Hi Folks, We want to build a report to show the top 10 or top 5 customers plus the others as a single category. The report also needs to be parameterized by product category, business type, region, etc. The chart type could be a stacked bar chart or also pie charts. Example:[img]http://i623.photobucket.com/albums/tt313/pauldj54/top%205.jpg[/img]The challenge is to generate the categories or subcategories dynamically using DAX.The values cannot be persisted in the dataset, because the needs to react dynamically to either, the ranking number (top 5 or top 10) and the contexts (by product, by business type, region, etc…)The equivalent in MDX would be a new dimension member. Have you faced a similar challenge?Any comment would be appreciated. Kind Regards,
↧
Dynamically generate the categories (dimension members) in a chart using DAX
↧
Need help on SSAS Tabular model
Hi All,I am currently working on SSAS tabular model and I am facing an issue when I slice on excel sheet as below:I have 2 fact tables(Turnover and Space). Both the fact tables have relation with Property table in similar terms. When I slice on Property (BuildingName) as column and add value for measure "Last 12 months Turnover"(which comes from Fact table Turnover), it works fine. Whereas , when I slice on Space(BuildingName) and add value for measure "Last 12 months Turnover"(which comes from Fact table Turnover), it doesn't work. It gives same value(Ex 144, 144, 144,) throughout. There is no relation between the two fact tables. But the client requirement is to create measure for "Last 12 months Turnover" even when we slice on Space.I even tried adding measures to relate the two tables(Facts), but it still doesn't work.I am not sure how to achieve this.Can somebody please help?Please let me know if you need more detailed description of the issue, I can send the screenshots of relations.Regards,Shweta
↧
↧
How to Create New Cube OFf Existing Cube Using Visual Studio
Hello,I Remote to a Windows 12 Server, that has SQL Server 2014 install, Visual 2013, SSDT, and Attunity 3.0 for Teradata. I'm new to this existing platform where there are production cubes and their SSIS packages.I need to copy an existing cube, and make a new one from it, where I plan to make some a number of one off changes. I have already copy/renamed some of the views on the SQL Server that I plan to use in this V2 one-off cube as the data source of Facts and Dimensions. I do not need to alter the source to SQL Server tables at this time.I open VS, go to File, Open, and go to Analysis Services, and give it the proper server/instance/database, but it doesn't connect. I can go to Management Studios and connect to Analysis services, and then see the production cubes, and the one I want to make a copy of to modify. I can in VS open the packages by going File, Open, and Project Solution, and see the solutions and their packages, but this seems incomplete as a "Cube", or a cube to copy.What is the best or correct way for me to copy a cube from production to make a one off version using Visual Studio? Thanks,JPQ
↧
Tabular -> Power BI DirectQuery Performance and Confusion
Hi,I hope somebody can help shed some light on my issue as I am struggling to find an answer.I am using SSAS 2014 SP1 - Tabular and Power BI Desktop/Service. My issue is that of performance when running a power BI report connected to a tabular model vs a direct sql connection. My scenarios are1. I create a proof of concept report using Power BI Desktop, connect directly to sql server using the Direct Query option, build my model in Power BI Desktop (very simple 1 fact table, 4 lookups). Everything runs quite fast - max refresh time so far of 3-4 seconds. Any changes to the report are instant. Everybody is impressed with the 'realtime BI' :-)2. To centralise everything, I then create a tabular model exactly the same in VS 2013, and switch on directquery mode, deploy (getting all excited at this point). Connect my report to the newly deployed model and hit refresh. About 2 minutes later I see the data. Any changes to the report yield the same waiting times.I have other tabular models running on my SSAS server absolutely fine, I only have one measure in my model which is a simple sum('Table'[Field]).I'm new to analysis services in general and cant really find much of an explanation as to why this would be, unless somebody knows of a software breaking bug/incompatibility. I'd really like to have everything in a tabular model for ease of documentation and training but this issue is crippling my OCD :-). To be honest, I'm not even sure where to begin troubleshooting.Does anybody have experience with this issue to shed some light on it?Thanks in advanceHenry
↧
Need some sample for SSAS InMemory Analytics and ColumnStore
I got some idea how the InMemory analytics with cube works, such as it executes in the RAM instead of Hard disk, so it is very fast, to get hands on both InMemory Analytics and ColumnStore in SSAS , please share me the link to get some sample in both. Thanks in advance
↧
↧
SSAS 2012 ROLAP to Netezza
Has anyone on this forum ever used SSAS ROLAP to Netezza?
