Abstract: A method implemented by a Broadband Network Gateway (BNG) of an Internet service provider to provide accessibility to a wide area network for a Residential Gateway (RG) upon a failure of a wireline connectivity between the BNG and the RG, the method including receiving a failure detect message indicating a connectivity failure at the BNG from the RG, deciding whether to re-route traffic by the BNG, sending a failure acknowledge message by the BNG to the RG notifying the RG that re-routing has been initiated, sending a traffic re-route request message by the BNG to a Packet Data Network Gateway (PDN GW) of a Long-Term Evolution (LTE) network requesting the PDN GW to re-route traffic, receiving a traffic re-route acknowledgement by the BNG from the PDN GW, and re-routing traffic between the RG and the BNG through the PDN GW by the BNG.

Abstract: A first virtual I/O server (VIOS) provides a cluster aware (CA) operating system (OS) executing on a processor resource of the first VIOS to register the first VIOS within a VIOS cluster. The first VIOS comprises a first field/failure data capture (FFDC) module that executes within the first VIOS and performs the functions of: receiving from an event listener a signal indicating that an FFFDC event/condition has been detected by the first VIOS; and automatically transmitting FFDC data to the shared storage repository for storage of the FFDC data within the shared storage repository. The FFDC module further performs the functions of: transmitting to one or more second VIOSes within the VIOS cluster, one or more messages to inform the one or more second VIOSes of an occurrence of the FFDC event/condition that was detected by the first VIOS.

Abstract: Providing high availability multi-way conferencing. Separate signaling and media components may be provided within an MCU or among a cluster of MCUs. A signaling server may control signaling aspects of a conference while a media server may provide media support for the conference. In the event of media server failure, the signaling server may assign a new media server to provide media support for the conference. A backup signaling server may also monitor the signaling server and may provide signaling support for the conference in the event of signaling server failure.

Abstract: A system comprises at least two random access memory (RAM) elements arranged to store data redundantly. The system further comprises RAM routing logic comprising comparison logic operably coupled to the at least two RAM elements and arranged to compare redundant data read from the at least two RAM elements, and check and validation logic, independent of the RAM routing logic, operably coupled to the at least two RAM elements and arranged to additionally detect an error in the redundant data read from the at least two RAM elements and provide an error indication signal to the RAM routing logic in response thereto. The RAM routing logic further comprises selection logic arranged to dynamically select redundant data from one of the at least two RAM elements based on the comparison of the redundant data and the error indication signal.

Abstract: An electric device is capable of transmitting and receiving data relative to an external device. The electric device includes an error detection unit for detecting an error occurred in the electric device; an error restoration command holding unit for holding an error restoration command to restore the error; and an error analyzing unit for analyzing the error occurred in the electric device and retrieving the error restoration command from the error restoration command holding unit according to an analysis result of the error. In the electric device, when the error detection unit detects the error, the error restoration command is retrieved from the error restoration command holding unit according to an analysis result analyzed with the error analyzing unit.

Abstract: A redundancy manager manages commands to peripheral devices in a computer system. These peripheral devices have multiple pathways connecting it to the computer system. The redundancy manager determines the number of independent pathways connected to the peripheral device, presents only one logical device to the operating system and any device driver and any other command or device processing logic in the command path before the redundancy manager. For each incoming command, the redundancy manager determines which pathways are properly functioning and selects the best pathway for the command based at least partly upon a penalty model where a path may be temporarily penalized by not including the pathway in the path selection process for a predetermined time. The redundancy manager further reroutes the command to an alternate path and resets the device for an alternate path that is not penalized or has otherwise failed.

Abstract: A system and method are disclosed for a resilient IP ring protocol. A system that incorporates teachings of the present disclosure may include, for example, a communication device using a resilient internet protocol (IP) ring protocol, including a controller (602) that manages operations of a transceiver (620) for exchanging messages in a communication system with adjacent ring nodes in a communication ring architecture (300). The controller can be programmed to use sub-second level signaling between adjacent ring nodes to determine when adjacent ring nodes have failed and enable failure options upon detection of an adjacent ring-node failure. Additional embodiments are disclosed.

