Resumes
Resumes
Sandip Agarwala Phones & Addresses
-
10630 Mine Ct, Cupertino, CA 95014
-
Foothill Ranch, CA
-
Cool, CA
-
355 Wolfe Rd, Sunnyvale, CA 94085 (408) 962-0460
-
Atlanta, GA
-
San Jose, CA
-
Santa Clara, CA
Publications
Us Patents
Online Management Of Historical Data For Efficient Reporting And Analytics
View pageUS Patent:
8306953, Nov 6, 2012
Filed:
Aug 31, 2010
Appl. No.:
12/872964
Inventors:
Sandip Agarwala - Sunnyvale CA, US
Sandeep Gopisetty - Morgan Hill CA, US
Stefan Jaquet - Morgan Hill CA, US
Sandeep Gopisetty - Morgan Hill CA, US
Stefan Jaquet - Morgan Hill CA, US
Assignee:
International Business Machines Corporation - Armonk NY
International Classification:
G06F 7/00
G06F 17/30
G06F 17/30
US Classification:
707688, 707736, 702179, 702187, 703 2
Abstract:
Embodiments for efficiently computing complex statistics from historical time series data are provided. A hierarchical summarization method includes receiving at least one stream of data and creating data blocks from the at least one stream of data. In another embodiment, a method for computing statistics for historical data includes accessing at least one online stream of historical data, the online stream of historical data including metadata, and creating data blocks from the at least one online stream of historical data. Each data block includes a pair of timestamps indicating a sampling start time and a sampling end time, a number of data samples spanned by the data block, a SUM(X) statistic, a SUM(XX) statistic, and a SUM(XY) statistic computed for the data samples spanned by the data block. Other methods are also presented, such as methods for efficiently and accurately calculating statistical queries regarding historical data for arbitrary time ranges, among others.
Allocation Of Storage Resources In A Networked Computing Environment Based On Energy Utilization
View pageUS Patent:
8407501, Mar 26, 2013
Filed:
Mar 28, 2011
Appl. No.:
13/073081
Inventors:
Sandip Agarwala - Sunnyvale CA, US
Eric K. Butler - San Jose CA, US
Sandeep Gopisetty - Morgan Hill CA, US
Kavita Chavda - Roswell GA, US
Eric K. Butler - San Jose CA, US
Sandeep Gopisetty - Morgan Hill CA, US
Kavita Chavda - Roswell GA, US
Assignee:
International Business Machines Corporation - Armonk NY
International Classification:
G06F 1/26
US Classification:
713320, 709226
Abstract:
Embodiments of the present invention provide an approach to provision storage resources (e. g. , across an enterprise storage system (ESS) such as a general parallel file system (GPFS) or the like) for different workloads in an energy efficient manner. The system evaluates different energy profiles/workloads' energy consumption characteristics of storage devices to determine an allocation plan that reduces the energy cost (e. g. , results in the lowest cost/energy consumption for handling a storage workload). In a typical embodiment, energy consumption characteristics for handling a particular storage workload will be determined. Thereafter, a type of storage device capable of handling the workload will be determined. Then, an allocation plan that results in the most efficient energy consumption for handling the workload will be developed. In general, the allocation plan is based upon the energy consumption characteristics and an energy efficiency algorithm.
End-To-End Provisioning Of Storage Clouds
View pageUS Patent:
8478845, Jul 2, 2013
Filed:
Aug 16, 2010
Appl. No.:
12/857005
Inventors:
Sandip Agarwala - Sunnyvale CA, US
Richard J. Ayala - Medford NJ, US
Kavita Chavda - Roswell GA, US
Sandeep Gopisetty - Morgan Hill CA, US
Richard J. Ayala - Medford NJ, US
Kavita Chavda - Roswell GA, US
Sandeep Gopisetty - Morgan Hill CA, US
Assignee:
International Business Machines Corporation - Armonk NY
International Classification:
G06F 15/177
G06F 15/167
G06F 15/173
G06F 15/167
G06F 15/173
US Classification:
709220, 709216, 709223, 709224, 709226
Abstract:
Embodiments discussed in this disclosure provide an integrated provisioning framework that automates the process of provisioning storage resources, end-to-end, for an enterprise storage cloud environment. Such embodiments configure and orchestrate the deployment of a user's workload and, at the same time, provide optimization across a multitude of storage cloud resources. Along these lines, input is received in the form of workload requirements and configuration information for available system resources. Based on the input, a set (at least one) of storage cloud configuration plans is developed that satisfy the workload requirements. A set of scripts is then generated that orchestrate the deployment and configuration of different software and hardware components based on the plans.
Method And System For Chargeback Allocation In Information Technology Systems
View pageUS Patent:
8515792, Aug 20, 2013
Filed:
Sep 25, 2009
Appl. No.:
12/567538
Inventors:
Sandip Agarwala - Sunnyvale CA, US
Ramani R. Routray - San Jose CA, US
Sandeep M. Uttamchandani - San Jose CA, US
Ramani R. Routray - San Jose CA, US
Sandeep M. Uttamchandani - San Jose CA, US
Assignee:
International Business Machines Corporation - Armonk NY
International Classification:
G06Q 10/00
US Classification:
705 711, 705 735, 705 722, 705 736, 705410
Abstract:
The invention provides a system and method for chargeback cost allocation in an information technology (IT) system including multiple resources. The method includes categorizing cost attributes of different resources into multiple levels defining a cost attribute hierarchy, defining different chargeback groups for resources with similar cost attributes and chargeback policies at different levels of the hierarchy, and performing chargeback cost allocation by allocating the cost for resources at each hierarchy level independently using chargeback policies defined for the resources at that hierarchy level.
