[MUSIC PLAYING] Hi, I'm Mark Davis, big data technologies lead with Dell. Today I'm going to talk about solving the problems associated with managing big data. So, let's start by going to the Dell XPS One touch screen to map out several reasons why solving big data problems on your own can be very challenging. First off, we have the problem of getting talent.
We have the problem of data scientists who are familiar with Hadoop and who know NoSQL technologies and other technologies that support them. These people are difficult to find. They're very expensive to hire. Once you've got people together, then the next question is, what do you do about hardware?
You have an amazing array of options when it comes to hardware. You have switches, you have SSDs, you have platter-based storage solutions, you have cloud virtualization technologies. This takes time to assemble, and it requires very complex coordination. All of this is very, very difficult to do on your own.
And finally, perhaps the most difficult part is getting the hardware and the data scientists together working on the software that's required. We're talking about Hadoop, we're talking about NoSQL technologies. The software might involve search technologies in addition to virtualization solutions, and text analytics, and other capabilities that all have to run together in a coordinated fashion.
And doing all this on your own is very, very difficult to do. We find people in the industry have been taking six months just to get started. At the end of six months, then they ask the question, how do we solve real business problems. At that point, that's another six months just to spin up trying to solve a specific problem.
And, so the Kitenga Analytic Suite is really designed to help bridge this gap, solve the problem of associated with trying to make sense of big data, and to make it easier to do, less expensive, less costly to do so. So, what does the Kitenga Analytic Suite do? Well, it supports the use of technologies like Hadoop and search in a coordinated fashion.
And to do this, there are various steps that are involved in using the solution. First off, we start by authoring solutions that run across Hadoop. That is, we create content-mining solutions. And when we say content-mining, we're talking about not just traditional RDBMS data. We're talking about unstructured data, semi-structured data, log files, documents, texts, tweets, Facebook postings.
All of the things that are very difficult to process in traditional ways using relational databases. And our solution, then, supports the dragging and dropping of those components together in a very simplified manner so that a solution can be created very rapidly that solves these problems.
We have out of the box components that do things like extract named entities from texts00 people, places, organizations-- that find social network relationships that are present within the data, and that convert all of your data, ultimately, into a searchable form. Once you've created this workflow, then the next step is to run it.
And so then, using Kitenga Analytic Suite, you press the Run button, and the workflow then executes across your cluster, your Hadoop infrastructure, and all of the processes are run in a pipeline fashion, feeding into each other to create your overall data product. While it's running, you can monitor the progress as well.
You can see how the individual nodes are doing, you can see how the overall cluster health is doing, and, when combined with other enterprise technologies, you can create a complete picture of how your system is performing. But that's not the end with Kitenga Analytics Suite.
The final stage is where people-- the data scientists and the data analysts-- get in touch with the data and start to use it to create end products. And so from out of the box, we support the ability to create searchable products and to analyze the data, pulling data directly off of Hadoop and HDFS and putting it into a visualization, a chart, a graph, an information visualization that shows relationships that are present within the data.
Also, the ability to index all that data into a searchable form to create a custom search experience so that you can visualize relationships that are present within the data with embedded graphics directly within the search experience. No more simple ranked list presentation of search results for enterprise search products.
But instead, the ability to have graphs, charts, and information visualization embedded directly within your solution. To learn how you can solve big data problems and transform big data into actionable intelligence, visit the URL you see on the screen. Thanks for watching.