Merve Unuvar

US Patent:

20140067446, Mar 6, 2014

Filed:

Aug 29, 2012

Appl. No.:

13/598185

Inventors:

YURDAER N. DOGANATA - Yorktown Heights NY, US
GEETIKA TEWARI LAKSHMANAN - CAMBRIDGE MA, US
MERVE UNUVAR - CAMBRIDGE MA, US

Assignee:

International Business Machines Corporation - Armonk NY

International Classification:

G06Q 10/06

US Classification:

705 712

Abstract:

A method for training a machine learning tool to generate a prediction in a business process includes receiving a business process model corresponding to the business process, the business process model including a plurality of tasks, identifying a cycling set at a decision point in the business process model, wherein the cycling set comprises at least one task that the business process model iterates through, and building a training table by determining a total number of sub-traces and a total number of variables from a plurality of execution traces of the business process model based on the cycling set identified at the decision point, wherein a new row of the training table is created for each of the sub-traces and a new column of the training table is created for each of the variables.

US Patent:

20140067732, Mar 6, 2014

Filed:

Sep 6, 2012

Appl. No.:

13/605723

Inventors:

GEETIKA TEWARI LAKSHMANAN - Cambridge MA, US
MERVE UNUVAR - Cambridge MA, US

Assignee:

INTERNATIONAL BUSINESS MACHINES CORPORATION - Armonk NY

International Classification:

G06F 15/18

US Classification:

706 12

Abstract:

A method for training a machine learning tool to generate a prediction in a business process includes receiving a business process model corresponding to the business process, the business process model including a plurality of tasks, identifying a cycling set at a decision point in the business process model, wherein the cycling set comprises at least one task that the business process model iterates through, and building a training table by determining a total number of sub-traces and a total number of variables from a plurality of execution traces of the business process model based on the cycling set identified at the decision point, wherein a new row of the training table is created for each of the sub-traces and a new column of the training table is created for each of the variables.

US Patent:

20130103441, Apr 25, 2013

Filed:

Oct 21, 2011

Appl. No.:

13/279067

Inventors:

Yurdaer N. Doganata - Hawthorne NY, US
Rania Yousef Khalaf - Cambridge MA, US
Geetika T. Lakshmanan - Cambridge MA, US
Merve Unuvar - Cambridge MA, US

Assignee:

International Business Machines Corporation - Armonk NY

International Classification:

G06Q 10/06

US Classification:

705 712

Abstract:

A method for generating predictions includes dividing a business process model into fragments, wherein the business process model includes task nodes and at least one decision node, determining the decision node in at least one of the fragments, determining a decision tree for each decision node, determining a probability for reaching a terminal node in each fragment, and merging the probabilities obtained from the fragments to find a probability of a future task.

US Patent:

20180131631, May 10, 2018

Filed:

Dec 4, 2017

Appl. No.:

15/830079

Inventors:

- Armonk NY, US
Malgorzata STEINDER - Leonia NJ, US
Asser N. TANTAWI - Somers NY, US
Merve UNUVAR - New York NY, US

International Classification:

H04L 12/911
G06F 9/50
G06F 9/455

Abstract:

Disclosed is a novel system and method for managing requests for an additional virtual machine. The method begins with operating at least one virtual machine accessing at least one computer resource associated with at least one physical machine within a computing cluster. One or more non-deterministic virtual machine requests for the computer resource are received. An over-utilization of the computer resource as a probability distribution function is modeled. In one example, the probability distribution function is a Beta distribution function to represent a one of a plurality of probability distribution functions. Next, an additional virtual machine on the physical machine associated with the computer resource is added in response to a probability of a utilization of the computer resource being greater than a probalistic bound on the over-utilization of the computer resource. Otherwise, the additional virtual machine is not added.

US Patent:

20160283270, Sep 29, 2016

Filed:

Mar 24, 2015

Appl. No.:

14/667374

Inventors:

- Armonk NY, US
- Barcelona, ES
Iqbal I. Mohomed - Stamford CT, US
Asser N. Tantawi - Somers NY, US
Merve Unuvar - New York NY, US

International Classification:

G06F 9/50

Abstract:

Techniques for workload management in cloud computing infrastructures are provided. In one aspect, a method for allocating computing resources in a datacenter cluster is provided. The method includes the steps of: creating multiple, parallel schedulers; and automatically selecting a resource allocation method for each of the schedulers based on one or more of a workload profile, user requirements, and a state of the datacenter cluster, wherein an optimistic resource allocation method is selected for at least a first one or more of the schedulers and a pessimistic resource allocation method is selected for at least a second one or more of the schedulers. Due to optimistic resource allocation conflicts may arise. Methods to resolve such conflicts are also provided.

US Patent:

20160212062, Jul 21, 2016

Filed:

Mar 31, 2016

Appl. No.:

15/086755

Inventors:

- Armonk NY, US
Malgorzata STEINDER - Leonia NJ, US
Asser N. TANTAWI - Somers NY, US
Merve UNUVAR - New York NY, US

Assignee:

International Business Machines Corporation - Armonk NY

International Classification:

H04L 12/911
G06F 9/50

Abstract:

Disclosed is a novel system and method for managing requests for an additional virtual machine. The method begins with operating at least one virtual machine accessing at least one computer resource associated with at least one physical machine within a computing cluster. One or more non-deterministic virtual machine requests for the computer resource are received. An over-utilization of the computer resource as a probability distribution function is modeled. In one example, the probability distribution function is a Beta distribution function to represent a one of a plurality of probability distribution functions. Next, an additional virtual machine on the physical machine associated with the computer resource is added in response to a probability of a utilization of the computer resource being greater than a probalistic bound on the over-utilization of the computer resource. Otherwise, the additional virtual machine is not added.

US Patent:

20160134557, May 12, 2016

Filed:

Nov 12, 2014

Appl. No.:

14/539543

Inventors:

- Armonk NY, US
Merve Unuvar - New York NY, US

International Classification:

H04L 12/911
G06Q 10/06

Abstract:

Embodiments of the invention provide a method, a system and a computer program product configured to automatically auto-scale a user compute instance to multiple cloud providers while considering a multiplicity of user requirements. The method, executed on a digital data processor, includes obtaining information, via a user interface, that is descriptive of user cloud computing related preferences, including a user cloud computing budgetary preference. The method further includes sensing properties of a plurality of clouds and making decisions, based at least on the obtained information and on the sensed properties, of when to scale up or scale down the user cloud instance, of selecting one of the plurality of clouds as where to scale the user cloud instance, and determining which resource or resources of the selected cloud to add or remove from the selected cloud. The method further includes automatically executing the decisions on the selected cloud.

US Patent:

20160134558, May 12, 2016

Filed:

Jun 22, 2015

Appl. No.:

14/745972

Inventors:

- Armonk NY, US
Merve Unuvar - New York NY, US

International Classification:

H04L 12/911
H04L 29/08

Abstract:

Embodiments of the invention provide a method, a system and a computer program product configured to automatically auto-scale a user compute instance to multiple cloud providers while considering a multiplicity of user requirements. The method, executed on a digital data processor, includes obtaining information, via a user interface, that is descriptive of user cloud computing related preferences, including a user cloud computing budgetary preference. The method further includes sensing properties of a plurality of clouds and making decisions, based at least on the obtained information and on the sensed properties, of when to scale up or scale down the user cloud instance, of selecting one of the plurality of clouds as where to scale the user cloud instance, and determining which resource or resources of the selected cloud to add or remove from the selected cloud. The method further includes automatically executing the decisions on the selected cloud.