so good afternoon and welcome to this webinar called ultra-fast self-service bi with tableau 8 and acting in vector wise my name is Jason leaders and I’m the vice president for Act II and asia-pacific my co-presenter today is Ellen Eldridge who is a senior sales consultant from tableau our agenda today is as follows firstly we will talk a little bit about acting and tableau and what makes us different from other solutions we will then discuss the bi journey and what are some of the challenges faced by organizations implementing bi then we’re going to show you exactly how vectorizing tableau 8 work together to streamline bi projects and save you a lot of time and money Allen will take us through a live demonstration connecting to a vectorized database and showing us just how easy it is to bring your data to life using tableau 8 will then open up the webinar for some Q&A so who are tableau and act in tableau helps people see and understand their data their products transform the way people use data to solve problems by delivering rapid speed to answer ease of use and user self-service all using visual best practice you’ll learn a lot more about tableau once you see a l’instant demonstration acting’s mission is to take action on big data we are focused on delivering enterprise-grade solutions for data integration and data management with over 10,000 customers globally we have operations across all continents servicing our customers today we’ll be focusing on our record-breaking vectorized database it enables organizations to solve performance issues deploy bi projects faster and spend dramatically less on delivering bi as you can see from the graph in front of you vectorize the blue part of the line has completely smashed the database performance benchmark since 2010 with the greatest margins ever recorded in the benchmarks history so let’s look at the BI journey and where we have come from in recent history static reporting was often good enough many organizations were simply content with monthly or perhaps weekly reporting that analysis was typically internally focused and gave us a picture of what happened in the past the information was usually pre-prepared by IT and deploy to business users and analysts using simple report generators to view this business data dynamically was considered a nice to have as organizations were heavily focused on a operational efficiency and standardizing on business processes however in recent years globalization constant economic pressures and rapidly changing business landscape has brought about a radical rethink garner calls this the 3 v’s the first is variety no longer is our structured controlled internal corporate data good enough businesses are demanding more data is available to us everywhere it’s inside emails documents and comes from social media feeds and external data providers the second is velocity businesses now need to react far more responsively to rapidly changing business climates they need to move from monthly and weekly reporting to daily hourly or even real-time analysis and the third is volume we all know that data is growing exponentially we’ve seen all the hype around big data and organizations needing to develop strategies to deal with this fast emerging challenge as a result businesses need to change their approach to VI no longer is it possible to generate business value simply by focusing on operational efficiency having a flexible and dynamic analytical capability is a business imperative that fosters innovation and competitive differentiation so how does conventional wisdom around bi and big data cope with his new paradigm well a typical approach to data warehousing and B is to build a consolidated view of corporate data in a single repository the primary aim of this approach was to provide a single consolidated view of the business however it always required considerable design and effort from very specialized IT professionals further adding to the complexity was a need to develop data analysis and performance capabilities

these schemers indexes aggregates and OLAP cubes were designed and built just to support reporting and bi functions so what about agility and the ability to change unfortunately with this level of analysis design and build required these pio projects are dreadfully slow and very time-consuming they require large teams of IT professionals and take months if not years to complete and as with all project work designs are rigid and inflexible so what was the impact of this in flexibility well the data warehouse Institute bi benchmark report shows us that the report that a report or a dashboard can take between six weeks in 2008 to seven weeks in 2011 to develop even more alarmingly thirty-three percent of organizations take over three months to add a new data source this is hardly flexible and agile bi so why he is building reports and dashboards quickly such a challenge at the heart of the current bi paddock paradigm is the need to build aggregated data sets to summarize and collect data to provide adequate VI performance the issue with these aggregations is that they are hard work they are difficult to design require expensive IT resources to build and manage and are very inflexible to business change most of all they take up wage too much time and money to design build and manage the 3 v’s variety velocity and volume have placed unprecedented strains on existing infrastructure further compounding the limitations of this aggregation approach so obviously agility is key agility in the form of being able to rapidly respond to business demands without the need for an expensive and time-consuming IT project so what so what if rather than trying to preempt the business questions that may get asked and then building a set of rigid reports and aggregated data sets we were able to answer these questions on the fly empowering your users to get their own answers without the need for IT involvement and eliminating the need for the six to seven weeks