affected’ adapter. For example, when processor node 102B fails or is taken off-line, processor node 102A looks to the master device table 170 for an alternate route to communicate to adapter 122B. In this example, the new route would be through I/O hub 112A down to switch assembly 120A and over to switch assembly 120B. The master device table 170 allows each processor node 102A, 102B to have complete connectivity to all adapters 122A, 122B at any given time.
In an exemplary embodiment, the device tables 150, 152 and the master device table 170 each get updated by firmware so that the surviving hub can keep track of each adapter and the surviving processor can identify which route to take to communicate to each ‘affected’ adapter. As such, with controlled failover PCI packets are not lost during a failover.
In accordance with an exemplary embodiment, each adapter 122A, 122B has a ‘bus master enable’ bit that selectively stops an adapter from generating new requests over the PCIe communication paths towards a hub. System firmware manipulates the bus master enable bit of each ‘affected’ adapter during a controlled failover or failback according to an exemplary embodiment. For example, during failover when processor node 102B is taken off-line or put back on-online (failback), the bus master enable bit of adapter 122B is turned off or reset stopping or suspending the ‘affected’ adapter (adapter 122B) from generating new requests over the PCIe bus. The bus master enable bit may be turned back on once traffic has properly moved from the failed processor node to the surviving processor node or once the failed processor node has been fixed and turned back on. Manipulating the bus master enable bit during controlled failover or failback prevents packets from being in the fabric and thus be potentially lost. System firmware also stops the processors from generating any new requests to the ‘affected’ adapters by marking all entries in the master function table 170 for ‘affected’ adapters as ‘temporarily unavailable’ according to an exemplary embodiment.
To ensure that all previous requests have completed, firmware queries a ‘busy bit’ on each of the adapters to make sure that the adapters have received all their completions according to an exemplary embodiment. Firmware also checks the status in the host to make sure the host has received all its completions according to an exemplary embodiment. As such, packets are blocked from flowing through the PCIe fabric (including switches) during the reconfiguration preventing packets from getting lost and thus potentially causing either more extensive recovery or a data integrity problem.
In accordance with an exemplary embodiment, all components within the I/O expansion network are assigned by firmware with unique PCI bus numbers and unique PCI address routing ranges in a way that allows unique address mapping and bus numbers both in the normal and failover case. As such, firmware does not have to reset every PCI-to-PCI bridge or every register making all the components unique across the whole failover domain.
In general, requests originating in the Root Complex towards the I/O adapter are ‘address routed,’ and the switches examine the PCIe address to determine the destination. To accomplish this, the PCIe address space is divided into multiple ranges, one range for each PCIe I/O adapter. Each south facing PCIe to PCIe bridge in the switch assemblies contains address registers specifying the PCIe address range that it will use to capture PCIe packets and send them downstream to an attached I/O adapter (or another PCIe switch). If the PCIe address in a packet falls within a bridge's address range, the packet is forwarded. If the PCIe address in the packet falls outside of the bridge's address range, the packet is not forwarded. In addition, the I/O adapter also examines the PCIe address in the packet to ensure that the packets are to be accepted by the I/O adapter.
System firmware is responsible for setting all the address routing registers in the bridges so that each bridge has a unique address range. It also sets the address range checking in the I/O adapter to have the same or subsets of the ranges set in the corresponding bridges. These address ranges in the I/O adapter are called BARs, or Base Address Registers. For example, if the system wants to send a request to adapter 122A attached to switch assembly 120A, it uses a PCIe address that falls within the address range set in bridge 8 (B8) and this address must also fall within the adapter's (adapter 122A) address ranges.
Even though the two root complexes in the failover domain, one in hub 112A and the other in hub 112B, could use the same PCIe address ranges in normal mode, in failover mode, the address ranges of both switches must be unique. It is therefore desirable to have unique address ranges over both root complexes in normal mode to minimize the amount of reconfiguration of address range registers in the bridges. Failover must be performed in a timely manner to avoid timeouts.
