Other

Veritas Operating Corporation

Symantec Operating Corporation

Paul L Borrill

Gregory L Slaughter

John A Muth

Andrew K Krumel

Girish Jorapurkar

Paul Czarnik

Computational nodes are connected to each other via multiple point to point communication ports to form a mesh network. When a node receives such a routing information, it will be a set of {Route ID, Forwarding Ports} pair, where the ports can be multiple targets including the node itself. When all the nodes within the routing path receive such information, the mesh system can handle the packet transfer for the given Route ID. Each node looks up the table with Route ID and determines where to forward the packet.

Classical implementation of entanglement

Quantum mechanics provides several features useful for datacenter networking. The no cloning theorem, which states that it is impossible to mate a duplicate of an arbitrary, unknown quantum state, can be used to detect eavesdroppers. Entanglement allows two parties to have common knowledge of a shared state. These properties are being used today for quantum key exchange and quantum computing, but they are currently too expensive for general use. Fortunately, we can use classical mechanisms to get a close enough approximation of these quantum properties to solve some important problems in distributed computing. Nothing we describe here is quantum mechanical. Rather, we show that it is possible to use classical mechanisms to emulate some properties of quantum mechanics, which enable us to address interesting problems in distributed computing. The engineering insight, is that we can get closer to achieving these properties than might be expected through conventional thinking. The key to obtaining the properties we desire is to make the inherently asynchronous system temporarily locally synchronous tor the operations we need. In this patent, we describe how to classically emulate the parts of the no cloning theorem and entanglement that we need for datacenter networking. We then demonstrate how those approximations to quantum behavior can be used to solve important problems in distributed computing, such as 'exactly-once' semantics in an environment where failures occur and can be healed without applications needing to know.

Classical implementation of quantum entanglement in datacenter network design

An entangled links mechanism to establish and maintain bipartite temporal intimacy between pairs of computers using an idempotent, reversible token method, which presents no observable external 'change' until communication of information needs to occur between the computers and which maintains the potential for “bounded (or unbounded) reversibility” in case the intended information dispatched by a source computational entity is not captured or properly accepted by a destination computational entity. The mechanism enables distributed computers in a network to remain continuously aware of each other's presence; to communicate on a logically nearest neighbor basis in a secure and reliable manner in which packets passed over these links do not conflict with normal traffic or cause the available resources of the link to exceeded; and that atomicity, isolation, and “reversible durability” may be maintained for transactions when perturbations occur.

Entangled links, transactions and trees for distributed computing systems

A computer-implemented system with a processor provides a reversible transfer of an atomic token from one side of an imperfect link to the other, such that if the protocol (or process) on either side fails at a critical moment, the atomic token will be found on both sides to be verifiably incomplete, unless the protocol has completed successfully past its 'irreversible threshold' on both sides.

Generalized reversibility framework for common knowledge in scale-out database systems

Quantum mechanics provides several features useful for datacenter networking. The no cloning theorem, which states that it is impossible to mate a duplicate of an arbitrary, unknown quantum state, can be used to detect eavesdroppers. Entanglement allows two parties to have common knowledge of a shared state. These properties are being used today for quantum key exchange and quantum computing, but they are currently too expensive for general use. Fortunately, we can use classical mechanisms to get a close enough approximation of these quantum properties to solve some important problems in distributed computing. Nothing we describe here is quantum mechanical. Rather, we show that it is possible to use classical mechanisms to emulate some properties of quantum mechanics, which enable us to address interesting problems in distributed computing. The engineering insight, is that we can get closer to achieving these properties than might be expected through conventional thinking. The key to obtaining the properties we desire is to make the inherently asynchronous system temporarily locally synchronous for the operations we need. In this patent, we describe how to classically emulate the parts of the no cloning theorem and entanglement that we need for datacenter networking. We then demonstrate how those approximations to quantum behavior can be used to solve important problems in distributed computing, such as 'exactly-once' semantics in an environment where failures occur and can be healed without applications needing to know.

An entangled links mechanism to establish and maintain bipartite temporal intimacy between pairs of computers, using an idempotent, reversible token method which presents no observable external 'change' until a communication of information needs to occur between the computers, and which maintains the potential for 'bounded (or unbounded) reversibility' in case the intended information dispatched by a source computational entity is not captured or properly accepted by a destination computational entity. The mechanism enables distributed computers in a network to remain continuously aware of each other's presence; to communicate on a logically nearest neighbor basis in a secure and reliable manner in which packets passed over these links do not conflict with normal traffic or cause the available resources of the link to be exceeded; and that atomicity, consistency, isolation, and &#x201c;reversible durability&#x201d; may be maintained for transactions when perturbations occur.

Computational nodes are connected to each other via multiple point to point communication ports to form a mesh network. When a node receives such routing information, it will be a set of {Route ID, Forwarding Ports} pair, where the ports can be multiple targets including the node itself. When all the nodes within the routing path receive such information, the mesh system can handle the packet transfer for the given Route ID. Each node looks up the table with Route ID and determines where to forward the packet.

Systems, methods, and apparatus, including computer program products, implementing protocols for building and maintaining multiple trees in a network and using the trees to route messages through the network. A node in the network initiates the building of a tree having itself as the root, and nodes in the network generally maintain representations of multiple trees, each tree having a distinct node as its root. A node can use the trees to route messages to the root nodes by sending the messages through root-ward ports for the respective trees.

Self-healing communication trees

A method and system for creating a peer-to-peer overlay network. The method includes discovering a set of node characteristics for each of a subset of existing nodes of a network. The method also includes scaling each of the discovered set of node characteristics to a value which is within a predetermined range of values. The method also includes generating a set of weighted values for each of the subset of nodes by applying a weighting function to each of the scaled values. In addition, the method includes generating a node rank value for each of the subset of nodes by summing the weighted values of each respective set of weighted values. Further, the method includes determining whether to connect to any of the subset of nodes based upon the node rank value.

Growing community of inventors

Palo Alto, CA, United States of America

Paul L Borrill