I got an opportunity to work with Verizon Services Operations VSO within the supply and chain department. It was a wonderful experience to work with some of the most experienced people of the industry.
- Used the SAP Security tool to create composite roles and assigned authorizations to different users within Verizon using these roles.
- Successfully created Access database with relationships and wrote queries to generate excel reports from the tables.
- Researched about the Oracle Application Integration Architecture (AIA) tool and created an evaluation report on the possibility of using the tool for the Procure to Pay Project.
- Created a summary status report on the VZW to Peg@s Project.
I worked on a project titled "Sorting of Products using PLC and Barcode". The project in consideration used a barcode reader connected to a Programmable Logic Controller used for sorting of products in an industry. The project thus revolves around the entire sorting process with the help of a PLC. We used Siemens PLS S7 - 200 CPU 226 for the project.
- Worked with a team on the project which involved Sorting of products on a conveyer belt with solenoid flaps using barcode reader interfaced to a programmable logic controller to improve efficiency in the packaging industry.
- Used MICROWIN software to program the PLC and created an RS 485 to RS 232C converter which is used for interfacing the barcode reader with PLC.
- Created a mimic which would act as an indicating control panel.
To view a presentation on the project, click on this link
Worked on Fabrication of Heat Exchangers for IPCL using fractional distillation process and procurement of critical components as per detail drawing.
- This project involved implementation of quicksort in C using sequential (single processor), message passing tool MPI (multiprocessors with distributed memory) and pthreads (mutliprocessors with shared memory).
- Multiprocessor implementaions were carried out using cluster computing and upto 40 processors were used in parallel to carry out mega bytes of sorting
- Benchmark results showed that mutliprocessor implementation methods drastically improved efficiency
- For 1 MB sort, Number of processes/ threads were varied and execution time was calculated. Results obtained are as follows. We can see that minimum execution time for MPI and Pthreads is obtained at approximately 10 processes / threads.
- Speed up analysis was also done for different set of experiments [Speed up = (Execution time using sequential algorithm) / (Execution time using multiprocessor using p processes)]. Results obtained are as follows
To view a complete report on this project, click on this link
- Project involved prediction of stock prices using technologies like PHP, Matlab, ASP.NET, MYSQL, web services and HTML
- The system collected data from web sources like Yahoo Finance, Google finance at a time interval of 15 minutes and update its database for a set of 25 companies.
- The system maintained a host of web services that fairly link different service modules with client interface.
- Different prediction models were built that gave a fair recommendation to the customer whether to buy/sell/hold stock. The system trained itself from a set of past data and simulated on a remaining set of test data to improve its precision.
To view a complete report on this project, click on this link
The project involved simulating a multicast network using C++ multithreading in Linux environment. The multicast system involved different modules as follows:-
- Discovery and Hello – The script file specifies the topology to be used for the multicast system. Each router that is initiated reads from the script file and hence knows of the number of interfaces it has and also it knows if a router or host is connected on each of its interface.
- RIP protocol – It uses RIP with split horizon with poisson technique to enable the RIP protocol to converge quickly. Everytime a router receives a RIP packet from its neighbors it updates its routing table and if this causes a change in the routing table then a routeupdate flag is made true.
- Multicasting – To implement enhanced forward and fork multicast, routing tables of neighbors are stored. Each node can send data, destined for a maximum of 3 destinations.
- ACK is used for RIP and data packets. Sequence number is used to implement Stop n Wait ARQ.