Abstract:
A server system receives messages from client computing devices. Each of the messages corresponds to a transaction. The server system assigns each respective transaction to a respective fresh virtual machine. Furthermore, the server system performs, as part of a respective virtual machine processing a respective transaction, a modification associated with the respective transaction to a shared database. The shared database is persisted independently of the plurality of virtual machines. In response to determining that processing of the respective transaction is complete, the server system discards the respective virtual machine. In response to determining that the respective transaction is associated with a cyber-attack, the server system uses checkpoint data associated with the respective transaction to roll back the modifications associated with the respective transaction to the shared database.
Abstract:
A server system receives messages from client computing devices. Each of the messages corresponds to a transaction. The server system assigns each respective transaction to a respective fresh virtual machine. Furthermore, the server system performs, as part of a respective virtual machine processing a respective transaction, a modification associated with the respective transaction to a shared database. The shared database is persisted independently of the plurality of virtual machines. In response to determining that processing of the respective transaction is complete, the server system discards the respective virtual machine. In response to determining that the respective transaction is associated with a cyber-attack, the server system uses checkpoint data associated with the respective transaction to roll back the modifications associated with the respective transaction to the shared database.
Abstract:
A survivable network is described in which one or more network device includes enhanced functionality to fight through cyber attacks. A Fight-Through Node (FTN) is described, which may be a combined hardware/software system that enhances existing networks with survivability properties. A network node comprises a hardware-based processing system having a set of one or more processing units, a hypervisor executing on each one of the processing units, and a plurality of virtual machines executing on each of the hypervisor. The network node includes an application-level dispatcher to receive a plurality of transaction requests from a plurality of network communication session with a plurality of clients and distribute a copy of each of the transaction requests to the plurality of virtual machines executing on the network node over a plurality of time steps to form a processing pipeline of the virtual machines.
Abstract:
A survivable network is described in which one or more network device includes enhanced functionality to fight through cyber attacks. A Fight-Through Node (FTN) is described, which may be a combined hardware/software system that enhances existing networks with survivability properties. A network node comprises a hardware-based processing system having a set of one or more processing units, a hypervisor executing on each one of the processing units, and a plurality of virtual machines executing on each of the hypervisor. The network node includes an application-level dispatcher to receive a plurality of transaction requests from a plurality of network communication session with a plurality of clients and distribute a copy of each of the transaction requests to the plurality of virtual machines executing on the network node over a plurality of time steps to form a processing pipeline of the virtual machines.
Abstract:
A server system receives messages from client computing devices. Each of the messages corresponds to a transaction. The server system assigns each respective transaction to a respective fresh virtual machine. Furthermore, the server system performs, as part of a respective virtual machine processing a respective transaction, a modification associated with the respective transaction to a shared database. The shared database is persisted independently of the plurality of virtual machines. In response to determining that processing of the respective transaction is complete, the server system discards the respective virtual machine. In response to a trigger, such as determining that the respective transaction is associated with a cyber-attack, the server system uses checkpoint data associated with the respective transaction to roll back the modifications associated with the respective transaction to the shared database.
Abstract:
A survivable network is described in which one or more network device includes enhanced functionality to fight through cyber attacks. A Fight-Through Node (FTN) is described, which may be a combined hardware/software system that enhances existing networks with survivability properties. A network node comprises a hardware-based processing system having a set of one or more processing units, a hypervisor executing on each one of the processing units, and a plurality of virtual machines executing on each of the hypervisor. The network node includes an application-level dispatcher to receive a plurality of transaction requests from a plurality of network communication session with a plurality of clients and distribute a copy of each of the transaction requests to the plurality of virtual machines executing on the network node over a plurality of time steps to form a processing pipeline of the virtual machines.
Abstract:
A survivable network is described in which one or more network device includes enhanced functionality to fight through cyber attacks. A Fight-Through Node (FTN) is described, which may be a combined hardware/software system that enhances existing networks with survivability properties. A network node comprises a hardware-based processing system having a set of one or more processing units, a hypervisor executing on each one of the processing units, and a plurality of virtual machines executing on each of the hypervisor. The network node includes an application-level dispatcher to receive a plurality of transaction requests from a plurality of network communication session with a plurality of clients and distribute a copy of each of the transaction requests to the plurality of virtual machines executing on the network node over a plurality of time steps to form a processing pipeline of the virtual machines.
Abstract:
This disclosure describes techniques for dynamically assembling and utilizing a pedigree of a resource. A pedigree of a resource is a set of statements that describe a provenance of the resource. As described herein, a document may include local pedigree fragments and optionally one or more pointers to remote pedigree fragments not locally stored in the document. A pedigree fragment, generally, is a data structure that specifies a direct relationship between a first resource, e.g., a primary resource, and a second resource from which an asserted fact of the first resource is derived. Because a pedigree fragment specifies such direct relationships, a set of pedigree fragments may be used to assemble the complete pedigree of resource.
Abstract:
A network node includes enhanced functionality to fight through cyber-attacks. A plurality of virtual machines run at the network node. The network node receives a plurality of transaction requests and distributes a copy of each of the transaction requests to the plurality of virtual machines over a plurality of time steps. Based on the first virtual machine having executed (n) transaction requests in the plurality of transaction requests, the node detects whether any of the virtual machines has been compromised. In response to detecting the plurality of virtual machines includes a compromised virtual machine, the network node isolates the compromised virtual machine. Furthermore, after isolating the compromised virtual machine, the network node may receive a subsequent transaction request and dispatch the subsequent transaction request to the compromised virtual machine. The compromised virtual machine may execute the subsequent transaction request.
Abstract:
A network node includes enhanced functionality to fight through cyber-attacks. A plurality of virtual machines run at the network node. The network node receives a plurality of transaction requests and distributes a copy of each of the transaction requests to the plurality of virtual machines over a plurality of time steps. Based on the first virtual machine having executed (n) transaction requests in the plurality of transaction requests, the node detects whether any of the virtual machines has been compromised. In response to detecting the plurality of virtual machines includes a compromised virtual machine, the network node isolates the compromised virtual machine. Furthermore, after isolating the compromised virtual machine, the network node may receive a subsequent transaction request and dispatch the subsequent transaction request to the compromised virtual machine. The compromised virtual machine may execute the subsequent transaction request.