↧
degenerative dimension in cube
Hinewish to Data warehousing and even newer to cube development. Ive built my sales dw and when building the FactOrder table ive included the orderid as a degenerative dimension. I did this thinking it would be pulled through into the cube and so i could see the orderid in the pivot table in excel, which would make checking the cube against the source data easier. I'm now finding that this won't be the case, only keys and measures are processed by the cube anything else it ignores. Obvioulsy have loads to learn but is there a way i can get this orderid into my pivot table?Thanks
↧
MDX to Convert Revenue To Positive
I need some help. Im new to MDX and SSAS. I need to convert every value for [FMSA Monthly Activity] under the [Dim FMS Heirarchy].[FMS Reporting Hierarchy].[FMSH Financial Page] starting under the Income Statement and all the levels below that member.I have been trying this?? Im I on the right path?-- Convert Revenue To PositiveSCOPE(descendants([Dim FMS Heirarchy].[FMS Reporting Hierarchy].[FMSH Financial Page].&[Income Statement],[Dim FMS Heirarchy].[FMS Reporting Hierarchy].[FMSH Financial Page],SELF_AND_AFTER),[Dim Date].[Current YTD]);This =(([Measures].[FMSA Monthly Activity].CurrentMember)*(-1));END SCOPE[img]https://photos.google.com/share/AF1QipPNbuhGcgrbErdhESJUm6_j-sohc3xhOPmYqEywnFaFWdyA0VDsA25fTzlOXgFpiA/photo/AF1QipMWays7eHLbvjptDgsykUbyAc82kuCHVb6bJXx9?key=S08yTGdQM2p5VFpkcWkwMkRqRHJOU0k5cmlIMmtR[/img][url=https://photos.google.com/share/AF1QipPNbuhGcgrbErdhESJUm6_j-sohc3xhOPmYqEywnFaFWdyA0VDsA25fTzlOXgFpiA/photo/AF1QipMWays7eHLbvjptDgsykUbyAc82kuCHVb6bJXx9?key=S08yTGdQM2p5VFpkcWkwMkRqRHJOU0k5cmlIMmtR] Picture shared here[/url]
↧
Rolling up granularity
I have the classic FactOrderLine table at the lowest granularity of items on orders. I have built this in my cube and when i connect with excel to get the total count i get the total count of all item on all orders as expected. Can I add measures to this fact table that will give me the total Orders without creating a new fact table at Order level granularity?quite new to cube design
↧
↧
Query used inside each entity in SSAS - Tabular model
HiI need the query for finding the list of query used inside each entity in SSAS - Tabular model.eg:select * from t_view used for t_view select * from t_view1 used for t_view1select * from t_view2 used for t_view2I need the output c1 c2t_view select * from t_viewt_view1 select * from t_view1
↧
Intro SSAS | SQL Server Analysis Services
Hello All,Thanks for accepting me to this skilled SQL Server forum. I have some knowledge on SQL Server Concepts, I would like to share my stuff of SQL Server and wanna expand my knowledge. Thanks,Sandra Lynn
↧
ParallelPeriod dynamic set very odd behaviour
Within an SSAS cube I administer, there are many Time dimensions with named sets. One of which returns previous week current year. Below example of set being used.[code="sql"]WITH DYNAMIC SET [Test] AS StrToSet ( " ParallelPeriod ( [Finance Date].[Year-Week].[Week], 1, StrToMember('[Finance Date].[Year-Week].[" + Cstr(datepart("yyyy",now())) + " Week " + Cstr(datepart("ww",now())) + "]') ) " )SELECT {[Test]} ON COLUMNSFROM [cube];[/code]Very simple. This should return Week 43 from the dimension.When I run this on a SQL Server 2012 instance it correctly returns Week43.I have restored the same SSAS cube to a 2016 Server, still pointing to the same datasource of the SQL Server 2012, and it's returning Week 42.I have tried a Full re-processing of the measures and dimensions but this hasn't helped.I have checked the system clock on the server.Any suggestions??
↧
Adding New Dim, in New Cube off of Old Cube, can I alter orig Prod Tables Structure?