of project work to deliver the answers we believe this is possible with the combination of tableau and vector wise a clean fresh approach to be I tableau provides fast intuitive analysis capabilities that allow end-users to freely interrogate data it efficiently delivers information to the end user in a way that they can understand the data and how best to use it importantly it allows them to freely select and interrogate the data without having to understand the underlying data structures it’s powerful visualization capability allows users to format the data in a myriad of graphical and tabular formats however all this analysis flexibility and power is only as effective as the underlying database vector wise is the magic ingredient that eliminates the need for IT expertise to deliver fast query performance it effortlessly plows through millions and millions of rows of data without the need for expensive and time-consuming indexes aggregates and OLAP cubes eliminating this performance bottleneck allows users with functionally rich bi tools like blow to freely analyze and interrogate any and all data without restriction so just how fast is vector lies vector wise is all about delivering record-breaking query performance just to give you some context on how well it does is on the screen in front of you are the results for the one terabyte t PCH benchmark what you see here is the staggering performance margin that vectorize achieves over its competitors with 445,000 queries per hour it was two and a half times faster than its nearest competitor but most importantly vectorize broke these records using dramatically less hardware the hardware costs are vectorized were a fraction of the cost of the other benchmarks the next fastest was almost eight times more expensive at four hundred and sixty thousand dollars and then you can see Oracle’s best effort came in at a

staggering 2.4 million dollars so what are the benefits you gain from using a fast database firstly you slash the cost of hardware in many cases we see vectorize consistently outperforming other databases on very small form factors when compared to much larger racks of servers on the screen in front of you you will see the hardware used from the one terabyte TB TCH benchmark that small dell server you see there outperformed the large Oracle RAC server that you see by a factor of two and a half times secondly you can achieve dramatic project time and cost savings taking advantage of this speed means you reduce the need for lengthy IT projects to implement cubes and aggregates or to continually tune the underlying databases for performance wouldn’t we much prefer to be educating our users on how to get the most out of tableaus functional richness rather than how many more aggregates or cues we need to build to get reporting so how in fact does this revolutionary performance impact the way we deliver bi today vector wise delivers a new paradigm for usable and agile old bi organizations can feed their businesses uses growing appetite for five stanzas by allowing them to access the data as it lies today in your data ecosystem no matter what the source systems are data can be fed directly into vectorized for analysis its outstanding performance allows users to interrogate data without the need to reformat or pre aggregate the data this capability coupled with tableau 8 mean that you can deliver unmatched flexibility and agility to your bi customers saving you valuable time and money let’s get Alan to show us how over to you Ellen great thanks Jason okay so what I’d like to go through in my part of the presentation is actually a live demonstration so we have three things that will be working with today I have a copy of tableau desktop installed on my local pc and i also have a copy of tableau server running locally on this pc and we’ll be using these as part of the demonstration and then we have our vectorized server which is actually running over in the in the US so what I’d like to do is fire up tableau and let’s connect through our vector wise environment so we’re going to connect in here and we’ll make a direct connection so there’s nothing that’s been predefined and this will be one of the I guess the key messages as we go through this demonstration the real power of vector wise is that it allows us to ask any question on the fly because there’s no need for aggregates or indexes or constructing an OLAP cube I as an end user I’m not constrained by any of these constructs in the questions that i can ask i’m free to ask whatever i’d like so you can see i have a wide variety of different data sources that I support and we’re going to jump in using our vector wise driver I’ll connect to this server all I have to do is provide the details for how to connect I pick the database that I want and you’ll see here that we have a number of different tables in here this is actually a small snowflake schema when I say small it’s actually about a hundred gigabytes of data and we have here just a single table for the purposes of this demonstration where we pre joined these tables together to rust just rather than having to build this construct from scratch so I’m going to connect directly to this table and at this point here you’ll see tableau offers as two choices we can either make a live connection which means we don’t move data anywhere we simply establish a communication path to the database and as we work in tableau we send queries to the database and it sends us back answers this is the powerful mechanism that we can leverage when we connect to vectorize because it’s able to answer these queries so quickly it’s enables us to have a real time exploratory experience working with our data many other tools might need us to import this data into memory before we get this kind of experience here tableau still offices that option here and we might choose to use this approach when we’re working with platform like vector wise if we wanted to create a portable set of data so if we wanted to take some data offline then we might do

