Hi, my name is Amanda Dark Angelo. Welcome to Weaving the fabric of Data. Uh I'm a senior con data consultant at CT I and co-founder of the group Salt City Data Community.Over the next 20 minutes, I'll be going over the basics of data fabric and the modern data platform with some time at the end for questioning, I will also provide my social media at the end. Please feel free to reach out. Ok. So a few years ago, I was working in the public sector and had this amazing opportunity to do analysis that really impacted the community. The goal was to change the responsibility of sidewalk maintenance from homeowners to the city government. This would save money overall and decrease significant stress on homeowners who at the time were required to maintain their sidewalks in front of their homes, either shoveling in the really intense snowy cold climate or replacing those sidewalks when they became damaged. But first, a financial analysis had to be done to understand if the city could afford to take on this responsibility. This is where I came in working through a variety of data. I calculated just how much the city would have to collect in fees to maintain sidewalk support. Eventually I was able to get to the solution. But getting there required connecting to three databases, two API S and adding a number of CS V data sets to the raw data before even starting the analysis.

And this had to be done each and every time any ad hoc analysis was done by leveraging integration and a data fabric that project would have taken days instead of weeks and the decision makers would have way more time to unpack the analysis. The data industry market value is expected to reach $4.5 billion. By the year 2026 40% of organizations in 2018 were planning to automate integration practices and policies. And by 2025 the global data sphere is expected to reach 100 and 81 Z bytes. Data fabric is an idea that an organization can and should be integrating all of that data that's being collected in every aspect of their business with the data industry becoming so integral to all organizations designing a data fabric early and re evaluating continuously will leave you with a product that quietly supports instead of causing more headaches.

Some of the key principles of a data fabric uh be cloud native if possible cloud native is a concept that an organization has never operated with on premise service or products. If you aren't a cloud native organization, be cloud first. This means as you evaluate new products, assess cloud based solutions more favorably than on premise solutions. Cloud products are more easily integrated with a data solution and often they already have their own API S or other connection capabilities built in. Next, you should be engaging your users early on with grassroots data governance, standardizing and making transparent the governance of your solution makes it more accessible and more easily understood by stakeholders. Always consider the right tool, right job principle. We now live in a data marketplace spanning from massive providers like Microsoft to niche projects like D BT. Ensure that along each step of your data solution, you evaluate whether the product directly supports your desired outcomes. At the beginning of your data journey, identify use cases that can be quickly and easily stood up early. A use case based approach decreases time to value and shows stakeholders in a tangible way. The value of the solution low barriers to entry are also key for non technical end users. Your solution should be designed around usability and tested by those end users at various points along the implementation to give notes and become familiar with the tool.

And finally, once your basic data fabric platform is put in place, you can use that platform to support all types of automation, machine learning and artificial intelligence. OK. So the data fabric really consists of 10 basic elements, data architecture, data modeling, data storage and operations, data security, documentation.

And content data warehouse and business intelligence, data integration management, master data management, data quality and reference and metadata. For a significant period of the accelerated data industry movement. These elements have been seen as separate niches that individuals could focus on throughout their careers within the data fabric.

However, that is no longer the case, professionals need to be well versed in all aspects of the industry. Even if in their day to day work, they tend to focus on only one or a few. This gives a holistic view to the data environment for all involved and allows for smoother implementation and a more comprehensive solution overall. With a more comprehensive solution. Though inherently comes more complexity, the best laid plans of mice and men often go awry and that exists in the data sphere as well in any architecture and design of a data fabric. There needs to be leeway to allow for technical complications on boarding of new technologies and resources and the overall acceptance of the platform by the organization at large. A few things to keep in mind during implementation, work time for training into your design. Uh there will need to be a lot of end user training and um and really that is one of the most important aspects of implementation of the solution. So be sure that your users are trained well and that's worked into your timeline. Emphasize to stakeholders that the designed architecture and the model are merely principles and guidelines as opposed to strict requirements and may change, you know, a as you see in the picture here, the reality is very different from the outset plan and finally be clear about scope and use cases at the outset.

You wanna start any data solution project with the end in mind. OK. Uh The data maturity model defines where organizations sit across a continuum of unaware, aware, defined implemented and optimized. These steps almost always need to be followed in order for a successful data solution.

And some solution models can only get you so far in maturity within organizations. These steps will likely look like one you are unaware of the data, you have completely uh two you're aware of the data you have and how it should be used. But most users outside a small team say your data team or your technology team aren't trained on this. So no one outside of that small team is aware, three data responsibilities are assigned to users outside the central data team and rules and policies are com communicated enterprise wide four standardization of the end to end data solutions is understood and enforced. And finally five we get to optimize which is where redundant data sets collection processes, analysis, et cetera are eliminated. And those policies previously created are refined and uh and bulked up a modern data platform takes into account all the ways data can be collected consolidated and used within an organization along the data maturity model that we just went over a data warehouse is an implemented solution at four.

While a modern data platform is an optimized solution at five along the data spectrum, no longer are reports and dashboards, the only outcomes expected of a data solution nor are structured relational data. The only desired inputs. A modern data platform allows for the full spectrum of data to be ingested including streaming audio video, geospatial and as many other data types, you can think of these data can then be understood and used to make decisions more quickly and work more complex techniques including data science and machine learning.

The modern data platform cannot exist functionally without a fully implemented data fabric. Each aspect of the end to end solution supports the increased complexity needed for the modern day uh modern data platform to be sustainable. While the legacy data warehouse style solution was more simplistic and therefore could have a less complex consolidation process. That is not the case for the modern data platform. Gaining insight from the wide casts net of data collection points requires defined layers of collection eltl curation, access and analysis along with a structured governance to ease barriers of entry when using the tool. So why does this matter? Um I'm gonna use an example for my life.

So I'm a knitter and knitting can build beautiful products. They can you can make a sweater blankets, hats, whatever. Uh but it can also be a nightmare. If I miss a stitch or lose track of my yarn. Once I do the entire project can pull apart extremely quickly, I've done all of that work and that is now effectively meaningless. That's the same reason. Each and every one of the elements of a data fabric need to be thoughtfully designed and carefully implemented with one piece missing or a weak link. Your data solution can begin to crumble and the work done to design and implement the tool just completely meaningless. So I want you to walk away from this uh this speaker session. Remembering that a functioning data solution is insightful and intuitive. You should always be keeping your end users in mind even a small break in the process or complication in their ability to use the tool for their needs will entice them to continue to use the lye solution to the data f that the data fabric was intended to replace the use and study of data.

L lies heavily in the ability to communicate the story of the data to non technical end users in an understandable way that allows them to do their jobs better without usability. An implemented data fabric is just a fancy paperweight. It's expensive with little to no function or purpose.

So again, a functioning data solution is insightful and intuitive. OK. Uh I hope you enjoyed my presentation on the data fabric and modern data platform. A final shameless plug my organization CT I is currently looking for all levels of data experience to join us as consultants.

If you're interested, please reach out through linkedin. Um And I'll be sure to get you the necessary information. I'll put my linkedin uh uh link in the chat here. Um And while I do that, I'd love to hear any questions that any of you have about data, the data fabric, modern data platform. Really? Anything that you wanna ask? OK. My, my linkedin is here in the chat. Um If anyone has questions after this presentation, you know, you think of something later, please feel free to reach out. Um Otherwise I thank all of you for joining me and I hope to see you uh throughout the networking sessions later in the conference.