Predictive queue control and allocation

Patent metadata

Field	Value
Publication number	US-10606604-B2
Application number	US-201715682739-A
Country	US
Kind code	B2
Filing date	Aug 22, 2017
Priority date	Aug 22, 2017
Publication date	Mar 31, 2020
Grant date	Mar 31, 2020

Abstract

A predictive queue control and allocation system includes a queue and a queue control server communicatively coupled to the queue. The queue includes a first and second allocation of queue locations. The queue stores a plurality of resources. The queue control server includes an interface and a queue control engine implemented by a processor. The interface monitors the plurality of resources before the plurality of resources are stored in the queue. The queue control engine predicts that one or more conditions indicate that a queue overflow will occur in the first allocation of queue locations. The queue control engine prioritizes the plurality of resources being received by the queue. The queue control engine may apply a machine learning technique to the plurality of resources. The queue control engine transfers the plurality of resources prioritized by the machine learning technique.

First claim

The invention claimed is:

1. A predictive queue control and allocation system comprising:
a queue comprising a first allocation of queue locations and a second allocation of queue locations, the queue configured to store a plurality of resources;
a queue control server communicatively coupled to the queue, the queue control server comprising:
an interface operable to:
monitor the plurality of resources before the plurality of resources are stored in the queue; and

a queue control engine implemented by a processor operably coupled to the interface, and configured to:
predict that one or more conditions indicate that a queue overflow will occur in the first allocation of queue locations before an overflow occurs, wherein predicting that the queue overflow will occur comprises:
determining a historical input rate for the first allocation of queue locations;
determining a current input rate for the first allocation of queue locations; and
determining that the current input rate exceeds the historical input rate for the first allocation of queue locations;

prioritize the plurality of resources being received by the queue according to a first resource type and a second resource type, wherein:
the first resource type is stored in the first allocation of queue locations; and
the second resource type is stored in the second allocation of queue locations;

apply a machine learning technique to the plurality of resources stored in the first allocation of queue locations, wherein:
applying the machine learning technique comprises:
 identifying information being analyzed for an application; and
 determining a correlation between the information being analyzed for the application and resource data in the plurality of resources stored in the first allocation of queue locations; and
the machine learning technique prioritizes resource data stored in the first allocation of queue locations as the first resource type or the second resource type based on the correlation between the information being analyzed for the application and the resource data in the plurality of resources stored in the first allocation of queue locations; and

transfer the plurality of resources prioritized by the machine learning technique as the second resource type from the first allocation of queue locations to the second allocation of queue locations.

2. The predictive queue control and allocation system of claim 1 , wherein the first resource type and the second resource type are determined by an application that utilizes the plurality of resources stored in the queue.

3. The predictive queue control and allocation system of claim 1 , wherein in order to determine that one or more conditions indicate that a queue overflow will occur in the first allocation of queue locations, the queue control engine is further configured to:
calculate an input rate of the plurality of resources being stored in the first allocation of queue locations;
calculate a processing rate of the plurality of resources being stored in the first allocation of queue locations;
determine a throughput rate based on the input rate and the processing rate.

4. The predictive queue control and allocation system of claim 3 , wherein the queue control engine is configured to prioritize resource data and apply the machine learning technique when the throughput rate is greater than a predetermined rate.

5. The predictive queue control and allocation system of claim 1 , wherein the plurality of resources comprise market trade data.

6. The predictive queue control and allocation system of claim 1 , wherein the queue further comprises a backup database configured to store the plurality of resources.

7. The predictive queue control and allocation system of claim 6 , wherein queue control server is further configured to:
prioritize the plurality of resources being received by the queue according to a third resource type;
apply the machine learning technique to the plurality of resources stored in the first allocation of queue locations, wherein the machine learning technique further prioritizes resource data stored in the first allocation of queue locations using the third resource type; and
transfer the plurality of resources prioritized by the machine learning technique as the second resource type from the first allocation of queue locations to the backup database.

8. A method for predictive queue control and allocation, comprising:
storing, at a queue, a plurality of resources, wherein the queue comprises first allocation of queue locations and a second allocation of queue locations;
monitoring, via an interface, the plurality of resources before the plurality of resources are stored in the queue; and
predicting, via a processor, that one or more conditions indicate that a queue overflow will occur in the first allocation of queue locations before an overflow occurs, wherein predicting that the queue overflow will occur comprises:
determining a historical input rate for the first allocation of queue locations;
determining a current input rate for the first allocation of queue locations; and
determining that the current input rate exceeds the historical input rate for the first allocation of queue locations;

prioritizing, via the processor, the plurality of resources being received by the queue according to a first resource type and a second resource type, wherein:
the first resource type is stored in the first allocation of queue locations; and
the second resource type is stored in the second allocation of queue locations;

applying, via the processor, a machine learning technique to the plurality of resources stored in the first allocation of queue locations, wherein:
applying the machine learning technique comprises:
identifying information being analyzed for an application; and
determining a correlation between the information being analyzed for the application and resource data in the plurality of resources stored in the first allocation of queue locations; and
the machine learning technique prioritizes resource data stored in the first allocation of queue locations as the first resource type or the second resource type based on the correlation between the information being analyzed for the application and the resource data in the plurality of resources stored in the first allocation of queue locations; and

transferring, via the processor, the plurality of resources prioritized by the machine learning technique as the second resource type from the first allocation of queue locations to the second allocation of queue locations.

9. The predictive queue control and allocation method of claim 8 , wherein the first resource type and the second resource type are determined by an application that utilizes the plurality of resources stored in the queue.

10. The predictive queue control and allocation method of claim 8 , wherein determining that one or more conditions indicate that a queue overflow will occur in the first allocation of queue locations, comprises:
calculating an input rate of the plurality of resources being stored in the first allocation of queue locations;
calculating a processing rate of the plurality of resources being stored in the first allocation of queue locations;
determining a throughput rate based on the input rate and the processing rate.

11. The predictive queue control and allocation method of claim 10 , wherein the method prioritizes resource data and applies the machine learning technique when the throughput rate is greater than a predetermined rate.

12. The predictive queue control and allocation method of claim 8 , wherein the plurality of resources comprise market tr

Assignees

• Bank Of America

Inventors

• Sharma Anuj
• Srivastava Gaurav
• Kelkar Vishal D

Classifications

• G06F9/5011 Primary

  the resources being hardware resources other than CPUs, Servers and Terminals · CPC title

• G06F9/3806

  using address prediction, e.g. return stack, branch history buffer · CPC title

• G06F9/3851 Primary

  from multiple instruction streams, e.g. multistreaming · CPC title

• G06F9/3855

  Physics · mapped topic

• G06F9/3856

  Reordering of instructions, e.g. using queues or age tags · CPC title