Abstract: A circuit-based digital communications network is provided for a large data center environment that utilizes circuit switching in lieu of packet switching in order to lower the cost of the network and to gain performance efficiencies. A method for transmitting data in such a network comprises sending a setup request for a path for transmitting the data to a destination node and then speculatively sending the data to the destination node before the setup request is completed.

Abstract: A cryptographically-secure component provides access-undeniability and verifiable revocation for clients with respect to downloaded content items from a server. A cryptographically-secure component is implemented in a client. When the client wants to purchase and download a content item from the server, the server requests an encryption key from the client. The client generates an encryption key that is bound to a state of the client that is associated with decrypting the content item. The server encrypts the content item using the encryption key and sends the encrypted content item to the client. Because the encryption key used to encrypt the content item is bound to the state associated with the client decrypting the content item, if the client desires to view the content item the client may first advance its state to the bound state to retrieve the decryption key.

Abstract: A cryptographically-secure component provides access-undeniability and verifiable revocation for clients with respect to downloaded content items from a server. A cryptographically-secure component is implemented in a client. When the client wants to purchase and download a content item from the server, the server requests an encryption key from the client. The client generates an encryption key that is bound to a state of the client that is associated with decrypting the content item. The server encrypts the content item using the encryption key and sends the encrypted content item to the client. Because the encryption key used to encrypt the content item is bound to the state associated with the client decrypting the content item, if the client desires to view the content item the client may first advance its state to the bound state to retrieve the decryption key.

Abstract: A system is disclosed for creating and implementing an access control policy framework in a weakly coherent distributed collection. A collection manager may sign certificates forming equivalence classes of replicas that share a specific authority. The collection manager and/or certain privileged replicas may issue certificates that delegate authority for control of item policy and replica policy. Further certificates may be signed that create one or more items, set policy for these one or more items, and define a set of operations authorized on the one or more items. The certificates issued according to the present system for creating and implementing a control policy framework cannot be modified or simply overridden. Once a policy certificate is issued, it may only be revoked by the collection manager or by a replica having revocation authority.

Abstract: A circuit-based digital communications network is provided for a large data center environment that utilizes circuit switching in lieu of packet switching in order to lower the cost of the network and to gain performance efficiencies. A method for transmitting data in such a network comprises sending a setup request for a path for transmitting the data to a destination node and then speculatively sending the data to the destination node before the setup request is completed.

Abstract: A distributed system includes full and partial replicas of a set of data items that may be inserted, modified, or deleted by any replica. Replicas may occasionally synchronize with other arbitrarily chosen replicas to learn about updates. A replica's knowledge includes one or more knowledge fragments, where each fragment indicates a set of items. A type of knowledge fragment, called a star knowledge fragment, contains versions associated with all items in the system. Star knowledge fragments are compact because the set of items stored at a replica need not be explicitly listed. Once all replicas know of all updates in the system, partial and full replicas will have the same compact star knowledge fragment.

Abstract: A distributed system synchronizes replica devices with respect to items that may be inserted, modified, or deleted by any of the replica devices. Replicas may synchronize with other replicas to learn about updates to items. Each replica device may include a high-fidelity replication platform and/or a low-fidelity replication platform. The low-fidelity replication platforms may synchronize low-fidelity versions of items among the replica devices, and the high-fidelity replication platforms may synchronize high-fidelity versions of items among the replica devices. Each replica device may include a fidelity manager that copies high-fidelity versions of items from the high-fidelity replication platform, generates low-fidelity versions of the items from the high-fidelity versions of the items, and adds the low-fidelity versions of the items to the low-fidelity replication platforms.

Abstract: A method and apparatus are provided for detecting data races that overcome the limitations of the prior art. In some embodiments, this is accomplished by detecting a first access to an object, determining whether the first access is associated with a suspicious pattern, automatically refining a pattern detection granularity from the object to a memory location within the object if a determination is made that the first access is associated with the suspicious pattern, and reporting the data race if a second access associated with the suspicious pattern occurs at the memory location after the pattern detection granularity is refined.

Abstract: A system is disclosed for synchronizing partially-replicated collections using the concept of item-set knowledge and move-out notifications. Move-out notifications are provided to allow removal of changed items from replicas when the change in the item takes that item outside of the interest set of the replicas.

Abstract: Criteria are defined for determining when additional stack dumps would be saved and presented (e.g., to the developer). This provides more information for the developer to use for debugging, and yet does not overburden the developer with irrelevant information. Suspicious activity worthy of an additional stack dumps may be defined as a thread accessing an already suspicious location that is different than the original thread, a write access being made and no write access having yet been made to the target location, or if the lockset being held by the thread accessing the location is omitted from the intersection of all locksets involved in stack dumps for the location.