Abstract: Failing bus lane detection using syndrome analysis, including a method for receiving a plurality of syndromes of an error detection code, the error detection code associated with a plurality of frames that have been transmitted on a bus that includes a plurality of lanes and is protected by the error detection code. The method includes performing for each of the lanes in each of the syndromes: decoding the syndrome under an assumption that the lane is a failing lane, the decoding outputting a decode result; determining if the decode result is a valid decode; and voting for the lane in response to determining that the decode result is a valid decode. A failing lane is then identified in response to the voting, with the failing lane being characterized by having more votes than at least one other lane on the bus.

Abstract: Process for placing a multimedia object comprising at least one elemental data stream in memory by a terminal receiving the said elemental stream or streams, the said stream or streams being received by the terminal in the form of data packets comprising a data part and a header comprising serial numbers, comprising a stage of placing the data parts of packets received in a data object, a stage of creating a hint track comprising elemental records relating to data packets, these elemental records being placed in memory in sequence according to the serial numbers of the packets received and including a reference to the data in the said packets within the data object, and further comprising a stage of creating elemental records corresponding to non-received packets within the said hint track, these elemental records being placed in memory in the sequence in the locations where the corresponding elemental records would have been placed in memory had the packets been received.

Abstract: Techniques are provided for a management application in a first virtual network to start a first cloud gateway in the first virtual network. First messages are sent to a second virtual network, the first messages comprising information configured to start a second cloud gateway and a first virtual switch in the second virtual network. A connection is established between the first cloud gateway and the second cloud gateway, where the first cloud gateway, the second cloud gateway, and the first virtual switch form a first scalable cloud network element. One or more second messages are sent to the second virtual network, the one or more second messages comprising information configured to start a virtual machine and a first virtual machine interface configured to allow the virtual machine to access processing resources in the second virtual network. Data are stored that associates the virtual machine with the first virtual switch.

Abstract: An apparatus, system, and method are disclosed for resetting and bypassing microcontroller stations. A command module asserts and de-asserts a reset line in response to a command. A reset module resets a microcontroller station if the command module asserts and de-asserts the reset line within a time interval. In addition, the reset module bypasses the microcontroller station if the command module asserts and holds the reset line for a time period exceeding the time interval.

Abstract: A cloud system and a method for monitoring and handling abnormal states of physical machines in the cloud system are disclosed. Each physical machine of the cloud system respectively executes a daemon program for monitoring operation states of the physical machine and providing the operation states to a management terminal in the cloud system. When the management terminal determines that any physical machine is having abnormal operation states, the management terminal provides a control instruction to the cabinet of the physical machine having abnormal operation states. The physical machine having abnormal operation states is compulsorily ejected from the cabinet. Thus, it is convenient to the administrator when replacing the physical machine having abnormal operation states onsite by shortening the time looking for the faulted physical machine.

Abstract: A first module calculates a failure occurrence risk index of each data storage area address. A second module calculates a power saving index of each data storage area address. A third module calculates an access speed index per unit data volume necessary to access each data storage area address. A fourth module generates a distribution table that represents the failure occurrence risk index, the power saving index, and the access speed index for each candidate address, with respect to data to be distributed. A fifth module selects a candidate address in the distribution table such that the power saving index and the access speed index meet restricting conditions and the failure occurrence risk index is minimized, and distributes the data to the candidate address.

Abstract: A computer-implemented method for diagnosing a network configuration of a computing device is described. A test network configuration is captured. A test network signature is generated from the test network configuration. A label is assigned to the test network signature. A determination is made as to whether the test network signature is labeled as an unsuccessful network signature. If the test network signature is labeled unsuccessful, one or more procedures to change the label are generated.

Abstract: A manycore networks-on-chip (NoC) formed by a plurality of clusters is provided. The manycore NoC includes redundant routers and connection channels therefore is fault-tolerant as long as the numbers of damaged routers and damaged connection channels are under predetermined thresholds. Moreover, the NoC can retain its original logical topology without isolating any core after resetting the connection channels in response to the damaged routers and connection channels.

Abstract: An information processing apparatus includes an execution determination unit and a control unit. The execution determination unit determines whether a series of processes including multiple processes is executable at an execution time of the series of processes. The control unit selectively provides at least one recovery device for substituting for the series of processes when it is determined that the series of processes is not executable.

Abstract: Methods and systems for escalating component failures in a cloud are provided. A cloud controller of a cloud receives an indication that a collection of virtual machines of the first cloud has failed based on a collection of virtual machines escalation policy. The cloud controller initiates relocating the collection of virtual machines to a second cloud.

