Implementations offload visual frames from a client device to a gateway or edge device for processing. The client device can receive streaming visual frames and a request to process the visual frames using a data service. The client device can offload visual frames to a gateway or edge device preloaded with a resource that corresponds to the requested data service. After the gateway or edge device processes the visual frames using the resource, the processed visual frames can be returned to the client device. In implementations, the offload device and client device are situated in a network such that a latency for the offload communications supports real-time video display. A gateway device manager can locate a gateway connected to the client device, and resources can be deployed (or the gateway can be prepopulated) so that the device can perform gateway services and edge processing services for the client device.

Disclosed herein are embodiments of a cloud data synchronization system enabling an user operating a mobile client device to download mission-specific data sets from a fixed cloud-based server system to a database of the mobile client device, and then use the downloaded data sets independently on the mobile client device when the mobile client device is disconnected from a network connecting to the fixed cloud-based server system. When connectivity to the fixed cloud-based server system is re-established by the mobile client device in an intermittent and bandwidth-limited communication network environment, the fixed cloud-based server system may provide bi-directional data synchronization between records of the fixed cloud-based server system and the mobile client device to update the data sets on the fixed cloud-based server system and the mobile client device while operating in the intermittent and bandwidth-limited communication network environment.

A user equipment (UE) or other device performs service discovery of edge computing resources in a cellular network system and dynamic offloading of UE application tasks to discovered edge computing resources. As part of the discovery process, the device (e.g., the UE) may request edge server site capability information. When performing dynamic offloading, the UE may obtain (collect and/or receive) information regarding channel conditions, cellular network parameters or application requirements and dynamically determine whether a task of the application executing on the UE should be offloaded to an edge server or executed locally on the UE. In making decisions between offloaded or local execution, the UE may use a utility function that takes into account factors such as relative differences in application latency, energy consumption and offloading cost.

A user equipment (UE) or other device performs service discovery of edge computing resources in a cellular network system and dynamic offloading of UE application tasks to discovered edge computing resources. As part of the discovery process, the device (e.g., the UE) may request edge server site capability information. When performing dynamic offloading, the UE may obtain (collect and/or receive) information regarding channel conditions, cellular network parameters or application requirements and dynamically determine whether a task of the application executing on the UE should be offloaded to an edge server or executed locally on the UE. In making decisions between offloaded or local execution, the UE may use a utility function that takes into account factors such as relative differences in application latency, energy consumption and offloading cost.

Embodiments of the invention are directed to enabling access transaction systems to accept different communication protocols. In some embodiment, an access device receives, from a portable device, an indication that a transaction is to be performed by exchanging transaction information between the portable device and a remote computer, wherein the remote computer is configured to communicate using a first communication protocol. Next, the access device determines that the portable device is configured to communicate using a second communication protocol. The access device then converts communications between the portable device and the remote computer from the second communication protocol to the first communication protocol to assist the portable device and the remote computer in exchanging the transaction information.

Embodiments of the invention are directed to enabling access transaction systems to accept different communication protocols. In some embodiment, an access device receives, from a portable device, an indication that a transaction is to be performed by exchanging transaction information between the portable device and a remote computer, wherein the remote computer is configured to communicate using a first communication protocol. Next, the access device determines that the portable device is configured to communicate using a second communication protocol. The access device then converts communications between the portable device and the remote computer from the second communication protocol to the first communication protocol to assist the portable device and the remote computer in exchanging the transaction information.

In an exemplary process for remote execution of machine-learned models, one or more signals from a second electronic device is detected by a first electronic device. The second electronic device includes a machine-learned model associated with an application implemented on the first electronic device. Based on the one or more signals, a communication connection is established with the second electronic device and a proxy to the machine-learned model is generated. Input data is obtained via a sensor of the first electronic device. A representation of the input data is sent to the second electronic device via the proxy and the established communication connection. The representation of the input data is processed through the machine-learned model to generate an output. A result derived from the output is received via the communication connection and a representation of the result is outputted.

In an exemplary process for remote execution of machine-learned models, one or more signals from a second electronic device is detected by a first electronic device. The second electronic device includes a machine-learned model associated with an application implemented on the first electronic device. Based on the one or more signals, a communication connection is established with the second electronic device and a proxy to the machine-learned model is generated. Input data is obtained via a sensor of the first electronic device. A representation of the input data is sent to the second electronic device via the proxy and the established communication connection. The representation of the input data is processed through the machine-learned model to generate an output. A result derived from the output is received via the communication connection and a representation of the result is outputted.

A method for fetching a content from a web server to a client device is disclosed, using tunnel devices serving as intermediate devices. The client device accesses an acceleration server to receive a list of available tunnel devices. The requested content is partitioned into slices, and the client device sends a request for the slices to the available tunnel devices. The tunnel devices in turn fetch the slices from the data server, and send the slices to the client device, where the content is reconstructed from the received slices. A client device may also serve as a tunnel device, serving as an intermediate device to other client devices. Similarly, a tunnel device may also serve as a client device for fetching content from a data server. The selection of tunnel devices to be used by a client device may be in the acceleration server, in the client device, or in both. The partition into slices may be overlapping or non-overlapping, and the same slice (or the whole content) may be fetched via multiple tunnel devices.

A method for fetching a content from a web server to a client device is disclosed, using tunnel devices serving as intermediate devices. The client device accesses an acceleration server to receive a list of available tunnel devices. The requested content is partitioned into slices, and the client device sends a request for the slices to the available tunnel devices. The tunnel devices in turn fetch the slices from the data server, and send the slices to the client device, where the content is reconstructed from the received slices. A client device may also serve as a tunnel device, serving as an intermediate device to other client devices. Similarly, a tunnel device may also serve as a client device for fetching content from a data server. The selection of tunnel devices to be used by a client device may be in the acceleration server, in the client device, or in both. The partition into slices may be overlapping or non-overlapping, and the same slice (or the whole content) may be fetched via multiple tunnel devices.