Abstract: A technique for performing stream output operations in a parallel processing system is disclosed. A stream synchronization unit is provided that enables the parallel processing unit to track batches of vertices being processed in a graphics processing pipeline. A plurality of stream output units is also provided, where each stream output unit writes vertex attribute data to one or more stream output buffers for a portion of the batches of vertices. A messaging protocol is implemented between the stream synchronization unit and the plurality of stream output units that ensures that each of the stream output units writes vertex attribute data for the particular batch of vertices distributed to that particular stream output unit in the same order in the stream output buffers as the order in which the batch of vertices was received from a device driver by the parallel processing unit.

Abstract: A system, apparatus, method, and computer-readable medium are provided for authorizing a computing node to participate in a group of computing nodes utilizing a shared group password. According to one method described herein, an invitation to join a group is transmitted to a tentative group member node. The invitation is used to establish a connection with a group member node of the group. The tentative group member node generates a hash of a group password and transmits the hash to the group member node. When the group member node receives the hash, the group member node compares the received value to a previously stored hash of the group password. If the previously stored value is identical to the value received from the tentative group member node, then the tentative group member node is authorized as a new member of the group. Otherwise the tentative group member node is not permitted to become a member of the group.

Abstract: Examples of achieving quality of service in a wireless local area network via meeting communication channel performance criteria to include a minimum signal-to-noise plus interference ratio and a proportional fairness limit are disclosed.

Abstract: Systems and methods are described that facilitate the management of contact information, at least some of the contact information related to entities in a serverless, peer-to-peer system. A contact store may store information regarding which other entities of a plurality of other entities are authorized to monitor presence of a user entity. Presence of an entity may generally indicate the willingness and/or ability of the entity to communicate and/or collaborate with other entities, for example. The contact store may also store information regarding which other entities of the plurality of other entities the presence of which should be monitored by the system. A user entity may be able to add contacts to and/or delete contacts from the contact store, for example.

Abstract: A set of collaborative contacts may be selected with an invitation user interface. An invitation to participate in a collaborative activity may be sent to the selected contacts from the invitation user interface or from a collaborative application. The invitation may be received by a collaborative services platform. A user of the collaborative services platform may be presented with the invitation along with an ability to choose one of accept, decline and ignore in response to the invitation. The user may choose to accept, decline or explicitly ignore the invitation, and a response may be sent to the sender of the invitation indicating the user's choice. An application programming interface for a collaborative invitation service of the collaborative services platform may include an invitation element, a send invitation element, a respond to invitation element and a cancel invitation element. The cancel invitation element may be used to cancel issued invitations.

Abstract: A security infrastructure and methods are presented that inhibit the ability of a malicious node from disrupting the normal operations of a peer-to-peer network. The methods of the invention allow both secure and insecure identities to be used by nodes by making them self-verifying. When necessary or opportunistic, ID ownership is validated by piggybacking the validation on existing messages. The probability of connecting initially to a malicious node is reduced by randomly selecting to which node to connect. Further, information from malicious nodes is identified and can be disregarded by maintaining information about prior communications that will require a future response. Denial of service attacks are inhibited by allowing the node to disregard requests when its resource utilization exceeds a predetermined limit. The ability for a malicious node to remove a valid node is reduced by requiring that revocation certificates be signed by the node to be removed.

Abstract: A security infrastructure and methods are presented that inhibit the ability of a malicious node from disrupting the normal operations of a peer-to-peer network. The methods of the invention allow both secure and insecure identities to be used by nodes by making them self-verifying. When necessary or opportunistic, ID ownership is validated by piggybacking the validation on existing messages. The probability of connecting initially to a malicious node is reduced by randomly selecting to which node to connect. Further, information from malicious nodes is identified and can be disregarded by maintaining information about prior communications that will require a future response. Denial of service attacks are inhibited by allowing the node to disregard requests when its resource utilization exceeds a predetermined limit. The ability for a malicious node to remove a valid node is reduced by requiring that revocation certificates be signed by the node to be removed.