Abstract: Devices, methods and instructions encoded on computer readable medium for implementation of a dual-adjacency between edge devices of a network site. A first edge device comprises one or more local interfaces configured for communication, via a local network, with one or more network devices co-located in a first network site. The first edge device also comprises one or more overlay interfaces configured for communication, via a core network, with one or more network devices located in one or more other network sites connected to the core network. The first edge device comprises a processor configured to establish, via at least one of the local interfaces, a site communication channel with a second edge device co-located in the first network site. The processor is further configured to establish an overlay communication channel, via at least one of the overlay interfaces, with the second edge device.

Abstract: Memory apparatus and methods utilizing multiple bit lanes may redirect one or more signals on the bit lanes. A memory agent may include a redrive circuit having a plurality of bit lanes, a memory device or interface, and a fail-over circuit coupled between the plurality of bit lanes and the memory device or interface.

Abstract: The network system is provided. The network system includes: a first processing apparatus configured to provide a specific service; a second processing apparatus configured to provide the specific service, the first processing apparatus and the second processing apparatus having one identical address; a client apparatus configured to utilize the specific service; and a network relay apparatus connected directly or indirectly via interfaces to the first processing apparatus, the second processing apparatus, and the client apparatus and configured to relay packet transmission between the client apparatus and the first processing apparatus or the second processing apparatus, wherein the network relay apparatus forwards a received packet, which is received via the interface connecting with the client apparatus to be sent to the address as a destination, to one processing apparatus in a state enabled to provide the specific service between the first processing apparatus and the second processing apparatus.

Abstract: A system that incorporates teachings of the present disclosure may include, for example, an edge device having a controller to receive a Session Initiation Protocol (SIP) message from a user endpoint device (UE) requesting communication services, forward the SIP message to a network element of a Server Office, receive from the network element a first error message indicating communication services at the Server Office are unavailable, replace the first error message with a second error message, the second error message indicating a temporary unavailability of communication services, and transmit the second error message to the UE. Additional embodiments are disclosed.

Abstract: A method and apparatus for achieving high availability for applications and optimizing power consumption within a datacenter is provided. In one embodiment, a method for providing high availability and optimizing power consumption within a datacenter comprises upon execution of an application at a target node amongst a plurality of nodes in a datacenter, selecting a failover target node amongst the plurality of nodes for the application, and reserving a failover capacity of at least one resource of the selected failover target node.

Abstract: Method and system for managing error related events while a system is processing input/output (“I/O”) requests for accessing storage space is provided. Various components are involved in processing the I/O requests. Some of these components may also have sub-components. Events related to the various components are classified with respect to their severity levels. Threshold values for a frequency of these events is set and stored in a data structure at a memory location. When an event occurs, the severity level and the threshold value for the event are determined from the data structure. The actual frequency is then compared to the stored threshold value. If the threshold value is violated and there is an alternate path to route the I/O request, then the affected component is restricted and the alternate path is used to route the I/O request.

Abstract: Techniques are provided for isolating faults in a software program by providing at least two code paths that are capable of performing the same operation. When a fault occurs while the one of the code paths is being used to perform an operation, data that indicates the circumstances under which the fault occurred is stored. For example, a fault-recording mechanism may store data that indicates the entities that were involved in the failed operation. Because they were involved in an operation that experienced a fault, one or more of those entities may be “quarantined”. When subsequent requests arrive to perform the operation, a check may be performed to determine whether the requested operation involves any of the quarantined entities. If the requested operation involves a quarantined entity, a different code path is used to perform the operation, rather than the code path from which the entity is quarantined.

Abstract: A network, in particular an Ethernet network, contains as network elements at least two network components that are interconnected by a network transmission line. Accordingly, at least one expansion unit having two external ports is disposed in the network line for extending the scope thereof, wherein the expansion unit forwards a failure of the network transmission line at one of the ports thereof to a port of the next subsequent network element.

Abstract: A method for managing multiple nodes hosting multiple memory segments, including: identifying a failure of a first node hosting a first memory segment storing a hypervisor; identifying a second memory segment storing a shadow of the hypervisor and hosted by a second node; intercepting, after the failure, a hypervisor access request (HAR) generated by a core of a third node and comprising a physical memory address comprising multiple node identification (ID) bits identifying the first node; modifying the multiple node ID bits of the physical memory address to identify the second node; and accessing a location in the shadow of the hypervisor specified by the physical address of the HAR after the multiple node ID bits are modified.

