Venkata Chaitanya Jonnalagadda

Hi. Uh, my name is. So I in the SQL, Lambda SQL, Python SQL, and other

So I utilized T SQL in, uh, complex SQL queries by trying to understand the ask first and then, uh, create a solution in my mind and try to implement that with T SQL. It could be incorporating the functions or different procedures while, uh, solving complex problems. Uh

So to optimize this equal query that is running too long, I would see to, uh, I would first try to see the cost where exactly, uh, the query is taking, uh, long, actually. So I would run the stats, uh, to see where, uh, the cost is getting more. And then I try to gather the stats or try to find fine tune the query by, uh, removing the redundant checks and, uh, filtering the, uh, and I would like to filter only the records that are needed and stuff retrieving, uh, by giving select star. So I will try to, uh, I would basically pull the required fields for the query, and I would, uh, I would use subqueries not necessary subqueries instead of in because in will take a lot of time in, uh, also in joins also. So wherever the cost is more for the or the time is taking more for the

So for for for retrieving data from, uh, from multiple tables in the database, I would do a join on the, uh, on the tables that, uh, have the data, and I will only pull the required fields that are necessary to be populated in the business report. Uh, this is how I handle data extraction by, uh, uh, collating from data from different table sources and pulling out into the report, so I can, uh, do it with a SQL, uh, with an SQL query by doing a join. And also I can do it with ETL by joining 2 forces to flat file, which is a data which is a file source for the report.

So designing a SQL database for an application that needs to handle large scale data, uh, which needs high readwrite for options is first, I would allow concurrency control, uh, to be limited. So, basically, multiple readout options are are going to be there. If there are going to be multiple operations as such, then I would allow the concurrency for multiple sessions, uh, by the same or different users to perform the read, write operations at the same time, uh, simultaneously without any stoppage or a blocker in the way. So and I would also, uh, design highly scalable database by, uh, by taking the offerings, which would help, uh, to perform the queries faster, and I would, I would incorporate Turing queries while creating the tables itself and try to optimize the runtime by writing efficient scripts for creating the tables and objects within a database.

Some of the issues that, uh, we might, uh, encounter while designing and implementing SQL database is to understand the scalability of the database, how much, uh, memory the database could use, and how many users or sessions can the database handle. So these all things come in mind when these all issues might arise when, uh, we start designing and implementing a SQL database. And I would resolve the issues by creating the, uh, database objects that are, uh, that are basically efficient by trying to utilize the memory with less, uh, memory as possible and try not and be cost effective while trying to run the SQL queries. So this is, uh, this can be happened by creating the tables and database objects and other database objects such as, uh, stored procedures, functions, and all, uh, with efficient scripts, actually.

So we can optimize the query by doing a select count. Instead of star, we can give the select count of a particular single field, which will equate to the same, uh, count that we are really looking for. So we can do a select count of, uh, field name from customers where country for US entity equal to New York. So by doing this, by eliminating the start, so we try to pull less records, which will give the same count as that which we get, uh, while giving a count star, or we can use a count of 1. That will solve the problem.

Can be made. So doing a select start from orders where order ID equal to 1 to 3. So we can We can try to pull the particular fields instead of giving a start, which would, uh, help improving the snippet and thereby, uh,

So the process which we would take to migrate a database from 1 to another is to, uh, do a cyclic, uh, check by transferring, uh, the database from one to another. Uh, so basically, we employ pluggable database, which will, uh, ensure to transfer data from a lot. Uh, since it's a large SQL database, we can transfer data from 1 to another using a pluggable database, which will ensure that the downtime is minimal since the Uh, original database from which we are transferring will still be up, and there'll be no, uh, no downtime or no restart of the services. Uh, so using a pluggable database can solve the problem.

So in the event of a SQL database fails, uh, the steps that we would take to restore the system by minimizing the data and data loss in downtime is to take backups while ensuring we do the data transfer beforehand. So, basically, when we, uh, deploy a pluggable database, uh, we simultaneously take a full backup, uh, on the, um, on the backup drive, which would, uh, help us to restore the data in case of any unexpected failures, uh, and it will minimize the data loss and downtime.