Abstract: A security infrastructure and methods are presented that inhibit the ability of a malicious node from disrupting the normal operations of a peer-to-peer network. The methods of the invention allow both secure and insecure identities to be used by nodes by making them self-verifying. When necessary or opportunistic, ID ownership is validated by piggybacking the validation on existing messages. The probability of connecting initially to a malicious node is reduced by randomly selecting to which node to connect. Further, information from malicious nodes is identified and can be disregarded by maintaining information about prior communications that will require a future response. Denial of service attacks are inhibited by allowing the node to disregard requests when its resource utilization exceeds a predetermined limit. The ability for a malicious node to remove a valid node is reduced by requiring that revocation certificates be signed by the node to be removed.

Abstract: Collaboration between collaborative endpoints may be facilitated by a serverless publication service of a collaborative services platform. The serverless publication service may be configured to accept communicative connections from the collaborative endpoints. Users of the collaborative services platform may publish their associated collaborative presences with the serverless publication service. For example, the collaborative presence of a user may include information with respect to valid collaborative endpoints for the user and collaborative capabilities at those endpoints. Subscriptions to published collaborative presences may be placed through the serverless publication service. A collaborative presence subscribe message may specify a subscription to a collaborative presence of a user. Upon receipt of a presence subscribe message, if there is no subscription policy with respect to the sender, the recipient may be queried for a subscription policy with respect to the sender.

Abstract: Systems and methods are described that facilitate the management of contact information, at least some of the contact information related to entities in a serverless, peer-to-peer system. A contact store may store information regarding which other entities of a plurality of other entities are authorized to monitor presence of a user entity. Presence of an entity may generally indicate the willingness and/or ability of the entity to communicate and/or collaborate with other entities, for example. The contact store may also store information regarding which other entities of the plurality of other entities the presence of which should be monitored by the system. A user entity may be able to add contacts to and/or delete contacts from the contact store, for example.

Abstract: Systems and methods are described that facilitate the management of contact information, at least some of the contact information related to entities in a serverless, peer-to-peer system. A contact store may store information regarding which other entities of a plurality of other entities are authorized to monitor presence of a user entity. Presence of an entity may generally indicate the willingness and/or ability of the entity to communicate and/or collaborate with other entities, for example. The contact store may also store information regarding which other entities of the plurality of other entities the presence of which should be monitored by the system. A user entity may be able to add contacts to and/or delete contacts from the contact store, for example.

Abstract: Technologies are described herein for relevance-based expiration of data. An initial expiration time for the data is computed based upon an expiration duration associated with the data. The expiration time for the data is periodically re-computed in order to extend the expiration time. A relevance value for the meeting data is computed and the expiration time is set as a function of the computed relevance value. The relevance value may be computed as a product of a user-settable relevance value for the meeting data and a dynamic relevance point for the meeting data. When the computed expiration time for meeting data has been reached, the meeting data is expired, such as through the deletion of the meeting data.

Abstract: A security infrastructure and methods are presented that inhibit the ability of a malicious node from disrupting the normal operations of a peer-to-peer network. The methods of the invention allow both secure and insecure identities to be used by nodes by making them self-verifying. When necessary or opportunistic, ID ownership is validated by piggybacking the validation on existing messages. The probability of connecting initially to a malicious node is reduced by randomly selecting to which node to connect. Further, information from malicious nodes is identified and can be disregarded by maintaining information about prior communications that will require a future response. Denial of service attacks are inhibited by allowing the node to disregard requests when its resource utilization exceeds a predetermined limit. The ability for a malicious node to remove a valid node is reduced by requiring that revocation certificates be signed by the node to be removed.

Abstract: A multi-tone synchronous collision resolution system permits communication nodes within a MANET to contend simultaneously for a plurality of available channels. The communication nodes contend for access using a synchronous signaling mechanism that utilizes multiple tones in a synchronous manner to resolve contentions. Contentions are arbitrated locally, and a surviving subset of communication nodes is selected. The communication nodes of the surviving subset then transmit data packets simultaneously across the available communication channels.

