Yeah, I can see your query would run faster but my database is far more complicated than this example. But you're right! I have to look at the structure, indexes, etc, of my database and see if I can make this query look like yours. However, I don't think it can look like yours because it's very dynamic query and I don't want to create different queries for each cases - I'm just modifing parts of the query. Maybe I shouldn't be so lazy and write a dedicated queries for each of the cases (even though it's against the best programming practices). ... No wonder why there's so many database jobs out there =) Thank you again! -- modified at 16:47 Friday 13th April, 2007

Is it smarter than the normal one? And does it cost more? If it does I'll stick with my dumb one and break the query into smaller pieces and build the 2 dimentional table withing the program. Takes 30 seconds. Thank you for all your help! LESSON LEARNED: NO CROSSTAB QUERIES ON HUGE Databases!!! Is that right?

Thank you for you effort on this. You've spend lot of time on this. I have lot of JOINs and conditions within each sub SELECT query. how many rows did you add to Tests? When I run my query on a smaller scale (ex: Just Toyotas years 2000-2006) it works fine (max 1 minute wait). When I do something like all SUV's then it goes more than 3 minutes and nobody wants to wait for that long. Thank you for everyone who tried to help! And here's my conclusion: DO NOT RUN a CROSSTAB query like this if you have complex database and each table has thousands and thousands of rows and especially if one of the tables has more than 1 million rows. This kind of cross tabbing in MS SQL just WILL NOT WORK (fast enough)! I don't know about other database engines but it seems like the engine SHOULD optimize it for you because compilers such as modern C++ compilers optimize codes. In and IDEAL database engine using this CROSSTAB example it should break down the years into threads and run them seperately - it saves lot of times. WHY? BECAUSE when I do this query using separately (dividing into smaller pieces by years) and if I run it it takes 10 seconds. Then I build the 2D table out of all this sub queries using any programming language - it takes another 10-20 seconds. So I'm going back to my old method. It also brings up another question: Can you do threads in MS SQL query and do some sort of wait on those? I figured it should do it in the background because all it takes is one semester worth of class with a project. I mean, the Microsoft operating systems are capable of doing threads (so it runs), so why they shouldn't integrate it into the SQL server? It's not a rocket science! CORRECT ME IF I'M WRONG. I COULD BE!!!

The structure itselft is not really complex. A. Cars: Make, Model, CarID, Country, BodyType, etc B. Tests: TestID, CardID, YearRan, etc C. EngineTests: EngineTestID, TestID, etc D. TestOutcomes: OutcomeID, EngineTestID, TestID, MyNumericValue, etc A - B - C - D "-" indicates one-to-many relationship I need averages for selected cars and selected test years. Not a rocket science. It's fairly large database. You can think of the scale as current car manufacturers data. As I said, it runs smooth if I do the select statement as 1 dimentional (by doing GROUP BY and turning it into 2D programmatically) And I haven't been resting for about a week now.

Basically my output of the query should return something like this: (The numbers are average of some tests that were ran on differnt model and year of cars. In order to calculate these averages I need the joins both places.) ______________|2000|2001|....|2006 Avalon________|__12|__12|....|__11 Camry_________|11.5|11.5|....|__10 Celica________|__10|__10|....|__10 Corolla_______|___9|___9|....|___7 Highlander____|____|____|....|__10 Land Cruiser__|__15|__14|....|__14 It used to be faster when I had this query in 1 dimentional form: Avalon |2000| 12 Avalon |2001| 12 ... Avalon |2006| 11 Camry |2000|11.5 ... When I had the query result in 1D format I just changed to 2 dimentional format programatically (using 2D array and binary search). I figured it would be faster if I do the whole thing in one step within the sql query. Maybe I should just stick with the old version. I mean how do you guys make crosstab queries in MS SQL?

If XP runs good it's definatly hardware and software compatability issue. Vista has lots of them. Go to your hardware manufacturer's website and look for drivers updated for Vista. PS: 1GB - 2GB of RAM in vista - just can't do you good. Make it 4GB RAM + at least 256 dedicated video memory (possible 512 if you have 27 inch + monitor) - and you'll love it. Some of your softwares might still won't work =P

search for sqlclient from your turbo c documentation.

For this kind of problems I use 2 dimentional array or some class that's similar to the 2D array as far as functionality goes. Then get the results from database 1 add it to the array, database 2 add it to the array ... database n add it to the array. You probably need to search the rows in order to keep the rows straight.

