Systems, computer-readable media and methods for aggregating data items from an out-of-order data stream over a sliding window efficiently. The method delays the value aggregation for certain time windows and computes partial aggregations that can be reused for the multiple time windows. Aggregations may have any value type such as Boolean, integer, strings, floating point, vector and map.

In one aspect, a computerized method includes the step of providing process monitor in a Gateway. The method includes the step of, with the process monitor, launching a Gateway Daemon (GWD). The GWD runs a GWD process that implements a Network Address Translation (NAT) process. The NAT process includes receiving a set of data packets from one or more Edge devices and forwarding the set of data packets to a public Internet. The method includes the step of receiving another set of data packets from the public Internet and forwarding the other set of data packets to the one or more Edge devices. The method includes the step of launching a Network Address Translation daemon (NATD). The method includes the step of detecting that the GWD process is interrupted; moving the NAT process to the NATD.

In one aspect, a computerized method includes the step of providing process monitor in a Gateway. The method includes the step of, with the process monitor, launching a Gateway Daemon (GWD). The GWD runs a GWD process that implements a Network Address Translation (NAT) process. The NAT process includes receiving a set of data packets from one or more Edge devices and forwarding the set of data packets to a public Internet. The method includes the step of receiving another set of data packets from the public Internet and forwarding the other set of data packets to the one or more Edge devices. The method includes the step of launching a Network Address Translation daemon (NATD). The method includes the step of detecting that the GWD process is interrupted; moving the NAT process to the NATD.

A traffic classification method and apparatus, a training method and apparatus, a device and a medium are provided. An implementation is: performing a preprocessing operation on each characteristic of one or more characteristics of an object to be classified; and inputting the one or more characteristics of the object to be classified into a traffic classifier to determine a traffic type of the object to be classified. The preprocessing operation includes at least one of: setting, in response to determining that a characteristic value of the characteristic is invalid data, the characteristic value to a null value; converting, in response to determining that the characteristic is a non-numeric characteristic, the characteristic value of the characteristic to an integer value, and normalizing, in response to determining that the characteristic is a non-port characteristic, the characteristic value of the characteristic.

A traffic classification method and apparatus, a training method and apparatus, a device and a medium are provided. An implementation is: performing a preprocessing operation on each characteristic of one or more characteristics of an object to be classified; and inputting the one or more characteristics of the object to be classified into a traffic classifier to determine a traffic type of the object to be classified. The preprocessing operation includes at least one of: setting, in response to determining that a characteristic value of the characteristic is invalid data, the characteristic value to a null value; converting, in response to determining that the characteristic is a non-numeric characteristic, the characteristic value of the characteristic to an integer value, and normalizing, in response to determining that the characteristic is a non-port characteristic, the characteristic value of the characteristic.

A traffic shaping circuit regulates packets transferred by a transmission resource into a network (e.g., a network on a chip) on behalf of a client. The packet transfers are selectively enabled or disabled based on a current budget value. The budget value is modified based on a packet-transfer cost in response to transferring a packet into the network. The rate of packet transfers into the network is monitored. A cost-adjustment signal is generated based on the rate of packet transfers. The packet-transfer cost is modified in response to the cost-adjustment signal for accounting for a subsequent-packet transfer into the network. The cost-adjustment signal may indicate an increase or decrease of the packet-transfer cost and/or a budget limit, both of which are read from a cost table comprising records ordered based on respective packet-transfer cost values. The packet-transfer cost and/or a budget limit are configurable.

A traffic shaping circuit regulates packets transferred by a transmission resource into a network (e.g., a network on a chip) on behalf of a client. The packet transfers are selectively enabled or disabled based on a current budget value. The budget value is modified based on a packet-transfer cost in response to transferring a packet into the network. The rate of packet transfers into the network is monitored. A cost-adjustment signal is generated based on the rate of packet transfers. The packet-transfer cost is modified in response to the cost-adjustment signal for accounting for a subsequent-packet transfer into the network. The cost-adjustment signal may indicate an increase or decrease of the packet-transfer cost and/or a budget limit, both of which are read from a cost table comprising records ordered based on respective packet-transfer cost values. The packet-transfer cost and/or a budget limit are configurable.

Various embodiments are directed to techniques for dynamically adjusting a maximum rate of throughput for accessing data stored within a volume of storage space of a storage cluster system based on the amount of that data that is stored within that volume. An apparatus includes an access component to monitor an amount of client data stored within a volume defined within a storage device coupled to a first node, and to perform a data access command received from a client device via a network to alter the client data stored within the volume; and a policy component to limit a rate of throughput at which at least the client data within the volume is exchanged as part of performance of the data access command to a maximum rate of throughput, and to calculate the maximum rate of throughput based on the stored amount.

Systems and methods provide for generating traffic class-specific congestion signatures and other machine learning models for improving network performance. In some embodiments, a network controller can receive historical traffic data captured by a plurality of network devices within a first period of time that the network devices apply one or more traffic shaping policies for a predetermined traffic class and a predetermined congestion state. The controller can generate training data sets including flows of the historical traffic data labeled as corresponding to the predetermined traffic class and predetermined congestion state. The controller can generate, based on the training data sets, traffic class-specific congestion signatures that receive input traffic data determined to correspond to the predetermined traffic class and output an indication whether the input traffic data corresponds to the predetermined congestion state. The controller can adjust, based on the congestion signatures, traffic shaping operations of the plurality of network devices.

Systems and methods provide for generating traffic class-specific congestion signatures and other machine learning models for improving network performance. In some embodiments, a network controller can receive historical traffic data captured by a plurality of network devices within a first period of time that the network devices apply one or more traffic shaping policies for a predetermined traffic class and a predetermined congestion state. The controller can generate training data sets including flows of the historical traffic data labeled as corresponding to the predetermined traffic class and predetermined congestion state. The controller can generate, based on the training data sets, traffic class-specific congestion signatures that receive input traffic data determined to correspond to the predetermined traffic class and output an indication whether the input traffic data corresponds to the predetermined congestion state. The controller can adjust, based on the congestion signatures, traffic shaping operations of the plurality of network devices.