Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Proximity Based Clustered Computing

IP.com Disclosure Number: IPCOM000244780D
Publication Date: 2016-Jan-14
Document File: 2 page(s) / 30K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method to enable proximity based computing where mobile devices are connected or are part of an ad hock opt in peer-to-peer network or grid/mesh network, and communicating by means of one of the wireless protocols. Devices are expected to share a portion of the associated compute resource with a compute resources coordinator, where a subset of available users within a certain locale are selected based on some desirable criteria, to participate in the completion of a compute task.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Proximity Based Clustered Computing

Mobile devices are continually increasing storage and connectivity capabilities; however, most mobile devices are underutilized. At the same time, the compute density of mobile devices can be quite high depending on the correlating user density (i.e. when mobile device users congregate at the mall, in traffic, at a sporting event, etc.).

A method is needed to pool mobile devices as a domain controller (DC).

The novel contribution is a system and method to enable proximity based computing. This is done where mobile devices connect as part of an ad hock opt-in peer-to-peer network or grid/mesh network, and communicate by means of one of the wireless protocols. In this situation, devices can share a portion of associated compute resources with a compute resources coordinator. The device (owner) must be willing to relocate to a specific venue when requested, when low-density compute is identified or anticipated.

A resources coordinator selects potential candidates for a compute campaign (see method (1)). The system queries the devices to determine which are fit for the task; the anonymity of each device is maintained if desired. The selection criteria for a pool of potential candidates are:


 Density of the pool


 Likelihood of identifying desirable candidates/device from selected pool


 Device owner's schedule (i.e. likelihood that the owner will remain at current location for the time needed to transfer and complete compute task)


 Device characteristics (e.g., specifications, remaining battery life, available bandwidth, interconnect speed, etc.)


 Stability of the device's position, if devices remain in a location threshold (e.g., erratic movement in position)


 Device registers as a device willing to participate


 Reputation of supplying resources for other tasks (i.e. must be good)


 Device (owner) is willing to relocate to a specific venue when requested, when low density compute is identified or anticipated

In a preferred embodiment, devices (or device owners) agree to allow a portion of the associated compute resource to be shared with the ad hoc gro...