That's a clever idea. But it does not work all the time: Say you have to sort on a field - then the id's gets shuffled. For example go to dell's outlet listings page: http://outlet.us.dell.com/ARBOnlineSales/topics/global.aspx/arb/online/en/InventorySearch?c=us&cs=22&l=en&lob=DIM&MODEL_DESC=XPS%20210&s=dfh[^] I wonder if they're doing the same thing here. Or are they just pulling all the record at once and doing the paging and sorting thing at the webpage level.

I have a big problem here. I have a query that reads (suedo): select [column], (select [a row] from [joined tables] where [condition]) as column2, (select [a row] from [joined tables] where [condition]) as column3, .... (select [a row] from [joined tables] where [condition]) as columnN from [main table] This is what I meant by crosstab query. MORE DETAILS: [a row] contains basic arithmetic computations, [joined tables] join 5 tables (one of them has nearly a million records) and [condition] inludes several LIKE statements (that might be contributing to the problem??? - but I don't think this is the MAIN reason) Problem is it gets too slow when the number of columns ([columnN]) increases. What do you suggest? Should I not do it as a crosstab, and run multiple smaller queries and put them in to one big table programmatically? Then I think, the sql server has to be smarter than me... Or is there something I'm missing? PS: the SQL database is a MS SQL and running on windows server 2003. Any help, hints, suggestions, questions are greatly appreciated.

Yes, I am using HTTP and I'd like to get that 404 response if the file does not exist. My question was how to make this request and how to read the response inside asp.net. In college I learnt application layer protocols and I'd just send the requests to remote server such as SMTP, HTTP and send me responses like 404, 500 or whatever. I just want to do the same thing in ASP.NET.

Yes, that's what exactly I want to do: Request the file from the server and the check the response. My question was how to make this request and how to read the response inside asp.net.

I tried to figure out (for couple of days now) how to check to see if a remote file exists on a different server where the web application is running. If it was on a same server I would just use server.mappath and I would have no problem. But the files are on a totally different server on different domain. So, what I'd like to do is just say checkIfThisFileExistsOnRemoteServer(http://someserver.com/directory/file.pdf) Is there a asp.net (C# or VB.NET, does not matter) way to do this. Thank you, in advance. - Chandman

I am trying to add an Event Procedure to a text field such that pressing the ENTER keyboard button activates the same procedure as clicking the submit button on the form. Apparently asp.net's textbox control does not have an ability to generate those kind of event. Or is there any way to do this? Thank you

Roath Kanel wrote:

There are many step to promote your standa alone server to become a primary domain controller. You have to configure, static IP address, DNS, configure friendly name, install DNS component and create forward look zone and revers look up zone (there is more technical detail to create these zone), test DNS service whether your configure properly then finally promote your stand alone server by typing DCPROMO command from the run dialog box.

our server is already DC. running DNS. has an static IP address. DHCP is working completely fine from our "DSL" Modem (supposedly our router). I am confident with the set up. I am just more concerned about the speed and I think there's just a little trick I have to do to get the accessing speed of shared files accross the network boosted up. I'll read those articles. Thank you again for all your help. Chandman

Sebastian Schneider wrote:

On 1: If you are in a single collision domain (network using hubs) you might be experiencing the effect of high load on such a kind of network. If you have seperate collision domains (network using switches), try enabling auto-negotiation for link-speed and -mode (i.e. "MBit/s" and "Duplex-Type"). Also check that everyone is in the same network (IP-network, i.e. everyones addresses are between 192.168.0.1 and 192.168.0.254). I do not think that splitting your network and putting a router in between is gonna help you very much.

We have a modem first - Main switch - hubs - computers We use hubs because we have 2 person in a office and only one ethernet plug into the wall. The Modem is our router, DHCP, DNS as well. Probably not a good solution, right? Q: How do I connect to our switch? I don't think there's a IP for it. When you say enable auto-negotioation, do you mean there's something that you switch manually on the switch?

Sebastian Schneider wrote:

On 2: Btw, your modem seems to be a router. Otherwise, it would probably not have an IP address at all. If it also has a DHCP-server, see that only one is running (modem/router or server), two might cause some trouble. If you want to use your server for DNS and DHCP, disable DHCP in the router and configure the server to look-up any unknown address via router / ISP-DNS. The clients should then be set to use the server for DNS.

