System and method for distributed computation using heterogeneous computing nodes

Patent metadata

Field	Value
Publication number	US-2016140359-A1
Application number	US-201414900061-A
Country	US
Kind code	A1
Filing date	Jun 9, 2014
Priority date	Jun 20, 2013
Publication date	May 19, 2016
Grant date	—

Abstract

This disclosure relates generally to the use of distributed system for computation, and more particularly, relates to a method and system for optimizing computation and communication resource while preserving security in the distributed device for computation. In one embodiment, a system and method of utilizing plurality of constrained edge devices for distributed computation is disclosed. The system enables integration of the edge devices like residential gateways and smart phone into a grid of distributed computation. The edged devices with constrained bandwidth, energy, computation capabilities and combination thereof are optimized dynamically based on condition of communication network. The system further enables scheduling and segregation of data, to be analyzed, between the edge devices. The system may further be configured to preserve privacy associated with the data while sharing the data between the plurality of devices during computation.

First claim

We claim:

1 . A system for distributed computation in a communication network, the system comprising:
a cluster monitoring module configured to receive data for computation; a privacy module configured to provide privacy measurement by performing sensitivity analysis on the data received by the cluster monitoring module; and a communication module configured to transfer the data for efficient load distribution during computation, and to preserve privacy, wherein the communication module is further configured to optimize bandwidth usage and energy consumption during the transfer of the data over the communication network.

2 . The system of claim 1 , further comprising a plurality of edge devices, wherein one or more of the cluster monitoring module, the privacy module, and the communication module are hosted on one or more of the edge devices among the plurality of edge devices.

3 . The system of claim 2 , wherein the plurality of edge devices further comprises:
a processor; a memory coupled to the processor, wherein the memory comprises the clustering module, the privacy module, and the communication module.

4 . The system of claim 1 , wherein the communication module is further configured to reduce traffic via group communication based communication scheme, to collaborate among the plurality of edge devices, and to dynamically utilize unused network capacity of the communication network.

5 . The system of claim 1 , wherein the bandwidth usage and energy consumption is optimized by compressing data by network encoding the data prior to the transfer of data, and formulating a relationship between length of coded data and number of edge devices from the plurality of edge devices.

6 . The system of claim 1 , wherein the privacy module further enables privacy preservation, wherein the privacy preservation is based on at least one factor selected from computation capability of the at least one edge device from the plurality of edge devices, privacy-utility requirement, and sensitivity of the data.

7 . The system of claim 1 , further comprising a partitioning module configured to segregate the data into smaller data sets, wherein the smaller data sets are analyzed simultaneously by the plurality of edge devices.

8 . The system of claim 7 , wherein size of the smaller data sets is governed by the computation capability of the plurality of edge devices, and network characteristics, wherein the network characteristics includes one or more of bandwidth available to the plurality of the edge devices, round trip latency.

9 . The system of claim 1 , wherein the cluster monitoring module is further configured to create a list of partitions and the size for the smaller data sets, and detect failure of edge devices from the plurality of edge devices.

10 . The system of claim 1 , further comprising a core scheduler module, wherein the core scheduler module enables scheduling for the computation of the data, wherein the scheduling of the data for computation is based on the list of partitions and the size for the smaller data sets created by the cluster monitoring module.

11 . A device for distributed computation, wherein the device is configured to engage a plurality of edge devices over a communication network, the device comprising:
a processor; a memory coupled to the processor, wherein the memory comprises a plurality of modules capable of being executed by the processor, and wherein the plurality of modules comprising:
a cluster monitoring module configured to receive data for computation;
a privacy module configured to provide privacy measurement by performing sensitivity analysis on the data received by the cluster monitoring module; and a communication module configured to transfer the data to the plurality of edge devices for efficient load distribution during computation, and to preserve privacy, wherein the communication module is further configured to optimize bandwidth usage and energy consumption during the transfer of the data over a communication network.

12 . A method for distributed computation, wherein a plurality of edge devices are used for distributed computation along with a backend server, the method comprising:
receiving data for computation from at least one edge device selected from the plurality of edge devices; segregating the data into smaller data sets of varying size, based on a list of partitions created; allocating the smaller data sets to the plurality of edge devices for subsequent analysis; and optimizing bandwidth and energy usage during allocation of the smaller data sets to the plurality of devices, while preserving privacy during said allocation.

13 . The method of claim 12 , further comprises detecting failure of a edge device from the plurality edge devices during the analyses.

14 . The method of claim 12 , wherein the smaller data sets of varying size are obtained based on the computation capability of the plurality of edge devices, and network characteristics such as the bandwidth available to the plurality of the edge devices and round trip latency.

15 . The method of claim 12 , wherein the bandwidth and the energy usage are optimized by enabling data transfer amongst the edge devices at the backend server, between the edge devices and the backend server and amongst the edge devices.

16 . The method of claim 12 , wherein the bandwidth and the energy usage are further optimized by network encoding the data to compress the data prior to the transfer of data, and formulating a relationship between length of coded data and number of edge devices from the plurality of edge devices.

17 . The method of claim 12 , wherein the allocation further comprises, scheduling of the data set for analysis based on the list of partitions, available edge devices, available bandwidth, other network channel characteristics, and desired level of privacy preservation.

Assignees

• Tata Consultancy Services Ltd

Inventors

• Mukherjee Arijit
• Bandyopadhyay Soma
• Ukil Arijit
• Bhattacharyya Abhijan
• Dey Swarnava
• Pal Arpan
• Paul Himadri Sekhar

Classifications

• H04W12/02

  Protecting privacy or anonymity, e.g. protecting personally identifiable information [PII] · CPC title

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• H04L63/0407

  wherein the identity of one or more communicating identities is hidden (cryptographic mechanisms or cryptographic arrangements for anonymous credentials or for identity based cryptographic systems H04L9/00) · CPC title

• G06F21/6263 Primary

  during internet communication, e.g. revealing personal data from cookies · CPC title

• G06F21/6245 Primary

  Protecting personal data, e.g. for financial or medical purposes · CPC title