Abstract: A method performed by network devices that includes operating in a normal mode, where the network devices form a virtual chassis that corresponds to a single logical network device; detecting when a failure within the virtual chassis occurs; executing a splitting process to form one or more new virtual chassis in correspondence to the failure; determining whether one of the one or more new virtual chassis operates as a functioning virtual chassis based on whether at least one of a set of criteria is satisfied, where the functioning virtual chassis operates according to resources configured for the virtual chassis; and operating as a nonfunctioning virtual chassis when it is determined that the one of the one or more virtual chassis does not satisfy the at least one of the set of criteria, where the nonfunctioning virtual chassis operates in a pass-through mode.

Abstract: A quorum service within a cluster infrastructure layer of a cluster environment comprising a plurality of nodes automatically triggers at least one automated fencing operation integrated within the quorum service, to reliably maintain a node usability state of each node of the plurality of nodes indicating an availability of each node to control and access at least one shared resource of the cluster. The quorum service reports the node usability state of each node as a cluster health status to at least one distributed application within an application layer of the cluster environment, to provide a reliable cluster health status of the plurality of nodes to the at least one distributed application for a failover of said at least one shared resource from control by a failed node from among the plurality of nodes to another node from among the plurality of nodes.

Abstract: A method, apparatus, and computer program product for handling a failure condition in a storage controller is disclosed. In certain embodiments, a method may include initially detecting a failure condition in a storage controller. The failure condition may be associated with a specific host and a specific storage device connected to the storage controller. The method may further include determining a failure ID associated with the failure condition. Using the failure ID, en entry may be located in a data collection and recovery table. This entry may indicate one or more data collection and/or recovery processes to execute in response to the failure condition. The method may then execute the data collection and/or recovery processes indicated in the entry. While executing the data collection and/or recovery processes, connectivity may be maintained between hosts and storage devices not associated with the failure condition.

Abstract: A first component, executing in a first data processing system, receives, over a data communication network using a first adapter, a first diagnostic heartbeat packet from a second adapter in a second data processing system. The first heartbeat packet comprises a header, a set of heartbeat parameters, and a set of diagnostic attributes. The first component determines, based on a set of values corresponding to the set of diagnostic attributes, that a soft network error condition exists in the data communication network. The soft network error condition is a network error condition that adversely affects the transmission of packets having certain properties in the data communication network. The first component stores the set of values in a state information record associated with the first component and re-routes data traffic from one link to a different link between the first and the second data processing systems.

Abstract: A first component, executing using a processor and a memory in a first data processing system, receives a diagnostic heartbeat packet from a second component executing in a second data processing system, wherein the diagnostic heartbeat packet is a packet comprising a header, a set of heartbeat parameters, and a set of diagnostic attributes. The first component determines, using a value of a diagnostic attribute in the diagnostic heartbeat packet, that a first communication link between the first and the second data processing systems is usable but includes a soft network error, wherein a soft network error condition is a network error condition that adversely affects transmission of packets having certain properties in the data communication network. The first component re-routes a synchronization message from the first component to the second component using a second communication link between the first and the second data processing systems.

Abstract: A connection expansion device connected to devices includes a plurality of ports to which devices are connected, a storage unit configured to record device information obtained from each port, and a processing unit configured to specify, based on the device information, a port in which an abnormal device exists, invalidate device information belonging to the port, and cause the storage unit to hold device information of a normal device.

Abstract: A connection control apparatus includes an active control unit that performs routing of data to a first device connected to a port of the connection control apparatus by using an address of the first device, and a standby control unit that, when the active control unit causes a failure, performs routing of data to the first device by using the address of the first device and by using route information indicating a connection route to a second device connected to a port of an adjacent connection control apparatus connected to the connection control apparatus, the standby control unit performing routing of data to the second device by using the route information upon the adjacent connection control apparatus being unable to perform routing.

