Search Results

Search found 65999 results on 2640 pages for 'large data volumes'.

Page 87/2640 | < Previous Page | 83 84 85 86 87 88 89 90 91 92 93 94  | Next Page >

  • Bogus InvalidOperationException (in a DataServiceRequestException)

    - by Andrei Rinea
    I am having a hard time with ADO.NET Data Services (formerly code-named Astoria) as it gives me a bogus exception when I try to insert a new entity from the silverlight client and trying in a clean project (the same code) doesn't. In both cases, however, data is correctly inserted into the database. Using Fiddler (an HTTP debugger I could see that there is no problem in the HTTP communication as I will show later in this question. The code : var ctx = new MyProject123Entities(new Uri("http://andreiri/MyProject.Data/Data.svc")); var i = new Zone() { Data = DateTime.Now, IdElement = 1 }; ctx.AddToZone(i); i.StareZone = new StareZone() { IdStareZone = 1 }; ctx.AttachTo("StareZone", i.StareZone); ctx.SetLink(i, "StareZone", i.StareZone); i.TipZone = new TipZone() { IdTipZone = 1 }; ctx.AttachTo("TipZone", i.TipZone); ctx.SetLink(i, "TipZone", i.TipZone); i.User = new User() { IdUser = 2 }; ctx.AttachTo("User", i.User); ctx.SetLink(i, "User", i.User); ctx.BeginSaveChanges(r =] ctx.EndSaveChanges(r), null); when run the last line (ctx.EndSaveChanges(r)) will throw the following exception : System.Data.Services.Client.DataServiceRequestException was unhandled by user code Message="An error occurred while processing this request." StackTrace: at System.Data.Services.Client.DataServiceContext.SaveAsyncResult.HandleBatchResponse() at System.Data.Services.Client.DataServiceContext.SaveAsyncResult.EndRequest() at System.Data.Services.Client.DataServiceContext.EndSaveChanges(IAsyncResult asyncResult) at MyProject.MainPage.[]c__DisplayClassd6.[]c__DisplayClassd8.[dashboard_PostZoneCurent]b__d5(IAsyncResult r) at System.Data.Services.Client.BaseAsyncResult.HandleCompleted() at System.Data.Services.Client.DataServiceContext.SaveAsyncResult.HandleCompleted(PerRequest pereq) at System.Data.Services.Client.DataServiceContext.SaveAsyncResult.AsyncEndRead(IAsyncResult asyncResult) at System.IO.Stream.BeginRead(Byte[] buffer, Int32 offset, Int32 count, AsyncCallback callback, Object state) at System.Data.Services.Client.DataServiceContext.SaveAsyncResult.AsyncEndGetResponse(IAsyncResult asyncResult) InnerException: System.InvalidOperationException Message="The context is already tracking a different entity with the same resource Uri." StackTrace: at System.Data.Services.Client.DataServiceContext.AttachTo(Uri identity, Uri editLink, String etag, Object entity, Boolean fail) at System.Data.Services.Client.MaterializeAtom.MoveNext() at System.Data.Services.Client.DataServiceContext.HandleResponsePost(ResourceBox entry, MaterializeAtom materializer, Uri editLink, String etag) at System.Data.Services.Client.DataServiceContext.SaveAsyncResult.[HandleBatchResponse]d__1d.MoveNext() InnerException: (there is no further information regarding the exception although the ADo.NET Data Service is configured to return detailed informations) However the row is inserted correctly and completely in the database. Using fiddler I can see that the request : <?xml version="1.0" encoding="utf-8" standalone="yes"?