that but if we just wanted to start analyzing our data all we have to do is make a live connection so I do that and I’m immediately connected to this database and we can immediately start querying it so I can come down here and just look at the number of records in under a second you can see here that it’s responded back and in this particular database there’s just a little under a billion records of data so this is quite a chunky amount of data that we’re working with and we’re going to be querying this live again I’m based in Melbourne and the server that we’re querying is based in the United States so we’re also seeing an efficient utilization of the bandwidth between these two locations as well so I’ve connected to this table it’s not particularly clean but you can see the tableaus started to make some understanding of the data that’s contained in this table it’s divided our fields in two dimensions which are categorical data elements these are the things that we want to group by or drill up and drill down through and our measures down here which are our quantitative data elements these are typically the things that we want to aggregate in some way we want to sum them or average them now it’s not perfect you can see here that there’s a number of these fields down here that have been miscategorized so we’ve got some pre-generated time fields we’ve got some ID fields down here and we’ve got some geography data some latitude and longitude for our stores in here these are all actually dimensions so it’s very simple for me to just take this select those fields and then drag them up and place them into the dimension bucket at the top up here now there’s a bunch of fields in here that we actually don’t really need for this part of the evaluation so any of these ID fields these date fields and you’ll see why we don’t need these a little bit further on in the demonstration but these are I guess constructs that have been built for use with other bi tools we really just need the the base date of the order date down here so we’ll get to play with that one so I’m going to remove all of these ID fields through here and these date fields in here we don’t need those so let’s get them out of the picture I just have to right-click and say hide so we can start to clean up any of these fields that we don’t necessarily need for the purposes of our demonstration right so now we’re ready to start asking questions and as I said before I’m not constrained in the questions that I can ask all I’m going to do is bring fields from the data panel over here into our workspace over here that will create a query which gets passed to vectorize and it will do the hard work and send me back the answers that I want so if I started with my line price down here in fact what we might do is we know that these three fields down here are actually currency field so I can just start to add some additional metadata into the way that this data should be handled now if I just double click on my price line down here you can see that we’ve got a little over 130 billion dollars of total transactions and it’s doing that calculation for us on the fly there are as I said before no aggregates that have been created so we can easily start to drill down on this information using our dimensions and start to break it down in different ways now there’s a couple of ways that we can bring our fields out into this workspace and one of the things that tableau is extremely powerful at is helping end users who are not visualization experts or data analysis experts to create meaningful visualizations as we go along and there’s a couple of ways that it does that at the heart of it is a feature of tableau called show me the best way to think of this is it’s a visual recommendation engine based on the data that is in the visual work space it will recommend an appropriate type of visualization so you can see here that it’s currently recommending a bar chart which is created as the appropriate type of visualization then if I wanted to continue to break this down so let’s bring our products family in here you can see a number of these other visualizations light up but it’s still recommending a bar chart so if I click on that you’ll see that it’s now got my three product families non consumable food and drink we can easily sort this data but you can see nann consumables are the significant contributor to our total revenue in here now we have information about our products stored over here in our data panel in our dimensions it would be really nice for me to be able to drill up and down through what is obviously a hierarchy we’ve got family category subcategory and product name now I can easily just bring these out so I could bring my product category and drag it and drop it into the workspace and we’ll send

the query again sought this information these are live queries again I have to emphasize that as we go through this presentation there’s no caching that’s that’s been done there’s no warming that’s been done to localize the data onto my pc this is just live queries going out to the vector wise database but you can see the response times of near real time I’m able to interact with the data very smoothly now I might want to create this hierarchy over here so in our data over here it’s simply a matter of grabbing the fields and dragging them one on top of the other so I’ll drag my category onto my product family this is my this is my product hierarchy in here so we’ll create that hierarchy and then I just need to bring the other fields in as necessary so the four fields so now I can start to drill up and drill down through these hierarchies over here now it might be good to look at other things you can see the length of the bar at the moment is currently representing the sum of the line price in here but we might also want to see if there are correlations related to profitability now I’ve got my margin field down here that’s my profit field