Hello everyone, I'm still a newbie,Learning new to me existing system, SQL Server 2012, SSAS and SSIS stacks, and VS 2013 with SSDT. I have an existing cube, and a two new dimensions needed to be added. The source view in a DW already has the columns I need to add to the cube, so now changes needed there. Perform a drop/rename on the temp table/views to be the original table/vies (having backup table copy of orig).Create a new solution off the existing cube, and add or make sure to wire up the new dimensions from the now new orig table/views. Deploy, update loads, etc..Is there a better way to add what seems to be two columns of dimensions into an existing cube? Thanks
↧
↧
MDX Previous Member and Filter
Welcome everybody. It has been a quite a time since I visited forum last time. I have a problem and hope that you can help with it.I need to introduce previous member value in MDX (e.g. previous week). I can write such a MDX Code:[quote]([Date].[Week Year].CurrentMember.PrevMember,[Measures].[New Users])[/quote]It works fine with Week Year dimension. Although, when I filter it (with e.g. current week flag that I have in my model) then it cannot get previous member as this is filtered out. Same issue with Lag. Can anybody help me with resolving that issue.With Regards
↧
SSAS Tabluar Partition Tables
Hi, still somewhat a SQL Server Stack newbie, I'm taking over an new/existing SQL Server 2012, SSIS and SSAS system, and the designer is VS 2013 and Data Tools. The developer/designer built several SSIS and SSAS packages, and has a number of daily, weekly, and monthly cubes running in place. All of these cubes use the Tabular Partition Table arrangement. There are no cubes created yet that do not use the partition tables to control partition data processing.If I create a new cube, and want to use the partition tables, what tables do I update with what values to take advantage of the existing partitions? I know the easy way, is copy an existing project that is closet to the new one, and that way the metadata and partition data comes with it. So even if I did that, surely I have new records to insert into the partition tables to ultimately get the new cube operations with the partition tables, and all the SPROCS that interface between the cube, tables and partitions. Thanks, JPQ
↧
SSAS Cube Clean-Up
Hello,I'm still somewhat new'ish to SQL Server and the SSIS and SSAS stacks. I taking over an recently built SQL Server 2012, VS 2013. All dev, Test and Prod is all on the same one instance. When I logon to the Server, and then logon to the SSAS service, expand Databases, I see all the active cubes which the user community connects to from Excel and PowerBI. I also see every test and practice cube ever made, and it's really becoming a mess, and eating up space.I can tell which are active or not, or are truly a production cube. What is the best method and practice to delete these and recover space, and not foul up the works? Thanks,JPQ
↧
How to externally access data warehouse - is SSAS needed?
Hi All,Thanks in advance for answering my questionWe have a client who wants us to build a Data Warehouse and provide reports via Tableau. Data servers will be hosted with us (remote from the client). Our IT doesn't want to provide VPN access or create AD accounts for the client on the serverQuestion is: Once we develop data warehouse, how do we provide access (via Tableau) to the client? Do we need to create SSAS cube in order to facilitate this or can they just query the data warehouse (via Tableau). If so, what would we need to do to make it happenAppreciate your timeKumar
↧
↧
XP_SQLAGENT_ENUM_JOBS vs. SP_HELP_JOB
Let me start at the beginning. Here is what we are doing:1. Control-M calls... 2. A Batch File that calls...3. A Stored Procedure that runs...4. A SQL Agent Job that processes...5. An SSAS cube.The status of the SQL Agent Job must ultimately be returned to Control-M.Yes, I know that we can have Control-M process the SSAS cube directly, thereby bypassing all of those steps. However, that is not what the company wants, not my decision :-).Anyway, I am using XP_SQLAGENT_ENUM_JOBS in my stored procedure to periodically retrieve the status of the job instead of SP_HELP_JOB. I know that XP_SQLAGENT_ENUM_JOBS is undocumented. However, I decided against using MSDB.dbo.sp_help_job because:1. It generally requires OPENROWSET to put the resultset into a table, which requires additional overhead of an OLEDB connection. 2. It outputs more columns than xp_sqlagent_enum_jobs, which might also impact performance. Is my reasoning sound? What are the counter arguments.
↧
Monitor Activty on Cube
Hello, what is the best way to see any users and/or current/active connections to a SSAS Cube? Or, if not there, to the SSAS server, to be able to see if any users are currently accessing the server or Cubes? I'm hoping to find a way to determine, and disconnect if need be, before I deploy SSIS and SSAS packages on a given Cube and kick off a new run. I'll be running at the end of the user time-frame window, just don't want to cut someone off in mid-step.Thanks, JPQ
↧
SSAS Tabular - Can't Unhide Measures
Hoping to get some help here. I'm not sure if this is a bug in Visual Studio or what. I have an SSAS tabular model setup using VS 2015 Update 3. So far I have 2 cases open up with Microsoft on bugs with their SSAS tabular designer.Here's my question regarding what looks like a 3rd case I'll have to open. I have a model with some measures that I have previously hidden from client tools by right clicking the measure and selecting 'Hide from Client Tools'. I now want to unhide them and can't do it no matter what I try. I right click and choose unhide and it just remains hidden. Any ideas how to get fix this?
↧