> <entry xmlns:d="http://schemas.microsoft.com/ado/2007/08/dataservices" xmlns:m="http://schemas.microsoft.com/ado/2007/08/dataservices/metadata" xmlns="http://www.w3.org/2005/Atom"> <category scheme="http://schemas.microsoft.com/ado/2007/08/dataservices/scheme" term="MyProject123Model.Zone" /> <title /> <updated>2009-09-11T13:36:46.917157Z</updated> <author> <name /> </author> <id /> <link href="http://andreiri/MyProject.Data/Data.svc/StareZone(1)" rel="http://schemas.microsoft.com/ado/2007/08/dataservices/related/StareZone" type="application/atom+xml;type=entry" /> <link href="http://andreiri/MyProject.Data/Data.svc/TipZone(4)" rel="http://schemas.microsoft.com/ado/2007/08/dataservices/related/TipZone" type="application/atom+xml;type=entry" /> <link href="http://andreiri/MyProject.Data/Data.svc/User(4)" rel="http://schemas.microsoft.com/ado/2007/08/dataservices/related/User" type="application/atom+xml;type=entry" /> <content type="application/xml"> <m:properties> <d:Data m:type="Edm.DateTime">2009-09-11T16:36:40.588951+03:00</d:Data> <d:Detalii>aslkdfjasldkfj</d:Detalii> <d:IdElement m:type="Edm.Int32">1</d:IdElement> <d:IdZone m:type="Edm.Int32">0</d:IdZone> <d:X_Post m:type="Edm.Decimal">587647.4705</d:X_Post> <d:X_Repost m:type="Edm.Decimal" m:null="true" /> <d:Y_Post m:type="Edm.Decimal">325783.077599999</d:Y_Post> <d:Y_Repost m:type="Edm.Decimal" m:null="true" /> </m:properties> </content> </entry> is well accepted and a successful response is returned : HTTP/1.1 201 Created Date: Fri, 11 Sep 2009 13:36:47 GMT Server: Microsoft-IIS/6.0 X-Powered-By: ASP.NET X-AspNet-Version: 2.0.50727 DataServiceVersion: 1.0; Location: http://andreiri/MyProject.Data/Data.svc/Zone(75) Cache-Control: no-cache Content-Type: application/atom+xml;charset=utf-8 Content-Length: 2213 <?xml version="1.0" encoding="utf-8" standalone="yes"?> <entry xml:base="http://andreiri/MyProject.Data/Data.svc/" xmlns:d="http://schemas.microsoft.com/ado/2007/08/dataservices" xmlns:m="http://schemas.microsoft.com/ado/2007/08/dataservices/metadata" xmlns="http://www.w3.org/2005/Atom"> <id>http://andreiri/MyProject.Data/Data.svc/Zone(75)</id> <title type="text"></title> <updated>2009-09-11T13:36:47Z</updated> <author> <name /> </author> <link rel="edit" title="Zone" href="Zone(75)" /> <link rel="http://schemas.microsoft.com/ado/2007/08/dataservices/related/CenterZone" type="application/atom+xml;type=feed" title="CenterZone" href="Zone(75)/CenterZone" /> <link rel="http://schemas.microsoft.com/ado/2007/08/dataservices/related/ZoneMobil" type="application/atom+xml;type=feed" title="ZoneMobil" href="Zone(75)/ZoneMobil" /> <link rel="http://schemas.microsoft.com/ado/2007/08/dataservices/related/StareZone" type="application/atom+xml;type=entry" title="StareZone" href="Zone(75)/StareZone" /> <link rel="http://schemas.microsoft.com/ado/2007/08/dataservices/related/TipZone" type="application/atom+xml;type=entry" title="TipZone" href="Zone(75)/TipZone" /> <link rel="http://schemas.microsoft.com/ado/2007/08/dataservices/related/User" type="application/atom+xml;type=entry" title="User" href="Zone(75)/User" /> <category term="MyProject123Model.Zone" scheme="http://schemas.microsoft.com ado/2007/08/dataservices/scheme" /> <content type="application/xml"> <m:properties> <d:IdZone m:type="Edm.Int32">75</d:IdZone> <d:X_Post m:type="Edm.Decimal">587647.4705</d:X_Post> <d:Y_Post m:type="Edm.Decimal">325783.077599999</d:Y_Post> <d:X_Repost m:type="Edm.Decimal" m:null="true" /> <d:Y_Repost m:type="Edm.Decimal" m:null="true" /> <d:Data m:type="Edm.DateTime">2009-09-11T16:36:40.588951+03:00</d:Data> <d:Detalii>aslkdfjasldkfj</d:Detalii> <d:IdElement m:type="Edm.Int32">1</d:IdElement> </m:properties> </content> </entry> Why do I get an exception? And, using this in a clean project does not throw the exception..