in fact it might even be good for me to come in here and just rename some of these so this is actually my sales field my margin field we can rename that let’s call that profit so we can make these much more readable for an end user and we can save these changes we can save this metadata and share it with other users so this is really a one-time effort of starting to clean up the data in here so I could bring my profit out here but doing that in a raw form is probably not going to be all that useful because we’ve made a lot of sales and obviously we’ll have a large amount of profit would probably make more sense for us to normalize this and this is where we can easily create calculated fields that allow us to extend our understanding of our base data so what I might like to do is do a simple calculation in here which is my the profit percentage in here so it’s simply a matter of some of profit / my some of if I can click this right sales and we’ll call that profit percent in here click on ok and again we can just tell it that when we work with this particular number over here it’s going to be a percent so now I can easily just bring this out into the workspace and I’m just dragging this on to the workspace and I’m holding down the mouse button at the moment again this is where tableaus visualization engine comes into play because it’s now saying you’re not explicitly telling me what to do with this data but if you just drop it onto the visualization will make some decisions for you so if I drop that here it goes off it execute the query and now you can see that it’s it’s using the profit percentage value which is we’ve calculated on the fly to color these bars so if I hover over this you can see that we have a very strong forty-eight percent profitability for our electronics but down here for our health foods our profitability is now down to about twenty-five percent so very quick and easy for us to create these interactive experiences again I could take some other data over here so I’ve got my customer region field for example I could drag this out onto the visualization and if I drop it across the top here it’s telling it that I would like a column for each one of these regions and now what we can see is for each of these regions the data is very heavily biased towards North America over here but if we wanted to drill in we could double-click and just start to look at the apac data so double-click is the indicator to focus in again for our geography we’ve got customer region if I’ve got my customer country I can drag that on top of here and we can make our geography hierarchy so again we can start to easily drill down to our customers city will bring that one in here as well so now I can drill down from the customer region and if I just click on the expand sign you can see that we’re drilling into look at the individual countries within the asia-pacific region so it’s very quick and very easy for me to start to create these visualizations all right so let’s just call that our product of you and as you can see the other thing that tableau is doing along the way is its its first way of representing data is to display it as a to display the information as a chart however we can easily at any stage come across here and just change this to say well we’d actually like to see this information as a crosstab so we can always I’m drilling to change the view from a graphical presentation to a crosstab representation if we wanted to do that and then it still retains its

interactivity so I can start to drill up and drill down exactly the same way as I could on the chart previously so this is very similar to a pivot table kind of experience so there’s my crosstab now we’ve seen that there are some fields in here that we’re working with in our data set that have spatial context so our customer region and our customer country field for example now in the underlying database there’s nothing special there’s been no geocoding that’s been done on the customer country dimension it’s just a text field that happens to have the names of countries in here but tableau has automatically recognized that this has spatial context you can see that by the little globe icon that’s being shown over here so if I just double click on that you can see that it now takes that data and automatically displays it on a map for me and the tableau engine has a built-in geocoding capability that says when we see the text field called Australia or the text field called India and it knows that those locations correspond to a location on the map and therefore we can plot this information and because we’ve created this hierarchy in here it’s easy for me to well maybe before we drill down let’s grab our sales over here drop it again using the show me feature to choose the size of the mark in here and in fact maybe what we need to do is just make these marks a little bit bigger so we can see what’s going on but maybe we’ll drill down at this point here and say I want to go down from the country level down to the city level now this is a query that takes a bit longer to run because it’s actually having to grab back a much more detailed data set so there’s some network latency that we’re dealing with in here so here are all of the cities that have been reported now this is a much busier chart so maybe we can clean this up a little bit we’ll add some transparency and we’ll put a border on our mark so it can really start to see this over plotting maybe we’ll grab our profit information and put that on the color so we can easily start to then correlate the sales as well as the prophet information together on the one chart you’ll also note that there is some unknown data down here so at this stage here we could either come in and we could correct these unknown locations or we could just filter out the stuff that we think is not relevant in here right now I can use that customer region over here and what I’d like to do is just add a quick filter which allows me to very easily