Similar to address routing, completions originating in the root complex towards the I/O adapter are ‘bus number routed,’ and the switches examine the PCIe bus number to determine the destination. To accomplish this, the PCIe bus number space is divided into multiple ranges, one range for each PCIe I/O adapter. Each south facing PCIe to PCIe bridge in the switch assembly contains bus number registers specifying the PCIe bus number range that it will use to capture PCIe packets and send them downstream to an attached I/O adapter (or another PCIe switch). If the PCIe bus number in a packet falls within a bridge's bus number range, the packet is forwarded. If the PCIe bus number in the packet falls outside of the bridge's bus number range, the packet is not forwarded. In addition, the I/O adapter also examines the PCIe bus number in the packet to ensure that the packets are to be accepted by the I/O adapter.
System firmware is responsible for setting all the bus number routing registers in the bridges so that each bridges has a unique bus number range. It also sets the bus number(s) in the I/O adapter to have the same or subsets of the ranges set in the corresponding bridges. The bus numbers in the I/O adapter are set by the bus number in certain Configuration Request packets. For example, if the system wants to send a completion to adapter 122A attached to switch assembly 120A, it uses a PCIe bus number that falls within the bus number range set in bridge 8 (B8) and this bus number must also match the bus number in adapter 122A.
Even though the two root complexes in the failover domain, one in I/O hub 112A and the other in I/O hub 112B, could use the same PCIe bus number ranges in normal mode, in failover mode, the bus number ranges of both switches must be unique. It is therefore desirable to have unique bus number ranges over both root complexes in normal mode to minimize the amount of reconfiguration of bus number range registers in the bridges. Failover must be performed in a timely manner to avoid timeouts.
Technical effects and benefits include the ability to control failover in a multi-processor system with reduced failover time using off-the-shelf PCIe fan-out chips (switch assemblies). This may minimize traffic disruption and failover latencies in a PCIe environment.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and/or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The corresponding structures, materials, acts, and equivalents of all means or step plus function elements in the claims below are intended to include any structure, material, or act for performing the function in combination with other claimed elements as specifically claimed. The description of the present invention has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the invention. The embodiment was chosen and described in order to best explain the principles of the invention and the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
As used herein, the term “controller” refers to an application specific integrated circuit (ASIC), an electronic circuit, a processor (shared, dedicated, or group) and memory that executes one or more software or firmware programs, a combinational logic circuit, and/or other suitable components that provide the described functionality.
The methods described herein can be performed or executed by any single processor, controller or a combination thereof or by multiple a number of processors, controllers or a combination thereof.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or computer program product. Accordingly, aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a “processor,” “controller” or “system.” Furthermore, aspects of the present invention may take the form of a computer program product embodied in one or more computer readable medium(s) having computer readable program code embodied thereon.
Any combination of one or more computer readable medium(s) may be utilized. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, 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), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the “C” programming language or similar programming languages. The program code 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).
Aspects of the present invention are described below 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 program instructions. These computer 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 program instructions may also be stored in a computer readable medium that can direct a computer, other programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the computer readable medium produce an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
As described above, embodiments can be embodied in the form of computer-implemented processes and apparatuses for practicing those processes. In exemplary embodiments, the invention is embodied in computer program code executed by one or more network elements. Embodiments include a computer program product 600 as depicted in
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 code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, 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 combinations of special purpose hardware and computer instructions.