Abstract: Embodiments of the present invention disclose a method for switching an NC link, a processor system, and a node, where the processor system includes more than two nodes capable of communicating with each other, each node includes a node controller NC chip, a host bus adapter HBA apparatus, and at least one CPU, the NC chip is connected to each CPU in a node where the NC chip is located, and the HBA apparatus is connected to each CPU in a node where the HBA apparatus is located; an NC link borne by the NC chip is corresponding to an HBA link borne by the HBA apparatus. By using an HBA apparatus to deploy a redundant link, the cost of deploying the redundant link is reduced effectively under a premise of ensuring the reliability of the processor system.

Abstract: Systems and methods may include providing a first silo configured to operate a first set of event-driven processing stages, and a second silo configured to operate a second set of event-driven processing stages; arranging at least one shared transaction cache between the first set of processing stages and the second set of processing stages of the respective first and second silos, where each processing stage of the first and second sets of processing stages can obtain or receive one or more transactions from the at least one shared transaction cache for processing, and where when a distressed processing stage in the first silo fails to process a first transaction designated for the distressed processing stage in a predefined amount of time, a non-distressed processing stage in the second silo obtains or receives the first transaction from the at least one shared transaction cache to process the first transaction on behalf of the distressed processing stage.

Abstract: Techniques are described for decentralizing a job scheduler in a distributed system environment. Embodiments of the invention may generally include receiving a job to be performed by a multi-nodal system which includes a cluster of nodes. Instead of a centralized job scheduler assigning the job to a node or nodes, each node has a job scheduler which scans a shared-file system to determine what job to execute on the node. In a job requiring multiple nodes, one of the nodes that joined the multi-nodal job becomes the primary node which then assigns and monitors the job's execution on the multiple nodes.

Abstract: It is provided a computer system comprising a plurality of computers; a PCI switch; and a plurality of I/O devices connected to the PCI switch, wherein the communication path includes a virtual switch and virtual bridges, and the PCI switch comprises a communication path generating module for setting the virtual switches and the virtual bridges, a virtual switch group management module for creating a virtual switch group including the at least one of the virtual switches, and setting an enabled flag to one of the virtual switches included in the virtual switch group, and a port management module for managing relation between each of the generated communication paths and the plurality of ports included in the each of the generated communication paths.

Abstract: Techniques for achieving high availability (HA) in a cloud environment are presented. Cloud storage provided to multiple tenants is accessed via a plurality of controllers via a switch. The controllers are organized in a ring and each controller is responsible for detecting failures in adjoining controllers within the ring. Storage services for the tenants are serviced without disruptions even when multiple nodes completely fail at the same time.

Abstract: High availability is provided in a storage system that offers expandability more inexpensively. Provided is a storage system including multiple expanders to be connected to multiple storage mediums, multiple cascades connected respectively to a prescribed number of expanders among the multiple expanders, and multiple control units for respectively controlling the multiple cascades. One end of the multiple cascades is connected with an inter-cascade link, and the inter-cascade link has a logically connected state and a logically disconnected state.

Abstract: A system and method for retaining routes in a control plane learned by an inter-domain routing protocol in the event of a connectivity failure between routers. Routers are classified as either route reflectors or originators. A connectivity failure between two routers is determined. A determination is then made whether to propagate a withdrawal of learned routes based on whether the connectivity failure occurred between a route reflector and an originator, two originators, or two route reflectors. A withdrawal of learned routes is propagated to neighboring routers if the connectivity failure occurred between two originators, or between a route reflector and an originator that is inaccessible via an intra-domain routing protocol. No withdrawal of learned routes is propagated if the connectivity failure occurred between two route reflectors, or between a route reflector and an originator that is accessible via an intra-domain routing protocol.

Abstract: A quorum service within a cluster infrastructure layer of a cluster environment comprising a plurality of nodes automatically triggers at least one automated fencing operation integrated within the quorum service, to reliably maintain a node usability state of each node of the plurality of nodes indicating an availability of each node to control and access at least one shared resource of the cluster. The quorum service reports the node usability state of each node as a cluster health status to at least one distributed application within an application layer of the cluster environment, to provide a reliable cluster health status of the plurality of nodes to the at least one distributed application for a failover of said at least one shared resource from control by a failed node from among the plurality of nodes to another node from among the plurality of nodes.