    Read the article

  • javafx tableview get selected data from ObservableList

    - by user3717821
    i am working on a javafx project and i need your help . while i am trying to get selected data from table i can get selected data from normal cell but can't get data from ObservableList inside tableview. code for my database: -- phpMyAdmin SQL Dump -- version 4.0.4 -- http://www.phpmyadmin.net -- -- Host: localhost -- Generation Time: Jun 10, 2014 at 06:20 AM -- Server version: 5.1.33-community -- PHP Version: 5.4.12 SET SQL_MODE = "NO_AUTO_VALUE_ON_ZERO"; SET time_zone = "+00:00"; /*!40101 SET @OLD_CHARACTER_SET_CLIENT=@@CHARACTER_SET_CLIENT */; /*!40101 SET @OLD_CHARACTER_SET_RESULTS=@@CHARACTER_SET_RESULTS */; /*!40101 SET @OLD_COLLATION_CONNECTION=@@COLLATION_CONNECTION */; /*!40101 SET NAMES utf8 */; -- -- Database: `test` -- -- -------------------------------------------------------- -- -- Table structure for table `customer` -- CREATE TABLE IF NOT EXISTS `customer` ( `col0` int(11) NOT NULL, `col1` varchar(255) DEFAULT NULL, `col2` int(11) DEFAULT NULL, PRIMARY KEY (`col0`) ) ENGINE=InnoDB DEFAULT CHARSET=latin1; -- -- Dumping data for table `customer` -- INSERT INTO `customer` (`col0`, `col1`, `col2`) VALUES (12, 'adasdasd', 231), (22, 'adasdasd', 231), (212, 'adasdasd', 231); /*!40101 SET CHARACTER_SET_CLIENT=@OLD_CHARACTER_SET_CLIENT */; /*!40101 SET CHARACTER_SET_RESULTS=@OLD_CHARACTER_SET_RESULTS */; /*!40101 SET COLLATION_CONNECTION=@OLD_COLLATION_CONNECTION */; my javafx codes: import java.sql.Connection; import java.sql.DriverManager; import java.sql.ResultSet; import java.sql.SQLException; import java.util.Map; import javafx.application.Application; import javafx.beans.property.SimpleStringProperty; import javafx.beans.value.ChangeListener; import javafx.beans.value.ObservableValue; import javafx.collections.FXCollections; import javafx.collections.ObservableList; import javafx.event.ActionEvent; import javafx.event.EventHandler; import javafx.scene.Scene; import javafx.scene.control.Button; import javafx.scene.control.TableCell; import javafx.scene.control.TableColumn; import javafx.scene.control.TableColumn.CellDataFeatures; import javafx.scene.control.TablePosition; import javafx.scene.control.TableView; import javafx.scene.control.TableView.TableViewSelectionModel; import javafx.scene.control.cell.ChoiceBoxTableCell; import javafx.scene.control.cell.TextFieldTableCell; import javafx.scene.layout.BorderPane; import javafx.stage.Stage; import javafx.util.Callback; import javafx.util.StringConverter; class DBConnector { private static Connection conn; private static String url = "jdbc:mysql://localhost/test"; private static String user = "root"; private static String pass = "root"; public static Connection connect() throws SQLException{ try{ Class.forName("com.mysql.jdbc.Driver").newInstance(); }catch(ClassNotFoundException cnfe){ System.err.println("Error: "+cnfe.getMessage()); }catch(InstantiationException ie){ System.err.println("Error: "+ie.getMessage()); }catch(IllegalAccessException iae){ System.err.println("Error: "+iae.getMessage()); } conn = DriverManager.getConnection(url,user,pass); return conn; } public static Connection getConnection() throws SQLException, ClassNotFoundException{ if(conn !=null && !conn.isClosed()) return conn; connect(); return conn; } } public class DynamicTable extends Application{ Object newValue; //TABLE VIEW AND DATA private ObservableList<ObservableList> data; private TableView<ObservableList> tableview; //MAIN EXECUTOR public static void main(String[] args) { launch(args); } //CONNECTION DATABASE public void buildData(){ tableview.setEditable(true); Callback<TableColumn<Map, String>, TableCell<Map, String>> cellFactoryForMap = new Callback<TableColumn<Map, String>, TableCell<Map, String>>() { @Override public TableCell call(TableColumn p) { return new TextFieldTableCell(new StringConverter() { @Override public String toString(Object t) { return t.toString(); } @Override public Object fromString(String string) { return string; } }); } }; Connection c ; data = FXCollections.observableArrayList(); try{ c = DBConnector.connect(); //SQL FOR SELECTING ALL OF CUSTOMER String SQL = "SELECT * from CUSTOMer"; //ResultSet ResultSet rs = c.createStatement().executeQuery(SQL); /********************************** * TABLE COLUMN ADDED DYNAMICALLY * **********************************/ for(int i=0 ; i<rs.getMetaData().getColumnCount(); i++){ //We are using non property style for making dynamic table final int j = i; TableColumn col = new TableColumn(rs.getMetaData().getColumnName(i+1)); if(j==1){ final ObservableList<String> logLevelList = FXCollections.observableArrayList("FATAL", "ERROR", "WARN", "INFO", "INOUT", "DEBUG"); col.setCellFactory(ChoiceBoxTableCell.forTableColumn(logLevelList)); tableview.getColumns().addAll(col); } else{ col.setCellValueFactory(new Callback<CellDataFeatures<ObservableList,String>,ObservableValue<String>>(){ public ObservableValue<String> call(CellDataFeatures<ObservableList, String> param) { return new SimpleStringProperty(param.getValue().get(j).toString()); } }); tableview.getColumns().addAll(col); } if(j!=1) col.setCellFactory(cellFactoryForMap); System.out.println("Column ["+i+"] "); } /******************************** * Data added to ObservableList * ********************************/ while(rs.next()){ //Iterate Row ObservableList<String> row = FXCollections.observableArrayList(); for(int i=1 ; i<=rs.getMetaData().getColumnCount(); i++){ //Iterate Column row.add(rs.getString(i)); } System.out.println("Row [1] added "+row ); data.add(row); } //FINALLY ADDED TO TableView tableview.setItems(data); }catch(Exception e){ e.printStackTrace(); System.out.println("Error on Building Data"); } } @Override public void start(Stage stage) throws Exception { //TableView Button showDataButton = new Button("Add"); showDataButton.setOnAction(new EventHandler<ActionEvent>() { public void handle(ActionEvent event) { ObservableList<String> row = FXCollections.observableArrayList(); for(int i=1 ; i<=3; i++){ //Iterate Column row.add("asdasd"); } data.add(row); //FINALLY ADDED TO TableView tableview.setItems(data); } }); tableview = new TableView(); buildData(); //Main Scene BorderPane root = new BorderPane(); root.setCenter(tableview); root.setBottom(showDataButton); Scene scene = new Scene(root,500,500); stage.setScene(scene); stage.show(); tableview.getSelectionModel().selectedItemProperty().addListener(new ChangeListener() { @Override public void changed(ObservableValue observableValue, Object oldValue, Object newValue) { //Check whether item is selected and set value of selected item to Label if (tableview.getSelectionModel().getSelectedItem() != null) { TableViewSelectionModel selectionModel = tableview.getSelectionModel(); ObservableList selectedCells = selectionModel.getSelectedCells(); TablePosition tablePosition = (TablePosition) selectedCells.get(0); Object val = tablePosition.getTableColumn().getCellData(newValue); System.out.println("Selected Value " + val); System.out.println("Selected row " + newValue); } } }); } } please help me..