Number | Name | Date | Kind |
---|---|---|---|
4104539 | Hase | Aug 1978 | A |
4644443 | Swensen et al. | Feb 1987 | A |
5027254 | Corfits et al. | Jun 1991 | A |
5170472 | Cwiakala et al. | Dec 1992 | A |
5282274 | Liu | Jan 1994 | A |
5430856 | Kinoshita | Jul 1995 | A |
5465332 | Deloye et al. | Nov 1995 | A |
5465355 | Cook et al. | Nov 1995 | A |
5535352 | Bridges et al. | Jul 1996 | A |
5551013 | Beausoleil et al. | Aug 1996 | A |
5574873 | Davidian | Nov 1996 | A |
5600805 | Fredericks et al. | Feb 1997 | A |
5617554 | Alpert et al. | Apr 1997 | A |
5742785 | Stone et al. | Apr 1998 | A |
5761448 | Adamson et al. | Jun 1998 | A |
5790825 | Traut | Aug 1998 | A |
5815647 | Buckland et al. | Sep 1998 | A |
5838960 | Harriman, Jr. | Nov 1998 | A |
5870598 | White et al. | Feb 1999 | A |
5949646 | Lee et al. | Sep 1999 | A |
5960213 | Wilson | Sep 1999 | A |
5963425 | Chrysler et al. | Oct 1999 | A |
6009261 | Scalzi et al. | Dec 1999 | A |
6023736 | Lambeth et al. | Feb 2000 | A |
6067595 | Lindenstruth | May 2000 | A |
6112311 | Beardsley et al. | Aug 2000 | A |
6205530 | Kang | Mar 2001 | B1 |
6301133 | Cuadra et al. | Oct 2001 | B1 |
6308255 | Gorishek, IV et al. | Oct 2001 | B1 |
6330656 | Bealkowski et al. | Dec 2001 | B1 |
6341064 | Bradley | Jan 2002 | B1 |
6349380 | Shahidzadeh et al. | Feb 2002 | B1 |
6408347 | Smith et al. | Jun 2002 | B1 |
6456498 | Larson et al. | Sep 2002 | B1 |
6538881 | Jeakins et al. | Mar 2003 | B1 |
6544311 | Walton et al. | Apr 2003 | B1 |
6594148 | Nguyen et al. | Jul 2003 | B1 |
6595018 | Goth et al. | Jul 2003 | B2 |
6625169 | Tofano | Sep 2003 | B1 |
6625648 | Schwaller et al. | Sep 2003 | B1 |
6643727 | Arndt et al. | Nov 2003 | B1 |
6721813 | Owen et al. | Apr 2004 | B2 |
6816590 | Pike et al. | Nov 2004 | B2 |
6845428 | Kedem | Jan 2005 | B1 |
6845469 | Hicks et al. | Jan 2005 | B2 |
6927975 | Crippen et al. | Aug 2005 | B2 |
6950438 | Owen et al. | Sep 2005 | B1 |
6973510 | Arndt et al. | Dec 2005 | B2 |
7003615 | Chiu et al. | Feb 2006 | B2 |
7004233 | Hasegawa et al. | Feb 2006 | B2 |
7007099 | Donati et al. | Feb 2006 | B1 |
7032052 | Sauber et al. | Apr 2006 | B2 |
7042734 | Hensley et al. | May 2006 | B2 |
7053502 | Aihara et al. | May 2006 | B2 |
7062594 | Sardella et al. | Jun 2006 | B1 |
7075788 | Larson et al. | Jul 2006 | B2 |
7093155 | Aoki | Aug 2006 | B2 |
7096308 | Main et al. | Aug 2006 | B2 |
7103808 | Kitamorn et al. | Sep 2006 | B2 |
7107331 | Gava et al. | Sep 2006 | B2 |
7107495 | Kitamorn et al. | Sep 2006 | B2 |
7134040 | Ayres | Nov 2006 | B2 |
7152136 | Charagulla | Dec 2006 | B1 |
7163546 | Mirizzi et al. | Jan 2007 | B2 |
7206946 | Sakakibara et al. | Apr 2007 | B2 |
7219181 | Carty | May 2007 | B2 |
7260620 | Halasz | Aug 2007 | B1 |
7313643 | Sakurai et al. | Dec 2007 | B2 |
7370224 | Jaswa et al. | May 2008 | B1 |
7447934 | Dasari et al. | Nov 2008 | B2 |
7457900 | Panesar | Nov 2008 | B2 |
7464174 | Ngai | Dec 2008 | B1 |
7478167 | Ould-Brahim et al. | Jan 2009 | B2 |
