Abstract: A passive optical network (PON) has an optical line termination (OLT) that terminates an optical fiber servicing a plurality of optical network units (ONUs). Each ONU has one or more traffic containers (TCONTs) addressable by the OLT. The PON dynamic bandwidth allocation (DBA) implements a scheduling hierarchy, including several scheduling layers, such that disjoint sets of TCONTs can be grouped together, then disjoint sets of groups can be grouped, and so on. In such hierarchy, the residential traffic can be grouped separately from the business traffic. Further, within either the residential or business group, traffic may be grouped to define scheduling layers (“sub-groups”) within the residential or business group. Scheduling in one group or sub-group is performed independently of the scheduling in other groups or sub-groups, subject to the available bandwidth for each group.

Abstract: The present disclosure generally pertains to systems and methods for passing timing information over packet networks. In one exemplary embodiment of the present disclosure, a transmitter inserts a timestamp into various packets being transmitted. A receiver determines the packet delay for each of these packets. The receiver monitors the packet delays of received packets over time to estimate a minimum packet delay for the network. Thereafter, the receiver uses the timestamps of received packets in order to update a local receiver clock signal so that the local receiver clock signal is synchronous to the clock signal used by the transmitter to generate the timestamps. However, the receiver filters the received timestamps such that the effects of packet delay variation to the synchronization of the local clock signals are reduced.

Abstract: A passive optical network (PON) has an optical line termination (OLT) that terminates an optical fiber servicing a plurality of optical network units (ONUs). Each ONU has one or more traffic containers (TCONTs) addressable by the OLT. The PON dynamic bandwidth allocation (DBA) implements a scheduling hierarchy, including several scheduling layers, such that disjoint sets of TCONTs can be grouped together, then disjoint sets of groups can be grouped, and so on. In such hierarchy, the residential traffic can be grouped separately from the business traffic. Further, within either the residential or business group, traffic may be grouped to define scheduling layers (“sub-groups”) within the residential or business group. Scheduling in one group or sub-group is performed independently of the scheduling in other groups or sub-groups, subject to the available bandwidth for each group.

Abstract: A communication system has a trunk extending from a network facility, such as a central office, with a plurality of distribution points positioned along the trunk. Each leg of the trunk defines a shared channel that permits peak data rates much greater than what would be achievable without channel sharing. As an example, the connections of each respective trunk leg may be bonded. Further, the same modulation format and crosstalk vectoring are used for each leg of the trunk. The crosstalk vectoring cancels both far-end crosstalk (FEXT) that couples between connections of a given trunk leg and crossover crosstalk that couples between one trunk leg and another. In addition, logic determines an amount of excess capacity available for each leg of the trunk and controls error correction based on the determined excess capacity.

Abstract: An optical communication system comprises a network interface device (NID) having a media converter coupled to an optical fiber of a passive optical network (PON). The media converter converts optical signals from the PON into electrical signals for communication across at least one non-optical channel, such as a conductive or wireless connection, to customer premises equipment (CPE), such as a residential gateway or other customer premises (CP) device. Rather than implementing an optical media access control (optical MAC) layer in the NID, an optical MAC layer for handling PON protocols and management is implemented by the CPE, thereby effectively extending the customer end of the PON across at least one non-optical connection to the CPE. By implementing the optical MAC layer at the CPE, the complexity of the NID is reduced thereby lowering the cost of the NID.

Abstract: A communication system comprises a plurality of line cards having transceivers coupled to a plurality of subscriber lines. Each line card has at least one transceiver within the same vectoring group, and each line card also has vector logic capable of cancelling crosstalk induced by a tone communicated by any member of the vector group. Further, the line cards are coupled to one another via a data connection across which a vectoring stream carrying vectoring information from one line card to the next. The bandwidth of the vectoring stream is reduced by dynamically adjusting time slots of the vectoring stream based on bit loading for the communicated tones.

Abstract: A communication system comprises a plurality of line cards having transceivers coupled to a plurality of subscriber lines. Each line card has at least one active transceiver within the same vectoring group, and each line card also has vector logic capable of cancelling crosstalk induced by an active transceiver that is a member of the vectoring group. Further, the line cards are coupled to one another via a ring connection across which vectoring information is passed from one line card to the next. In the event of a failure of one of the line cards, the failed card is bypassed by the vectoring stream so that the operational line cards can continue crosstalk vectoring operations despite such failure.

Abstract: A communication system comprises a plurality of line cards having transceivers coupled to a plurality of subscriber lines. Each line card has at least one active transceiver within the same vectoring group, and each line card also has vector logic capable of cancelling crosstalk induced by an active transceiver that is a member of the vectoring group. Further, the line cards are coupled to one another via a ring connection across which vectoring information is passed from one line card to the next. In the event of a failure of one of the line cards, the failed card is bypassed by the vectoring stream so that the operational line cards can continue crosstalk vectoring operations despite such failure.

Abstract: A controller at a distribution point (DP) of a communication system is coupled to a plurality of customer premises (CP) transceivers via drop connections in a point-to-multipoint architecture. Each drop connection is coupled to at least one switch that operates under the control of the controller for selectively isolating the drop connection from the controller, as well as the CP transceivers of other drop connections. In this regard, by controlling the states of the switches, the DP controller can control to which of the CP transceivers it is communicatively connected, and during operation the DP controller controls the switches such that it is communicatively connected only to the CP transceivers for which communication is desired or needed during a particular time interval.