Cost And Power Efficient Storage Area Network Provisioning
View pageUS Patent:
8515967, Aug 20, 2013
Filed:
Mar 29, 2010
Appl. No.:
12/749435
Inventors:
Sandip Agarwala - Sunnyvale CA, US
Harsha D. Gunatilaka - Tucson AZ, US
Ramani R. Routray - San Jose CA, US
Harsha D. Gunatilaka - Tucson AZ, US
Ramani R. Routray - San Jose CA, US
Assignee:
International Business Machines Corporation - Armonk NY
International Classification:
G06F 7/00
US Classification:
707748, 707769, 711101, 711117, 711147, 713300
Abstract:
Various embodiments for efficiently provisioning a storage area network (SAN) are provided. In one embodiment, SAN information is provided to an engine for optimization. The SAN information includes at least one of SAN configuration information, SAN usage information, at least one cost profile, and at least one chargeback model. Based on the SAN information, those of an available plurality of storage resources not meeting at least one storage criterion are filtered. The filtered storage resources are ranked on a cost basis. A resource configuration graph is constructed based on the ranked storage resources. The resource configuration graph is traversed to obtain a plurality of possible SAN configuration plans. At least one power profile is applied to the plurality of possible SAN configuration plans to rank the plurality of possible SAN configuration plans by energy consumption.
Performance Isolation For Storage Clouds
View pageUS Patent:
8554917, Oct 8, 2013
Filed:
Aug 20, 2010
Appl. No.:
12/859788
Inventors:
Sandip Agarwala - Sunnyvale CA, US
Sandeep Gopisetty - Morgan Hill CA, US
Seshashayee S. Murthy - Yorktown Heights NY, US
Sandeep Gopisetty - Morgan Hill CA, US
Seshashayee S. Murthy - Yorktown Heights NY, US
Assignee:
International Business Machines Corporation - Armonk NY
International Classification:
G06F 15/173
US Classification:
709226
Abstract:
Embodiments of the present invention provide performance isolation for storage clouds. Under one embodiment, workloads across a storage cloud architecture are grouped into clusters based on administrator or system input. A performance isolation domain is then created for each of the clusters, with each of the performance isolation domains comprising a set of data stores associated with a set of storage subsystems and a set of data paths that connect the set of data stores to a set of clients. Thereafter, performance isolation is provided among a set of layers of the performance isolation domains. Such performance isolation is provided by (among other things): pooling data stores from separate performance isolation domains into separate pools; assigning the pools to device adapters, RAID controller, and the set of storage subsystems; preventing workloads on the device adapters from exceeding capacities of the device adapters; mapping the set of data stores to a set of Input/Output (I/O) servers based on an I/O capacity and I/O load of the set of I/O servers; and/or pairing ports of the set of I/O servers with ports of the set of storage subsystems, the pairing being based upon availability, connectivity, I/O load, and I/O capacity.
Chargeback Reduction Planning For Information Technology Management
View pageUS Patent:
20110078695, Mar 31, 2011
Filed:
Sep 25, 2009
Appl. No.:
12/567582
Inventors:
Sandip Agarwala - Sunnyvale CA, US
Ramani R. Routray - San Jose CA, US
Sandeep M. Uttamchandani - San Jose CA, US
Ramani R. Routray - San Jose CA, US
Sandeep M. Uttamchandani - San Jose CA, US
Assignee:
International Business Machines Corporation - Armonk NY
International Classification:
G06F 9/46
US Classification:
718104
Abstract:
Reducing cost chargeback in an information technology (IT) computing environment including multiple resources, is provided. One implementation involves a process wherein resource usage and allocation statistics are stored for a multitude of resources and associated cost policies. Then, time-based usage patterns are determined for the resources from the statistics. A correlation of response time with resource usages and outstanding input/output instructions for the resources is determined. Based on usage patterns and the correlation, a multitude of potential cost reduction recommendations are determined. Further, a multitude of integrals are obtained based on the potential cost reduction recommendations, and a statistical integral is obtained based on the statistics. A difference between the statistical integral and each of the multiple integrals is obtained and compared with a threshold to determine potential final cost reduction recommendations. A final cost reduction recommendation is then selected from the potential cost reduction recommendations.
Comprehensive Bottleneck Detection In A Multi-Tier Enterprise Storage System
View pageUS Patent:
20120233310, Sep 13, 2012
Filed:
Mar 9, 2011
Appl. No.:
13/044048
Inventors:
Sandip Agarwala - Sunnyvale CA, US
David P. Goodman - Loveland CO, US
Sandeep Gopisetty - Morgan Hill CA, US
David D. Weck - Lexington KY, US
David P. Goodman - Loveland CO, US
Sandeep Gopisetty - Morgan Hill CA, US
David D. Weck - Lexington KY, US
Assignee:
International Business Machines Corporation - Armonk NY
International Classification:
G06F 15/173
US Classification:
709224
Abstract:
Embodiments of the present invention provide approaches (e.g., online methods) to analyze end-to-end performance issues in a multi-tier enterprise storage system (ESS), such as a storage cloud, where data may be distributed across multiple storage components. Specifically, performance and configuration data from different storage components (e.g., nodes) is collected and analyzed to identify nodes that are becoming (or may become) performance bottlenecks. In a typical embodiment, a set of components distributed among a set of tiers of an ESS is identified. For each component, a total capacity and a current load are determined. Based on these values, a utilization of each component is determined. Comparison of the utilization with a predetermined threshold and/or analysis of historical data allows one or more components causing a bottleneck to be identified.