7480303 | Ngai | Jan 2009 | B1 |
7519647 | Carlough et al. | Apr 2009 | B2 |
7525957 | Scherer et al. | Apr 2009 | B2 |
7535828 | Raszuk et al. | May 2009 | B2 |
7546386 | Arndt et al. | Jun 2009 | B2 |
7558348 | Liu et al. | Jul 2009 | B1 |
7565463 | Johnsen et al. | Jul 2009 | B2 |
7594144 | Brandyberry et al. | Sep 2009 | B2 |
7613847 | Kjos et al. | Nov 2009 | B2 |
7617340 | Gregg | Nov 2009 | B2 |
7617345 | Clark et al. | Nov 2009 | B2 |
7624235 | Wadhawan et al. | Nov 2009 | B2 |
7627723 | Buck et al. | Dec 2009 | B1 |
7631097 | Moch et al. | Dec 2009 | B2 |
7660912 | Gregg | Feb 2010 | B2 |
7664991 | Gunda et al. | Feb 2010 | B1 |
7676617 | Kloeppner | Mar 2010 | B2 |
7729316 | Uhlik | Jun 2010 | B2 |
7836254 | Gregg et al. | Nov 2010 | B2 |
7873851 | Linnell et al. | Jan 2011 | B1 |
7975076 | Moriki et al. | Jul 2011 | B2 |
8032684 | Pettey et al. | Oct 2011 | B2 |
8041811 | Calippe et al. | Oct 2011 | B2 |
8046627 | Takubo | Oct 2011 | B2 |
8082466 | Tanaka et al. | Dec 2011 | B2 |
8140917 | Suetsugu et al. | Mar 2012 | B2 |
8510592 | Chan | Aug 2013 | B1 |
20020112067 | Chang et al. | Aug 2002 | A1 |
20030056155 | Austen et al. | Mar 2003 | A1 |
20030058618 | Soetemans et al. | Mar 2003 | A1 |
20030097503 | Huckins | May 2003 | A1 |
20030177221 | Ould-Brahim et al. | Sep 2003 | A1 |
20030198180 | Cambron | Oct 2003 | A1 |
20030200477 | Ayres | Oct 2003 | A1 |
20040024905 | Liao et al. | Feb 2004 | A1 |
20040088604 | Bland et al. | May 2004 | A1 |
20040117534 | Parry et al. | Jun 2004 | A1 |
20040130868 | Schwartz et al. | Jul 2004 | A1 |
20040133819 | Krishnamurthy et al. | Jul 2004 | A1 |
20040136130 | Wimmer et al. | Jul 2004 | A1 |
20040199700 | Clayton | Oct 2004 | A1 |
20050033895 | Lueck et al. | Feb 2005 | A1 |
20050071472 | Arndt et al. | Mar 2005 | A1 |
20050091438 | Chatterjee | Apr 2005 | A1 |
20050116546 | Zeighami et al. | Jun 2005 | A1 |
20050144533 | LeVangia et al. | Jun 2005 | A1 |
20050146855 | Brehm et al. | Jul 2005 | A1 |
20050160214 | Sauber et al. | Jul 2005 | A1 |
20050162830 | Wortman et al. | Jul 2005 | A1 |
20050182862 | Ritz et al. | Aug 2005 | A1 |
20050213513 | Ngo et al. | Sep 2005 | A1 |
20050276017 | Aziz et al. | Dec 2005 | A1 |
20050289271 | Martinez et al. | Dec 2005 | A1 |
20050289278 | Tan et al. | Dec 2005 | A1 |
20060067069 | Heard et al. | Mar 2006 | A1 |
20060085150 | Gorin | Apr 2006 | A1 |
20060085573 | Pike et al. | Apr 2006 | A1 |
20060087813 | Becker et al. | Apr 2006 | A1 |
20060087814 | Brandon et al. | Apr 2006 | A1 |
20060095607 | Lim et al. | May 2006 | A1 |
20060195644 | Arndt et al. | Aug 2006 | A1 |
20060206639 | Tee et al. | Sep 2006 | A1 |
20060230208 | Gregg et al. | Oct 2006 | A1 |
20060236054 | Kitamura | Oct 2006 | A1 |
20060253619 | Torudbakken et al. | Nov 2006 | A1 |
20060288130 | Madukkarumukumana et al. | Dec 2006 | A1 |
20060291447 | Siliquini et al. | Dec 2006 | A1 |
20070069585 | Bosco et al. | Mar 2007 | A1 |
20070073955 | Murray et al. | Mar 2007 | A1 |