focus in on these regions so now I could turn this into just a single value selector and if I just focus on the anz area you’ll see that the chart now updates automatically whenever we change this quick filter so if we’re looking at Australia New Zealand or if we’re looking at oh well before we do that let’s make our chart a little easier to understand we’ll take the city name and we’ll stick that on the label and maybe it makes sense for us to take the sales and put that on the label as well so as we can easily see what what’s going on if I hover I get a little pop-up tooltip that tells me the information but I can now start to flip around across the various customer regions here if I want to be able to zoom in because clearly there’s a very dense set of data here then I could just very easily use the map tools to drill in to be able to see what’s going on or even perhaps just double click on the map to be able to drilling but everything’s dynamic and you’ll see tableau makes a good job of presenting the information in a readable way all right so there’s my map move on and let’s create another chart this time I’m going to again start with my sales data in here and we’ve already seen what happens when we start to break it down by what we call a nominal data set something that is just categorical in nature like product family or category in here but we can also start to work with temporal data so I could take this order date for example and just drag it out and drop it here and you’ll see the tableau is automatically built out a whole bunch of time awareness of time intelligence in here so we can start to look at our information by date part we can look at our information continuously so I’m going to plot this information at the month level in here again a relatively chunky query that we’re having to execute it’s got to bring back a reasonable amount of data so it’s off asking that question of the database but here’s our result back in a couple of seconds and we can see our data actually has a relatively cyclic shape to it we could change the way that we’re looking at this information so at the moment we’re looking at it just by the sum of sales but we could easily change this to any other kind of aggregation function I

might like to actually duplicate this data set so we’ll have two coffees of it and this time rather than just looking at the raw information what I’d like to do is bring in a more complex type of calculation we call these things table calyx and where I could perhaps do a running total so you can see that we now get a running total of the the sales across the same time period here it’s actually quite a difficult expression to do in sequel but bringing into tableau we can do this kind of thing in just a couple of clicks so what I could now do on that running total is we can say well let’s break that running total down by the product family and will actually change that to be an area chart so we get more of a stacked view and we can see what’s going on at that level down here on my sales up here what I might like to do is change this from a line chart to a bar chart so I’m starting to take a little bit more control over how the data is being presented so here’s my bar chart and now what I’d like to do is combine this with some additional data one of the points that Jason made in his introduction was that it’s not uncommon when you have I guess a traditional approach to how you deal with your data or how you deal with your business intelligence tools that it takes a long time to introduce new data there’s no agility in there but with tableau it makes it very easy for me to combine this data that we’re pulling back live from vector wise with some additional data that I’ve just got sitting over here in an Excel spreadsheet so what I’m going to do is bring my Excel spreadsheet into my workbook we’ll just grab my budget data again we’ll just do a live connection and what you can see is that tableau automatically recognizes that there is commonality between these two data sets these are linking fields in here so now I can just grab that budget data and bring it into the visualization and will change the mark type over here to be against and will change its color so that it stands out it’s black and we’ll make sure that we synchronize our axes and will free up a bit more space and what we’ve just done in a few clicks here is created a chart that combines information being brought back from our vector wise database in the States and combine that done a data match up on the fly we started it’s coming from an Excel spreadsheet that’s just sitting locally on my desktop on my PC that’s incredibly powerful for a lot of data analysts if you think of the kind of what-if scenario problems that you might like to tackle or even just combining well-governed data that’s coming out of our structured data environment with information that we’ve just pulled together on the fly it’s an incredibly powerful capability we call this in in tableau blending so we’ve combined data from multiple data sources in here and the last thing that I might like to do before I move along here is to just take my order date over here will create again a filter on this so we can filter this based on the year and we’ll choose all of the years at this stage and then I’ll just surface that as a quick filter object over here so we can change that on the fly right the last thing that I’d like to do now is to pull all of these things together and create a single view that shows all of these elements on one page we call this thing a dashboard so I’ve got a dashboard and I’m just going to adjust this so that it has an automatic sizing for my screen and we can easily start to bring this information in so i’ll get my list of products I’ll bring my map and you can see that as I just drag these objects out onto the canvas tableau allows me to easily place them side-by-side with each other maybe I’ll drop this one down here at the bottom so here’s my blending view