Abstract: Systems and methods are disclosed for scaling and accelerating decentralized execution of transactions. In one implementation, transactions are divided into transaction segments. A first transaction segment is executed and relevant initialization state for the first transaction segment is determined. A second transaction segment is executed based on the execution of the first transaction segment. Based on the execution of the second transaction segment and an output of the execution of the first transaction segment, a In second initialization state is determined. The first transaction segment and the first initialization state are provided to a first execution shard. The second transaction segment and the second initialization state are provided to a second execution shard. A validation of result(s) of the transactions is received. The validation is computed based an output of the execution of the first transaction segment and an output of the execution of the second transaction segment.

Abstract: A method for designing a Network Function Virtualization (NFV) architecture includes accepting a definition of multiple Virtual Network Functions (VNFs), and of one or more packet types having respective occurrence probabilities, wherein each packet type is associated with a respective subset of the VNFs that are to be applied to packets of that packet type. Information on multiple available physical computers, each capable of running only a partial subset of the multiple VNFs, is further accepted. The VNFs are allocated to the physical computers by applying an optimality criterion to definition and the information.

Abstract: Packets received by a network device are assigned to packet flows defined by shared packet characteristics. The packet flows belong to a first category containing packet flows that are to be granted access to the shared resource periodically and with a fixed time interval between subsequent grants of access to the shared resource, or a second category containing packet flows not included in the first category. A periodic schedule that allocates the shared resource to the packet flows in the first category is determined. The periodic schedule includes one or more empty time slots during which the shared resource is not granted to any packet flow in the first category. Access to the shared resource is granted to packet flows in the first category according to the periodic schedule, and is granted packet flows in the second category during the empty time slots in the periodic schedule.

Abstract: A TCAM database is partitioned into a plurality of sections. A set of rules to be stored in the TCAM database is analyzed to identify respective subsets of rules that match respective possible bit value combinations corresponding to a subset of bits in the rules, and to identify, in the subsets of rules, two or more subsets that share one or more rules. Then, it is determined whether two or more subsets that include shared rules can be written to a same section in the TCAM database, without exceeding a maximum number of rules that can be written to the same section, when one or more duplicates of one the the shared rules is omitted from the TCAM database. When it is determined that two or more subsets can be written to the same section, the two or more subsets are written to the same section in the TCAM database.

Abstract: Systems and methods are provided for a multi-core processor for processing different types of data units. A system includes a classifier configured to classify incoming data units into different type data units. A plurality of processing cores are selectably configurable into plural processing pipelines, respective processing pipelines including connected processing cores, ones of the processing cores being selectably programmed to execute a respective processing operation on a received incoming data unit, different ones of the processing pipelines defined by a selectable number of processing cores. A distributor is configured to distribute the different types of data units to one of the pipelines among the plural pipelines at least as a function of the classified type of the data units and the programmed processing operations of processing cores in the pipelines.

Abstract: A method for designing a Network Function Virtualization (NFV) architecture includes accepting a definition of multiple Virtual Network Functions (VNFs), and of one or more packet types having respective occurrence probabilities, wherein each packet type is associated with a respective subset of the VNFs that are to be applied to packets of that packet type. Information on multiple available physical computers, each capable of running only a partial subset of the multiple VNFs, is further accepted. The VNFs are allocated to the physical computers by applying an optimality criterion to definition and the information.

Abstract: A method for communication includes configuring a multi-level fat-tree network to include at least three levels of switches, including multiple modules arranged externally in a tree topology. Each module contains a respective group of the switches arranged in an internal tree extending over at least two of the levels of the network. A subset of the modules is selected to be active in carrying the communication traffic. The network is operated so as to convey communication traffic among the switches via the active modules, while the modules that are not in the selected subset remain inactive.

Abstract: A method for communication includes configuring a multi-level fat-tree network to include at least three levels of switches, including multiple modules arranged externally in a tree topology. Each module contains a respective group of the switches arranged in an internal tree extending over at least two of the levels of the network. A subset of the modules is selected to be active in carrying the communication traffic. The network is operated so as to convey communication traffic among the switches via the active modules, while the modules that are not in the selected subset remain inactive.

Abstract: A method for communication includes estimating a characteristic of communication traffic to be carried by a fat-tree network. Responsively to the estimated characteristic, a subset of the spine switches in the highest level of the network is selected, according to a predetermined selection order, to be active in carrying the communication traffic. In each of the levels of the spine switches below the highest level, the spine switches to be active are selected based on the selected spine switches in a next-higher level. The network is operated so as to convey the traffic between the leaf switches via the active spine switches, while the spine switches that are not selected remain inactive.

Abstract: A method for communication includes estimating a characteristic of communication traffic to be carried by a fat-tree network. Responsively to the estimated characteristic, a subset of the spine switches in the highest level of the network is selected, according to a predetermined selection order, to be active in carrying the communication traffic. In each of the levels of the spine switches below the highest level, the spine switches to be active are selected based on the selected spine switches in a next-higher level. The network is operated so as to convey the traffic between the leaf switches via the active spine switches, while the spine switches that are not selected remain inactive.