20070078996 | Chen et al. | Apr 2007 | A1 |
20070115230 | Tajiri et al. | May 2007 | A1 |
20070136554 | Biran et al. | Jun 2007 | A1 |
20070168636 | Hummel et al. | Jul 2007 | A1 |
20070183393 | Body et al. | Aug 2007 | A1 |
20070186074 | Bradford et al. | Aug 2007 | A1 |
20070226386 | Sharp et al. | Sep 2007 | A1 |
20070226523 | Chang | Sep 2007 | A1 |
20070234018 | Feiste | Oct 2007 | A1 |
20070239925 | Koishi | Oct 2007 | A1 |
20070245041 | Hua et al. | Oct 2007 | A1 |
20070271559 | Easton et al. | Nov 2007 | A1 |
20070273018 | Onozuka et al. | Nov 2007 | A1 |
20070274039 | Hamlin | Nov 2007 | A1 |
20080043405 | Lee et al. | Feb 2008 | A1 |
20080069141 | Bonaguro et al. | Mar 2008 | A1 |
20080091851 | Sierra | Apr 2008 | A1 |
20080091868 | Mizrachi et al. | Apr 2008 | A1 |
20080091915 | Moertl et al. | Apr 2008 | A1 |
20080114906 | Hummel et al. | May 2008 | A1 |
20080126648 | Brownlow et al. | May 2008 | A1 |
20080126652 | Vembu et al. | May 2008 | A1 |
20080147943 | Freimuth et al. | Jun 2008 | A1 |
20080148295 | Freimuth et al. | Jun 2008 | A1 |
20080162865 | Koufaty et al. | Jul 2008 | A1 |
20080168208 | Gregg | Jul 2008 | A1 |
20080189577 | Arndt et al. | Aug 2008 | A1 |
20080209114 | Chow et al. | Aug 2008 | A1 |
20080222406 | Tabuchi | Sep 2008 | A1 |
20080235425 | Belmar et al. | Sep 2008 | A1 |
20080239687 | Leigh et al. | Oct 2008 | A1 |
20080239945 | Gregg | Oct 2008 | A1 |
20080259555 | Bechtolsheim et al. | Oct 2008 | A1 |
20080263246 | Larson et al. | Oct 2008 | A1 |
20080270853 | Chagoly et al. | Oct 2008 | A1 |
20080288661 | Galles | Nov 2008 | A1 |
20090037682 | Armstrong et al. | Feb 2009 | A1 |
20090070760 | Khatri et al. | Mar 2009 | A1 |
20090125666 | Freking et al. | May 2009 | A1 |
20090144462 | Arndt et al. | Jun 2009 | A1 |
20090144731 | Brown et al. | Jun 2009 | A1 |
20090182966 | Greiner et al. | Jul 2009 | A1 |
20090182969 | Norgaard et al. | Jul 2009 | A1 |
20090210527 | Kawato | Aug 2009 | A1 |
20090210646 | Bauman et al. | Aug 2009 | A1 |
20090222814 | Astrand | Sep 2009 | A1 |
20090234987 | Lee et al. | Sep 2009 | A1 |
20090240849 | Corneli et al. | Sep 2009 | A1 |
20090249039 | Hook et al. | Oct 2009 | A1 |
20090276774 | Kinoshita | Nov 2009 | A1 |
20090328035 | Ganguly | Dec 2009 | A1 |
20100005234 | Ganga et al. | Jan 2010 | A1 |
20100027559 | Lin et al. | Feb 2010 | A1 |
20100077117 | Asnaashari | Mar 2010 | A1 |
20100115329 | Tanaka et al. | May 2010 | A1 |
20100146089 | Freimuth et al. | Jun 2010 | A1 |
20100157463 | Arizono et al. | Jun 2010 | A1 |
20100211714 | LePage | Aug 2010 | A1 |
20100312894 | Awad et al. | Dec 2010 | A1 |
20110131359 | Pettey et al. | Jun 2011 | A1 |
20110320653 | Lais et al. | Dec 2011 | A1 |
20110320675 | Gregg et al. | Dec 2011 | A1 |
Number | Date | Country |
---|---|---|
1885096 | Dec 2006 | CN |
101196615 | Jun 2008 | CN |
101571631 | Nov 2009 | CN |
102193239 | Sep 2011 | CN |
57191826 | Nov 1982 | JP |
5981724 | May 1984 | JP |
6279557 | Apr 1987 | JP |
0553973 | Mar 1993 | JP |