Abstract: A security infrastructure and methods are presented that inhibit the ability of a malicious node from disrupting the normal operations of a peer-to-peer network. The methods of the invention allow both secure and insecure identities to be used by nodes by making them self-verifying. When necessary or opportunistic, ID ownership is validated by piggybacking the validation on existing messages. The probability of connecting initially to a malicious node is reduced by randomly selecting to which node to connect. Further, information from malicious nodes is identified and can be disregarded by maintaining information about prior communications that will require a future response. Denial of service attacks are inhibited by allowing the node to disregard requests when its resource utilization exceeds a predetermined limit. The ability for a malicious node to remove a valid node is reduced by requiring that revocation certificates be signed by the node to be removed.

Abstract: A security infrastructure and methods are presented that inhibit the ability of a malicious node from disrupting the normal operations of a peer-to-peer network. The methods of the invention allow both secure and insecure identities to be used by nodes by making them self-verifying. When necessary or opportunistic, ID ownership is validated by piggybacking the validation on existing messages. The probability of connecting initially to a malicious node is reduced by randomly selecting to which node to connect. Further, information from malicious nodes is identified and can be disregarded by maintaining information about prior communications that will require a future response. Denial of service attacks are inhibited by allowing the node to disregard requests when its resource utilization exceeds a predetermined limit. The ability for a malicious node to remove a valid node is reduced by requiring that revocation certificates be signed by the node to be removed.

Abstract: Application programming interfaces and methods that provide graph management in a peer-to-peer (P2P) network are provided. More specifically, new and improved P2P application programming interfaces (APIs) and methods for the creation and access of graphs, the retrieval of node and graph information, the addition, modification, deletion and management of records (data), the importation and exportation of graph data, the direct communication between graph nodes, the addition of a security provider to a graph, the setting and retrieval of presence information, the registering for event notifications, and other utility and support functions are presented. Each of these interfaces utilize various parameters that are passed in from an application program that uses these interfaces to manage and utilize graphs. The interfaces return a value that indicates the success or failure of the function. For failures, the interfaces provide an indication as to the problem resulting in the failure.

Abstract: A security infrastructure and methods are presented that inhibit the ability of a malicious node from disrupting the normal operations of a peer-to-peer network. The methods of the invention allow both secure and insecure identities to be used by nodes by making them self-verifying. When necessary or opportunistic, ID ownership is validated by piggybacking the validation on existing messages. The probability of connecting initially to a malicious node is reduced by randomly selecting to which node to connect. Further, information from malicious nodes is identified and can be disregarded by maintaining information about prior communications that will require a future response. Denial of service attacks are inhibited by allowing the node to disregard requests when its resource utilization exceeds a predetermined limit. The ability for a malicious node to remove a valid node is reduced by requiring that revocation certificates be signed by the node to be removed.

Abstract: A system and method for providing security to a graph of interconnected nodes includes a grouping multiplexing layer configured to monitor calls to the system, a graphing dynamic link layer configured to transmit and receive data to and from the graph, and a group security manager coupled to the grouping multiplexing layer and coupled to the graphing dynamic link layer; the group security manager is configured to perform security-related acts via interacting with a group database to propagate security-related information to members of a group within the graph. The group security manager is configured to provide role-based authorization on publication of one or more records and provide membership control for admission to a graph of interconnected nodes. The group security manager provides membership control by providing credentials to potential members of the graph to enable a connection and by providing a governed system for renewal and revocation of members.

Abstract: An authentication mechanism uses a trusted people store that can be populated on an individual basis by users of computing devices, and can comprise certificates of entities that the user wishes to allow to act as certification authorities. Consequently, peer-to-peer connections can be made even if neither device presents a certificate or certificate chain signed by a third-party certificate authority, so long as each device present a certificate or certificate chain signed by a device present in the trusted people store. Once authenticated, a remote user can access trusted resources on a host device by having local processes mimic the user and create an appropriate token by changing the user's password or password type to a hash of the user's certificate and then logging the user on. The token can be referenced in a standard manner to determine whether the remote user is authorized to access the trusted resource.