Abstract: A method, performed by a first mobility management entity (MME) device in a network, includes receiving, from a second MME device, standby database information associated with user equipment (UE) registered with the second MME device; detecting that the second MME device has failed or lost connectivity; designating that the UEs registered with the second MME device will be registered with the first MME device, in response to detecting that the second MME device has failed or lost connectivity; detecting a request to activate a particular UE registered with the second MME device; and paging the particular UE to register with the first MME device, using the standby database information and in response to detecting the request to activate the particular UE.

Abstract: The disclosure relates to communication technologies and discloses a method and a system for emergency switching. In accordance with the embodiments of the present invention, by configuring a mapped IP address in a previous level network device which can map addresses, the embodiments of the present invention enable corresponding devices in respective service processing subsystems to be backup devices to each other. When a problem occurs in a device of a service processing subsystem, as long as an IP address, which is mapped to the network device, in the previous level network device is mapped to a corresponding device of another service processing subsystem, and the corresponding device in another service processing subsystem acts as a backup device to process a service of the device in which the problem occurs, thus enabling simple and fast starting of a backup device when a problem occurs in the original device.

Abstract: Provided are systems and methods for accessing a storage device from a node when a local connection failure occurs between the node and the storage device. A failure is determined to have occurred at a first node access path between a first node and a storage device that prevents an application at the first node from accessing the storage device from the first node access path. An access request is sent from the first node to a second node. The second node has a second node access path to the storage device. A determination is made that the second node can communicate with the storage device. The storage device is accessed by an application at the first node via the second node access path.

Abstract: A multi-cluster system includes a plurality of computers; and a plurality of system storage apparatuses each of which is coupled to the plurality of computers; wherein at least one of the plurality of system storage apparatuses includes a first circuit that detects a connection information that includes connection-failure information indicating a connection failure in a connection with at least one of the plurality of computers, and a second circuit that reports the connection information detected by the first circuit to the plurality of computers; and each of the plurality of computers includes a third circuit that receives the connection information from each of the plurality of system storage apparatuses, and a fourth circuit that executes processing to disconnect a system storage apparatus, based on the connection information received by the third circuit.

Abstract: A management system, which manages a host computer and a storage system, holds cluster information, specifies an active-state host computer and an inactive-state host computer based on a backup instruction specifying a virtual host identifier, determines the need for executing a replication for disaster recovery use, and when necessary, executes this replication for disaster recovery use in combination with a replication for backup use.

Abstract: Methods and apparatus relating to home agent data and memory management are described. In one embodiment, a scrubber logic corrects an error at a location in a memory corresponding to a target address by writing back the corrected version of data to the target location. In an embodiment, a map out logic maps out an index or way of a directory cache in response to a number of errors, corresponding to the directory cache, exceeding a threshold value. Other embodiments are also disclosed.

Abstract: In one embodiment, a technique for routing traffic in networks represented by logical topologies, such as Multi Chassis Port Channel (MCPC) or Multi Chassis Ether Channel (MCEC) topologies, is provided. By modifying a port priority vector (PPV) to include an additional “Switch ID” field that identifies a designated bridge ID or a local switch ID, depending on whether the corresponding port is used as an MCT, a routing protocol designed to avoid loops in routing paths, such as STP, may avoid blocking MCT ports.

Abstract: Exemplary embodiments relate to a method for swapping PPRC secondary device definitions from a subchannel set other than zero to subchannel set zero. The method comprises identifying a PPRC primary and secondary device, wherein the PPRC primary device definitions reside at subchannel set zero and the PPRC secondary device definitions reside at a subchannel set other than subchannel set zero, initiating a PPRC device definition substitution operation to substitute the PPRC primary device definitions within the subchannel set where the PPRC secondary device definition resides and substitute the PPRC secondary device definitions within the subchannel set where the PPRC primary device definitions resides, storing the control block information for each PPRC device whose device definitions are to be substituted, and storing the PPRC primary device definitions within the subchannel set where the PPRC secondary device definition resides and storing the PPRC secondary device definitions within subchannel set zero.

Abstract: A method for providing fault-tolerant network communications between a plurality of nodes for an application, including providing a plurality of initial communications pathways over a plurality of networks coupled between the plurality of nodes, receiving a data packet on a sending node from the application, the sending node being one of the plurality of nodes, the data packet being addressed by the application to an address on one of the plurality of nodes, and selecting a first selected pathway for the data packet from among the plurality of initial communications pathways where the first selected pathway is a preferred pathway.