2007087082 | Apr 2007 | JP |
2007241526 | Sep 2007 | JP |
2010134627 | Jun 2010 | JP |
WO9600940 | Nov 1996 | WO |
2009027189 | Mar 2008 | WO |
Entry |
---|
Final Office Action Received Mar. 2, 2012 for U.S. Appl. No. 12/821,221. |
International Search Report for PCT/EP2011/059810, Sep. 14, 2011. |
Non Final Office Action dated Sep. 22, 2011 for U.S. Appl. No. 12/821,221. |
PCI Express Base Specification Rev 1.0a, Apr. 15, 2003 p. 1-2, 31, 35-36, 43-44, 49-51, 55, 59-62, 74, 101. |
Dolphin Interconnect Solutions; MySQL Acceleration Solutions; Solid State Storage; Embeded and HPC Solutions; “DXH510 PCI Express Host Adapter”; ww.dolphinics.com/products/pent-dxseries-dsh510.html downloaded Jun. 10, 2010. |
J. Regula, “Using Non-transparent Bridging in PCI Express Systems”, PLX Technology, Inc., pp. 1-31, Jun. 1, 2004. |
Jack Regula “Ethernet Tunneling through PCI Express Inter-Processor Communication, Low Latency Storage IO Source”; www.wwpi.com; Publisher: Computer Technology Review, Jan. 19, 2009. |
Robert F. Kern, “IBM System z & DS8000 Technology Synergy”, IBM ATS Americas Disk Storage; Jul. 21, 2009, pp. 1-25. |
Szwed et al.; “Managing Connected PCI Express Root Complexes”; Dated: Dec. 23, 2009—6 pages. |
Final Office Action mail date Jun. 15, 2011 for U.S. Appl. No. 12/821,221. |
U.S. Appl. No. 12/821,221, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,222, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,224, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,226, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,239, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,242, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,243, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,245, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,247, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,248, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,250, filed Jun. 23, 2010. |
U.S. Appl. No. 12/821,271, filed Jun. 23, 2010. |
Baumann, Andrew, et al., “The Multikernel: A New OS Architecture for Scalable Multicore Systems,” Oct. 2009, SOSP'09, Oct. 11-14, 2009, Big Sky, Montana, USA, pp. 29-43. |
Crawford et al. “Accelerating Computing with the Cell Broadband Engine Processor”; CF'08, May 5-7, 2008; Ischia, Italy; Copyright 2008 ACM 978-1-60558-077. |
Darren Abramson et al.; “Intel Virtualization Technology for Directed I/O”; Intel Technology Journal, vol. 10, Issue 3, Aug. 10, 2006; pp. 1-16. |
Huang, Wei et al., “A Case for High Performance Computing with Virtual Machines,” ISC '06, Jun3 28 30, Carins, Queensland, Australia, pp. 125-134, Jun. 3, 2006. |
“Intel (registered trademark) Itanium (registered trademark) Architecture Software Developer's Manual,” vol. 2, Rev. 2.2, Jan. 2006. |
“z/VM: General Information Manual,” IBM Publication No. GC24-5991-05, May 2003. |
“DMA Engines Bring Mulicast to PCI Express Systems,” http://electronicdesign.com, Aug. 13, 2009, 3 pages. |