Abstract: A discrete multi-tone (DMT) transceiver communicates tones across a subscriber line. Vectoring is employed in an effort to reduce the effects of crosstalk. However, for some tones, such as tones significantly affected by radio frequency interference (RFI) or other forms of alien noise, vectoring may actually introduce distortions such that the vectoring degrades rather than improves overall signal quality. Control logic of the DMT transceiver is configured to sense when tones are affected by significant levels of alien noise and to exclude such tones from vectoring, thereby improving signal quality for such tones. The control logic also may lower the constellation densities of such tones in order to accommodate the vectoring exclusions applied to such tones.

Abstract: A communication system comprises a plurality of line cards having transceivers coupled to a plurality of subscriber lines. Each line card has at least one active transceiver within the same vectoring group, and each line card also has vector logic capable of cancelling crosstalk induced by an active transceiver that is a member of the vectoring group. In the event of a vectoring fault that prevents a line card from receiving vectoring information from at least on other line card, the vector logic is configured to disable vectoring for the interferers affected by the error in order to prevent vectoring operations based on obsolete vectoring coefficients from adversely affecting the quality of the communicated signals. The transceivers communicating signals affected by the suspended vectoring operations are also configured to adjust their constellation density profiles, thereby reducing their data rates, to accommodate the increased noise level resulting from the loss of vectoring.

Abstract: A system and method provides for failover of guest operating systems in a virtual machine environment. During initialization of a computer executing a virtual machine operating system, a first guest operating system allocates a first memory region within a first domain and notifies a second guest operating system operating in a second domain of the allocated first memory region. Similarly, the second guest operating system allocates a second region of memory within the second domain and notifies the first operating system of the allocated second memory region. In the event of a software failure affecting one of the guest operating systems, the surviving guest operating system assumes the identity of the failed operating system and utilizes data stored within the shared memory region to replay to storage devices to render them consistent.

Abstract: A system and method reclaims unused storage space from a data container, such as a logical unit number (LUN) of a storage system. In particular, a novel technique is provided that allows a storage system to reclaim storage space not used by a client file system for which the storage system maintains storage, without requiring assistance from the client file system to determine storage usage. In other words, storage system may independently reclaim storage space not used by the client file system, without that file system's intervention.

Abstract: The present invention provides a system and method for fast restart of a guest operating system executing on a virtual machine operating system in a virtual machine environment. During initialization, the guest operating system saves a set of checkpoint information to persistent storage. Upon detection of an error condition during operation, the guest operating system begins a re-initialization procedure in accordance with an illustrative embodiment of the present invention. During the re-initialization procedure, the guest operating system retrieves the checkpoint information and configures itself using the retrieved information. By utilizing the retrieved information, the guest operating system avoids the need to perform lengthy configuration discovery routines, thereby shortening the re-initialization time substantially.

Abstract: A system and method accelerates update of a metadata search database using PCPI differencing. After first populating the search database, a search agent generates a PCPI and utilizes a PCPI differencing technique to quickly identify changes between inode files of first and second PCPIs. The differences are noted as modified metadata and are written to a log file, which is later read by the search agent to update the search database.

Abstract: A system and method for nearly in-band search indexing. A network switch (or other intermediate network device) is configured to provide port mirroring so that data access requests directed to a storage system are forwarded to both the storage system and to a search appliance. The search appliance collects index information from the received data access requests to update a search index. As the search appliance is nearly in-band, i.e., not directly in-line of the data access request path, no increase of latency occurs for processing data access requests by the storage system.

Abstract: A system and method provides for failover of guest operating systems in a virtual machine environment. During initialization of a computer executing a virtual machine operating system, a first guest operating system allocates a first memory region within a first domain and notifies a second guest operating system operating in a second domain of the allocated first memory region. Similarly, the second guest operating system allocates a second region of memory within the second domain and notifies the first operating system of the allocated second memory region. In the event of a software failure affecting one of the guest operating systems, the surviving guest operating system assumes the identity of the failed operating system and utilizes data stored within the shared memory region to replay to storage devices to render them consistent.

Abstract: A system and method provides for failover of guest operating systems in a virtual machine environment. During initialization of a computer executing a virtual machine operating system, a first guest operating system allocates a first memory region within a first domain and notifies a second guest operating system operating in a second domain of the allocated first memory region. Similarly, the second guest operating system allocates a second region of memory within the second domain and notifies the first operating system of the allocated second memory region. In the event of a software failure affecting one of the guest operating systems, the surviving guest operating system assumes the identity of the failed operating system and utilizes data stored within the shared memory region to replay to storage devices to render them consistent.

Abstract: An interworking function (IWF) is coupled to a switch of a packet network and communicates with the network switch based on an Ethernet clock signal or some other type of clock signal. A primary reference clock (PRC) of the network generates a PRC signal, and a timing analyzer determines timing information indicative of timing relationships between the Ethernet clock signal and the PRC signal. The timing analyzer periodically transmits such timing information, and the IWF uses the timing information to generate a PRC signal that is traceable to the network PRC signal.

Abstract: A system and method for nearly in-band search indexing. A network switch (or other intermediate network device) is configured to provide port mirroring so that data access requests directed to a storage system are forwarded to both the storage system and to a search appliance. The search appliance collects index information from the received data access requests to update a search index. As the search appliance is nearly in-band, i.e., not directly in-line of the data access request path, no increase of latency occurs for processing data access requests by the storage system.