Abstract: A distributed system includes full and partial replicas of a set of data items that may be inserted, modified, or deleted by any replica. Replicas may occasionally synchronize with other arbitrarily chosen replicas to learn about updates. A replica's knowledge includes of one or more knowledge fragments, where each fragment indicates a set of items. A type of knowledge fragment, called a star knowledge fragment, contains versions associated with all items in the system. Star knowledge fragments are compact because the set of items stored at a replica need not be explicitly listed. Once all replicas know of all updates in the system, partial and full replicas will have the same compact star knowledge fragment.

Abstract: A hierarchical system is disclosed for synchronizing partially-replicated collections that provides guaranteed paths of information to all replicas in a collection. Each partial replica is assigned a parent replica to act as a proxy on its behalf, and with which the replica synchronizes regularly. Each parent replica is responsible for one or more child replicas. Parent replicas have filters which are either the same as, or more inclusive than, their child replicas, and parent replicas thus store and synchronize all objects that are of interest to their one or more child replicas.

Abstract: A system is disclosed for synchronizing partially-replicated collections using the concept of item-set knowledge and move-out notifications. Move-out notifications are provided to allow removal of changed items from replicas when the change in the item takes that item outside of the interest set of the replicas.

Abstract: A system is disclosed for creating and implementing an access control policy framework in a weakly coherent distributed collection. A collection manager may sign certificates forming equivalence classes of replicas that share a specific authority. The collection manager and/or certain privileged replicas may issue certificates that delegate authority for control of item policy and replica policy. Further certificates may be signed that create one or more items, set policy for these one or more items, and define a set of operations authorized on the one or more items. The certificates issued according to the present system for creating and implementing a control policy framework cannot be modified or simply overridden. Once a policy certificate is issued, it may only be revoked by the collection manager or by a replica having revocation authority.

Abstract: A system is disclosed for synchronizing partially-replicated collections while keeping synchronization overhead low by using the concept of item-set knowledge. Item-set knowledge uses knowledge fragments, which associate knowledge vectors with item-sets. An item-set consists of an explicitly represented list of items. In a partial replica, this item set may be the items known to a replica within which a filter is applied limiting the items known to some subset of the overall items in the collection.

Abstract: A system is disclosed for recovery from a compromise of a replica in a weakly-consistent distributed collection. The system employs a collection manager for revoking a compromised replica, and one or more archival replicas for storing time-stamped versions. Upon a compromise, versions tainted by the compromised replica may be expunged from the collection. Thereafter, versions determined to be unaffected by the compromise may be returned to the collection using the time-stamped versions stored in the one or more archival replicas.

Abstract: A method and apparatus are provided for detecting data races that overcome the limitations of the prior art. In some embodiments, this is accomplished by detecting a first access to an object, determining whether the first access is associated with a suspicious pattern, automatically refining a pattern detection granularity from the object to a memory location within the object if a determination is made that the first access is associated with the suspicious pattern, and reporting the data race if a second access associated with the suspicious pattern occurs at the memory location after the pattern detection granularity is refined.

Abstract: Managing end stations locations in a bridged network includes providing a smart bridge, and performing communications operations using a network flood active port connections table, a network flood reaching table, a best path forwarding table and a best path sourcing table. An end station locations revision manages a host location (HLOC) table. End stations locations revision is invoked upon detection of changes in any end station location. The end stations locations revision exchanges location revision grandparent (LRG), a location revision child (LRC) table and a location revision wavefront distribution protocol. The locations revision passes the location revision request from the smart bridge over a topology of the bridged network. A location revision wavefront starts at the root and, spreads a location revision request for revising the HLOC of the other smart bridges. The wavefront spreads by wavefront protocol.

Abstract: A bridged network and a method for performing bridged network topology acquisition. The bridged network includes network segments and smart bridges having one or more ports. The network segments are interconnected by the smart bridges via the ports. Each of the smart bridges being configured to perform steps to accomplish the network topology acquisition. The method also performs steps to accomplish the network topology acquisition. The steps include an inventory construction that compiles a list of the ports that are connected to each of the network segments. Among the ports that are connected to any of the network segments the inventory construction respecting each port operates cooperatively so as to cause the list respecting each of the ports to converge rapidly to the state in which all the lists are similar and yielding a uniform segment inventory.

Abstract: In a distributed computer system having components which may fail intermittently, a fault monitor, coupled to one component, detects whether that component is working. Whenever the fault monitor detects that the component has failed, i.e., changed from working to broken status, the status filter transmits without delay a "broken" signal. Whenever the fault monitor detects that the component has changed from broken to working status, however, the status filter transmits a "working" signal only after a recovery time interval corresponding to a computed skepticism level, and only if the component does not fail during that recovery time interval. The status filter increases the computed skepticism level, and redetermines the recovery time interval in accordance therewith, each time that the component fails after the status filter has transmitted a "working" signal.