“I/O Virtualization and AMD's IOMMU,” AMD Developer Central, http://developer.amd.com/documentation/articles/pages.892006101.aspx, Aug. 9, 2006. |
“IBM Enhances the IBM eServer zSeries 990 Family of Servers,” Hardware Announcement, Oct. 7, 2003, pp. 1-11. |
Internet Article, “Large Page Support in the Lunux Kernel,” http://lwn.net./Articles/6969/<retrieved on Jan. 26, 2010>. |
K. Vaidyanathan et al.; “Exploiting RDMA Operations for Providing Efficient Fine-Grained Resource Monitoring in Cluster-Based Servers”; Jun. 2006; pages -10; Downloaded: Apr. 13, 2010 at 18:53:46 UTC from IEEE Xplore. 1-4244-0328-6/06. |
Mysore, Shashidhar et al., “Understanding and Visualizing Full Systems with Data Flow Tomography” SPOLOS '08, Mar. 1-5, 2008, Seattle, Washington, USA, pp. 211-221. |
Narayanan Ganapathy et al.; Papers-USENIX Annual Teleconference (No. 98); Entitled: “General Purpose Operating System Support for Multiple Page Sizes” 1998; pp. 1-9. |
U.S. Appl. No. 12/821,221, Non-Final Office Action mail date Jan. 10, 2011. |
Paulsen, Erik; “Local Memory Coaxes Top Speed from SCSI Masters”; Electronic Design, v. 41, (Apr. 15, 1993) p. 75-6+. |
Swift, Micael M. et al., “Improving the Reliability of Commodity Operating Systems,” ACM Transactions on Computer Systems, vol. 23, No. 1, Feb. 2005, pp. 77-110. |
Talluri et al., “A New Page Table for 64-bit Address Spaces,” ACM SIGOPS Operating Systems Review, vol. 29, Issue 5 (Dec. 1995), pp. 194-200. |
VTdHowTo—Xen Wiki; Downloaded—Apr. 16, 2010; pp. 1-5; http://wiki.xensource.com/xenwiki/VTdHowTo. |
Winwood, Simon, et al., “Multiple Page Size Support in the Linux Kernel”, Proceedings of Ottawa Linux Symposium, 2002. |
Xu, Min et al., “Towards a VMM-based Usage Control Framework for OS Kernel Integrity Protection,” SACMAT '07, Jun. 20-22, 2007, Sophia Antipolis, France, pp. 71-80. |
z/Architecture Principles of Operation, Feb. 2009; pp. 1-1344. |
z/VM: Running Guest Operating Systems, IBM Publication No. SC24-5997-02, Oct. 2001. |
Final Office Action dated Jul. 19, 2012 for U.S. Appl. No. 12/821,250. |
Non Final Office Action recieved May 8, 2012 for U.S. Appl. No. 12/821,243. |
Non Final Office Action received Mar. 15, 2012 for U.S. Appl. No. 12/821,242. |
Non Final Office Action dated Apr. 3, 2012 for U.S. Appl. No. 12/821,226. |
Non Final Office Action for U.S. Appl. No. 12/821,224 dated May 15, 2012. |
Non-final Office Action dated Jun. 5, 2012 for U.S. Appl. No. 12/821,221. |
Notice of Allowance recieved Jun. 1, 2012 for U.S. Appl. No. 12/821,248. |
Restriction Requirement received Jun. 18, 2012 for U.S. Appl. No. 12/821,239. |
Non Final Office Action received Feb. 16, 2012 for U.S. Appl. No. 12/821,250. |
Final Office Action dated Jun. 14, 2013 for U.S. Appl. No. 12/821,239. |
Informational Materials for IDS, date May 8, 2013, 4 pages. |
International Search Report of the Patent Cooperation Treaty for International Application No. PCT/CN2013/070828, mailing date Apr. 24, 2013, 13 pages. |
Number | Date | Country | |
---|---|---|---|
20110320861 A1 | Dec 2011 | US |