he is my map here’s my list of my products across the side over here we can freeform lay out some of this stuff so if I wanted to take this element over here just holding down the shift key we can drag this and place that easily onto the chart so we can free up some space so we have this concept of free form layout of dashboards where we can drag and drop objects I’ll remove some of these we don’t necessarily need them and now we’ve we’ve started to get this dashboard that shows a combined view again its interactive so I can use these filters to change things maybe I’ll change my time filter to be a slider so it makes it easier for me to flip from year to year across here but you’ll notice that these are attached to just single objects this is only affecting the map at the moment if I wanted this to affect everything over here then it’s very easy for me to come along here and say I would like this to apply to all of the worksheets that are driving off this same set of data so now it’s easy for me

to be able to flip backwards and forwards across these and you’ll see that all three objects are now changing in response to this filter similarly with my time filter down here I could change this to apply to all of these objects as well so we can create these filters very easily to be dynamic across these data sources in here I can also use these objects as filters against one another so if I wanted to let’s use my map as a filter or maybe even use my product over here as a filter then as I click on these objects and maybe let’s look at my sports data down here when I click on that it’s going to actually filter these other two objects to just show the data for sports and then if I just wanted to drill in on New Zealand I can circle around that and then I’ll drilling to just show the data for museum in fact it’s actually now the intersection of those two elements so we have this ability to create these dashboards with these associative filters sort of go running across each of the objects in here right so I’m relatively happy with the the resulting dashboard that I’ve created the last thing that I’d like to do is to now take this and I’d like to start sharing it with other people so let me just give this a known or just called a dashboard and we’ll publish this particular workbook up onto my server so we’ll publish this workbook as i said i have a local server that’s running down here so I’m just going to log into my local server it’s secured so it’s going to we can restrict who has the ability to publish content up onto this particular server so we’ll drop it up here will call this VW demo one and I can control who can view this I can control which elements I’m publishing up onto the environment maybe we’ll just publish these two tabs here and whether I need to send any external files i’m going to leave this checked on because i have that spreadsheet sitting on my desktop obviously when we publish this to a server you’d want to be able to push that spreadsheet file across with the data set I’m just going to click publish and what is now doing is pushing this workbook up onto my tableau server so it has to do a couple of queries in here just to understand the structure of the data into the app that creates and some thumbnail views but what that’s now done is published it to my server if I open this now in a browser window you’ll see that it’s popped up my web browser I’m using google chrome and i’m going to log in using those same credentials now i can see those two views that we published the product view on the dashboard view if i click on the dashboard view but we’re now looking at is that exact same workbook that i had previously and it’s now been made available to a multitude of users if I want to just through a web browser interface takes a few seconds to render it the first time because it actually has to go off and pull those map tiles back to be able to draw the map for me and it has to run those those queries the first time but there it is it’s brought that information back and now I can start to just interact with this same as I could previously so dragging across the objects and and creating filters works if i change my year and slice from from year to year this interactivity is all available to me through a web browser experience as well so it continues to be a rich experience for my end users rather than just a static view of this data being presented back to them it also now becomes a very dynamic piece of content which I can reuse across a number of other platforms so I can take this I could once it’s been published up into tableau server it’s available on mobile devices like iPads and androids it’s also available in a variety of different formats so I can come up here and either just copy from here or i can grab this share button if i wanted to then send this particular workbook to somebody else just copy that link and then it’s just a matter of emailing to somebody they’d get that link and if they now put that into their address bar over here they jump straight in and see the dashboard as it was when i copied and sent that particular URL to them we can also take this information and we could start to say well I’d like to actually take this particular workbook and snapshot this workbook as a PDF file so just by typing PDF on the URL I’m actually telling tableau to automatically generate this particular workbook as a PDF file so I could then email that off to other people as well so in a just a matter of minutes we’ve