Yes, our Modem is our only DHCP, because I uninstalled the DHCP service from the server. I tried to use our server as DHCP but I thought the internet would not work when the server is shut down. Server is on all the time but I just didn't want the other computers to be dependent on it. Q: What do you think about this situation? PS: Our internet works fine though - no one's complaining about the internet speed. Thanks Sebastian, Chandman -- modified at 14:13 Friday 9th June, 2006

Thank you for your response Roath.

Roath Kanel wrote:

Chandman wrote:

pro and couple of home editions

What do you mean by home edition? Is it windows xp home edition?

Yes we have Windows XP Professional and few Home editions aswell. But I found out that if I set the Workgroup same as our domain name, the computer becomes a part of the network.

Roath Kanel wrote:

Chandman wrote:

1. Accessing people's shared folders became slower - sometimes up to a minute. Sometimes it goes really smooth

In my opinion, this issue relate to name resolution process... From here, i want to ask you a question, does your network have a DNS server? DNS server is the server that use to store a DNS name with IP address of each computer...

Yes. Our Windows Server 2000 runs a DNS server. How should I go about configuring it? In our server - in the DNS console, I just noticed that we have our server ip address for the forward lookup zones. But we don't have anything in the reverse lookup zones. Is that an issue? I am also thinking that I should set all computer's preffered DNS to our server - not our modem. But I have to have some way to forward that to our gateway if it's outgoing request outside of our LAN. So I'm not so sure on this eigther. About router: I've seen many people having a [ WAN - modem - router - switch - computers] set up. Though, our current configuration is set up in a way that we don't have a router in between our switch and modem - therefore our modem is acting like a router, hosting DHCP, and acting as a DNS. I feel like having it set up so that when the sub netmask is 255.255.255.0 forward it to our server (DNS) else forward it to our modem or directly to our ISP DNS server. But I'm not sure about how to do it and whether I should do it. Other thing that might be useful to know is that in some room we have 2 person working in there and they share the connection via hub. It's kind of expensive for us to replace all those hubs with small switches just to test, because once again we're non-profit. PS: Our internet works fine though - no one's complaining about the internet speed. Thank you again for all your replies so far. Chandman -- modified at 14:14 Friday 9th June, 2006

Hello everyone, I work at a non-profit org. and we have a small office network with 1 server (Windows Server 2000) and about 20 computers (fast, slow, dell, hp but all running Windows OS - pro and couple of home editions). Our Windows Server 2000 is on a Dell Dimension E510 with Pentium HT 2.8 GHz and 1GB of RAM. It runs as a Domain controller and running Active Directory on it. Our DHCP is our modem - just so that DHCP works even when our server is turned off. We recently switched from use of workgroup to domain after we got the server and couple of problems seem to be rising, which I need some help on. 1. Accessing people's shared folders became slower - sometimes up to a minute. Sometimes it goes really smooth. I think getting a router might solve this problem but we never had this problem before we installed the server on one of our computers 2. In our TCP/IP config we have our modem as "Preferred DNS" and our ISP DNS as "Alternate DNS". I am not sure if this is the way it should be. Seems like our Server is running DNS too. I don't know if I can disable it - though non of our computers are configured to look for DNS in the server (our local E510). 3. When a domain user wants to log on to the their computer it asks if they want to lon on to the the local computer or the domain. I was just wondering what's the difference? Which one should I use? Right now everyone is loggin onto their own local accounts because that's where they have their files and everything because that's the only place that we could log on to before we got the server. How should I go about to solve this problems. Any help, ideas, comments, questions is appreciated. Thank you for your time, Chandman

I did it long time ago. I used XML file. As far as I can remember when a new session starts log it to the XML file and when ever session ends remove it from the XML file. I think it was in the Global.asax file.

I am sure somebody here can answer my questions. 1. We have a small office network (Windows). When a domain user wants to log on to the their computer it asks if they want to lon on to the the local computer or the domain. I was just wondering what's the difference? Which one should I use? 2. When a user wants to access other user's shared file sometimes it takes almost a minute. What would be the reason? (We have a server that runs Domain Controller, but our DHCP is our router - just so that DHCP works even when our server is turned off.) Any help would be appreciated. THank you, Chandman