Shakti Kumar

Yeah. Hello. My name is Shakti Kumar. Thank you so much for providing me this opportunity. So, a brief introduction about myself would be like, I'm a computer science engineer by education, and currently, I'm working as a senior business analyst at Dell Technologies. I'm associated with Dell Technologies from almost 9 and a half years. And, as a part of the senior member in the BI team of Dell Technologies, my prime role and responsibility is to drive sales of Dell NCP products across the globe, driving and designing different strategies about sales, how to increase sales across the globe, collaborating with stakeholders, project managers, understanding their requirements, getting in touch with report users, such as senior leadership, and designing sales dashboards as per the requirements, defining and designing new KPIs, which is important for the business to measure multiple campaigns running across the globe. Apart from this, as a senior member of the team, I manage our junior colleagues, and my prime role and responsibility is to drive them across sales campaigns, make them understand the vision and mission of our work, and how it improves sales of building and securing products across the globe. So that's about myself. Thank you so much.

So there are different methods that I will use to improve query performance in a SQL database. First, I'll identify the exact issue causing the performance problem in our database and report. Then, I'll use a performance analyzer to determine which visualizations, such as graphs, charts, or any other visualization, are creating a performance issue. Alternatively, I can use query folding techniques in Power BI to optimize queries. If necessary, I'll perform calculations, like creating conditional columns, before importing the data into the Power BI dataset to remove unnecessary steps and import the data in the required format for our database. To improve performance, I will remove unnecessary data that is not part of the leadership and is taking up space. These are the techniques I will use to improve SQL performance or query performance in our SQL database.

So, generally, instead of the method, there are basically 2 methods that Power BI provides the user in order to import the data. One is import query, another is direct query. So if I need to enhance the efficiency of data retrieval in Power BI, I will try to use the import query because it is based on historical data and not on live data. So we can do anything, and we can do any kind of modification in the data while loading it into the Power BI dataset. And it has nothing to do with the live data, so it will make the data retrieval faster as compared to using the direct query option. So for all these databases, like Microsoft SQL Server, Postgres SQL, or Oracle. I'm gonna use a direct query from the database in order to retrieve the data to Power BI.

Whatever. So when I'm working with any complex tax expiration, first of all, it depends on the kind of complex city I'm going to use for the tax expiration. It depends on the business requirement, like what is the actual use and what is the actual data that we want to put in our dashboard. So that's how we define the complexity of any DAX expression. And when it comes to communicating with our team when working with any DAX expressions, the key consideration would be to explain why we are using that particular DAX expression and what is the need to use the DAX expression if we are not able to get it directly from our data. So, definitely, we will lose the DAX expression. So, yeah, these are the metrics that I will consider in order to explain to my team members. And apart from that, there are a few more things I would like to inform my team members: whenever we are creating any DAX equation, we will try to make sure there is lesser dependency on any column or not use multiple data to calculate one single DAX expression, which will lead to complexity in the relationship. So these are the things that I will consider when I'm trying to create any DAX expression or when I'm trying to communicate the same to my team members as well when we are working with any kind of complex DAX expressions. Thank you so much.

So, generally, I would like to give this example of which is very much directly linked to my day in and day out work. So whenever we try to provide any kind of security layer or any kind of access control to our Power BI dashboard, So, basically, there are multiple things that we check, like whether we are going to provide the security on the basis of hierarchy, or it's kind of page level security that we are going to provide or this kind of low level security that we are going to provide to our stakeholders. So these are the things or the requirement or the key consideration that we will check with our stakeholders to see what kind of security level they are looking for in order to access the Power BI dashboard. Because sometimes what will happen is the senior leaders will need access to each and every dashboard. But when you are looking for any stakeholders, we don't want to provide them a huge set of data like the overall numbers. It's just program specific, what are the programs they cater to. So these are the things that we will see and we will look into when we are providing any kind of security layers or any kind of security to our Power BI dashboards. And in order to test that application while publishing it to the dashboard, we generally use a Power BI desktop version where we have an option to manage rules, and then we can get access. Then we can define on the basis of users how they're going to use the dashboard - it's based on the read function, write function, or both. So, yeah, these are the things that we will look at in order to provide any kind of security layers to our Power BI dashboard. Thank you.

To ensure the security of the dashboard being viewed by multiple users, we would take the following steps. We would define certain rules for all the users in terms of read, write, or both. We will provide a base level filter, base level view, or the complete dashboard view. These are the things we would consider when providing security onto our Power BI dashboard, especially when it's used by multiple users. When defining any kind of security, we would ensure that each user has a specific level of access. Thank you.

So I guess, in this DAX, basically, what this DAX formula is doing, we are trying to calculate the sales of only 2022 by providing this calculate function into our DAX expression. But I believe for getting this, getting the output for this particular information, what we are looking for, there's no need for that formula because we can provide a filter here on a visual level filter and we can provide that automatically, and it should give us the total sales of 2022. So here, we have an opportunity to save the measure we have created here by using this DAX formula. And it's kind of not required, which will definitely impact performance less if we're not creating it and we're just providing a visual filter for year 2022, I think.

I'm sorry, I'm not able to see that we're trying to pull out the two data. Like, one is from the table. Another one is our Excel file sales data. We are joining that table from the customer ID. And, all the tables are utilizing no. I have no idea. I need to work on this one. Thank you.

First of all, we need to see what is first and foremost, we need to get to the root cause of the performance issue, whether it's the issue is getting impacted by the DATs, what we have created, or there is some other factor, which is creating the performance issue in our Power BI dashboard. And after that, we'll try to build a logic around it, so how to improve the performance of the report. But when I say, if I want to design a Power BI reporting solution that actually reflects complex business logic without compromising on report performance, then I would say as I said earlier, when we are building DAX, we need to be very specific. Like, what are the values we need to calculate, and what are the values which are dependent on them. So, basically, we need to identify the dependent variable. And if that is our dependent variable, we'll try to create a DAX which is dependent on as few as possible. And generally, when we are trying to use any DAX expression, we first see the relationship between the tables. If it is a bidirectional or one-to-many, generally, one-to-many will be fine. We can go ahead with the DAX. But if it is many-to-one or many-to-many, we generally avoid creating a DAX on those kinds of relationships. So these are the criteria or the things that we need to look into when we are creating a DAX or when we are creating any Power BI dashboards, what is the relationship between the data sets, and how we can try to build a dash without depending on much more data sets. So, lesser the dependency, the faster will be the query. Thank you. Thank you so much.

So you can say that if you can put the Power BI onto a Power BI service, then it's not related to and there should not be, like, if it's a database maintenance activity is happening, then we can stop the auto refresh of our Power BI report, whether it is monthly, weekly, depending on the business requirement. But, when the report is on Power BI service, it has no dependency on, it has the dependency on Power BI dataset, but unless Anurag tries to refresh the dataset, it will not create any impact. So if any database maintenance activity is going on and we break the connection from our Power BI service to the Power BI datasets, then I don't think it will have any impact on our Power BI report during the database maintenance activities. Thank you.

I'm not familiar with this kind of situation in my day-to-day work, and definitely, I'll look into it. Okay? I'll look into it. Thank you. Thank you so much.