we’ve gone through the process of connecting to a vectorized database done a series of analytic views and all the way along here again I just need to re-emphasize that that’s been a live connection against quite a large volume of data if we needed to import that data into an in-memory model we’d probably still be sitting here waiting for that too we’d still be sitting waiting for that to complete in this case here we’ve actually been able to create this dashboard create interactivity across the components and then we’ve been able to publish it up and make it available through web browsers and make it available to mobile devices and a bunch of other things that we can do just by taking those links and controlling them so at that point Jason I’d like to hand back to you and you can close us out thanks for that Ellen that’s a great demonstration I think it showed remarkable flexibility and ease of use especially on such a large data set just fio folks that but data set is just on a billion rows of data so like never before vectorizing tableau allow you to empower your end-users give your users what they want when they want it in a simple easy cost-effective way of exploring data to create reports in minutes and we saw Alan do that very effectively secondly it’s simple to implement and you avoid the dreaded crippling IT projects to deliver outstanding value to your users at a fraction of the cost so finally now we can focus on the business questions and not the underlying IT those of you interested in trying out vectorizing tableau 8 fear cells please go to either tableau software com or acting calm / vector wise to download either product alternatively if you require any further information please don’t hesitate to contact us on the email addresses on the screen there and you’re also very free to contact either myself or Allen directly so that concludes our presentation for today I’d like now to open up the session for your questions so please feel free to enter your questions into the chat window and we’ll do our best to respond just give you a couple of seconds to get the questions through so the question come through here Alan I think it’s best for you to answer does tableau support of mobile devices like iPads and tablets and yes it does as I said in the presentation once we take a workbook and we publish it into tableau server the way the tableau server works and as it was working when we were viewing it in the web browser and is it renders the workbook using pure HTML so there’s no active content there’s no silver light or flash or java that you require so the visualizations that tableau server creates when it delivers them to the end-users work across any platform that can render HTML so your desktop browsers or any mobile devices going beyond just rendering it onto the mobile device we also recognize that the requests coming from a mobile device are coming from a touch oriented device so for the interactivity controls things like the quick filter objects or being able to zoom in and zoom out on a map for example will move across and start to recognize gestures so you can pinch to zoom or you can swipe to be able to scroll or on the quick filter objects will replace the ones designed for a keyboard and mouse with ones designed for being manipulated through a touch device the key point here I guess is just that the experience is tuned for a mobile device but you don’t need to republish or alter the workbook in any way to get it to work across those platforms thanks Alan this one is so I guess one for me what makes a vector wise different to be faster than other databases like column and MPP look I guess there’s two parts of this answers the tech cool reason and and then what does that deliver in terms of its outcomes vectorize is very unique in the way it leverages modern chip architectures and

it means that it’s able to compute far greater amounts of data than anything else in the market and it’s very unique in the way it does it so what that means for users and VI and consumers of VI is that they are able to implement databases far more simply it means that you can just take data out as it lies from your source system without having to redesign the schema specifically for reporting or to sit there and tune the database to get any sort of performance out of it what vectorize enables you to do is just drop the table into vectorize and report on its straight away without having to build for example a MPP cluster or anything like that so with very very small modest hardware you know the sort of dell servers in HP and IBM Intel based service there’s no specialized hardware there’s no you know special chipset you require or special backbone or anything like that it’s it’s it’s beauties and it’s elegance and it’s able to simply just find a server somewhere load it with your linux or windows and just run it and it does it so cheaply and so effortlessly is what makes it very attractive we find a lot of people doing proof of concepts are overwhelmed by how simple in fact it is as a couple more coming through the chat window what kind of data structures does it have does it look like a dimensional architecture does it like a SAS scheme at data well yeah that’s actually a very interesting question so the answer is it is a relational database model underneath it it’s for all intensive purposes vectorized looks like any other relational database user state of SQL the SQL you run against Oracle sequel server or ingress is exactly to what you will run against effective the data structures underneath are whatever you want like any relational database you can have it normalized or denormalized it can be schema star schema or snowflake schema what do we recommend we don’t recommend anything it will happily process data at an astonishing rate that you will not need to worry about how this schema is designed underneath don’t take my word for it try it for yourselves we’ve had a number of customers GE power and water is a good example who came to us asking for recommendations on how they should structure their data they had a couple of terabytes of data in the end we said we don’t have a preference try it for yourselves they tried three or four different ways and in the end the overhead of doing the transformation just wasn’t worth any performance the the minor performance gains that they got from it so so the answer is it can look like any schemer you like next question does vector I support sorry Jason just before you just probably move on I guess I’ll just parallel the same response from tableau as well that from the point of view of analyzing the data the structure of the schemer from our point of view again is irrelevant so whether it’s flat stars snowflake third normal we’re happy to work with any of those and again we just construct the appropriate sequel which we pass to the database okay there you go look there’s a there’s a few questions and the technology look I’m happy to do more about the technical difference it is quite a it’s a presentation in its own right it’s quite a fundamental step change in the way works underneath the covers the end result though is that it just behaves like any other database any other relational database it’s identical in its behavior all it does is does it dramatically quicker there’s no as I said it’s standard SQL anti sequel standard there’s no magic in the way you need to program it in the way you need to set it’s delightfully simple for those of you who want more in-depth technical information are more than happy to take you through it ETL processes are fully supported we we support all the main ETL tools obviously there are circumstances where you know different heterogeneous systems need to communicate we appreciate that but again there’s nothing there’s nothing there that as I said it weighs like any other relational database if you want to use each of your tools and do transformations and knock yourself out data volumes there’s a

question here around how small and how big look typically we find anywhere from a hundred megabytes of raw system data up until about 3 30 terabytes of raw system data is all very doable from a vectorized point of view it’s quite flexible in its pricing it can be priced either by the processor cores or using data pricing in the form of terabyte data pricing if you want to know about specific pricing for you please feel free to drop me an email and I’ll put you in touch with your with the right person the other question here can we get data out of sources such as who do the answer is yes we have a very clever Hadoop connector that is able to run the MapReduce jobs across the the the Hadoop cluster and consolidate the output data and load it very very quickly into vectorize we believe it’s the fastest Hadoop loader in the market today it can typically do about two and a half terabytes an hour which is on a very very modest sub $15,000 server which is which is also very nice as well so again very cost-effective and certainly supports our price performance leadership position another question is here tableau supports iphone and ipad i think that’s correct right Ellen you said that’s correct that’s correct will support both of those platforms and we have a native app that’s available for ipad on iphone you would just use safari to be able to browse on the the workbooks that the only thing to bear in mind when you’re delivering content down to a smartphone is that the screen real estate is significantly constrained so you just need to take that into account when you’re designing a dashboard that you intend to deliver to a to a smartphone platform otherwise you just find yourself scrolling around a lot it’ll still work it just is obviously a lot of scrolling and zooming to be able to see things one more here scale in terms of users for tableau it scales too many thousands of users we have customers who are using it in environments where they’re delivering dashboards to tens of thousands of employees or external users and of course it’s built on the same technology as our tableau public which is our cloud-based service and that has hundreds of thousands of users accessing that okay and one more for you does tableau reports do tableau reports published to the web run on need basis refresh i’m assuming of a static or dynamic so the approach that we showed today is a live connection to the database so any time I come along as an end user and I view a workbook I’m viewing the data that is currently stored in the underlying database now there’s some caching layers in between that we can tune so I might look at a workbook and then if i look at it five minutes later i might still be inside that casing window so i look at the information that’s in the case unless i explicitly override that and tell it no i do want it to go back and refresh the data so the concept of scheduling a report which i guess is a central concept in traditional bi architectures and that kind of concept doesn’t really exist inside tableau if you’re using the extract model which I talked about earlier if you wanted to bring the data into our in memory architecture then the concept of refreshing the extract makes sense and scheduling that periodically make sense but when we’re working with a live connection against our high-performance database like vector wise then anytime I look at that particular workbook I’ll be looking at the data that’s in the underlying database great thanks all right I think we’ll do one last question we’re getting one minute to the hour does vectorize run on Windows and Linux the answer is yes either flavor of the mainstream Linux products or Windows is fine the only thing you need to do is run an x86 chip and you can buy that hardware from anybody whether it be Dell HP IBM or you know a white box supplier in Taiwan suit yourself it’s part of its attraction attractive and price

performances of fact that you are not tied into a hardware and operating system vendor all right well thank you everyone thank you and we really appreciate you making the effort to join the they’d start we are very grateful for that and apologize profusely again for the issues that we had of course if there are any questions please do not hesitate to contact either organizations and please feel free to get on to either of our websites for the trial downloads and to find out any more information that you might need thank you again and thank you Ellen for your great demonstration and we’ll see you all next time