    Read the article

  • SQL Monitor’s data repository: Alerts

    - by Chris Lambrou
    In my previous post, I introduced the SQL Monitor data repository, and described how the monitored objects are stored in a hierarchy in the data schema, in a series of tables with a _Keys suffix. In this post I had planned to describe how the actual data for the monitored objects is stored in corresponding tables with _StableSamples and _UnstableSamples suffixes. However, I’m going to postpone that until my next post, as I’ve had a request from a SQL Monitor user to explain how alerts are stored. In the SQL Monitor data repository, alerts are stored in tables belonging to the alert schema, which contains the following five tables: alert.Alert alert.Alert_Cleared alert.Alert_Comment alert.Alert_Severity alert.Alert_Type In this post, I’m only going to cover the alert.Alert and alert.Alert_Type tables. I may cover the other three tables in a later post. The most important table in this schema is alert.Alert, as each row in this table corresponds to a single alert. So let’s have a look at it. SELECT TOP 100 AlertId, AlertType, TargetObject, [Read], SubType FROM alert.Alert ORDER BY AlertId DESC;  AlertIdAlertTypeTargetObjectReadSubType 165550397:Cluster,1,4:Name,s29:srp-mr03.testnet.red-gate.com,9:SqlServer,1,4:Name,s0:,10 265549387:Cluster,1,4:Name,s29:srp-mr03.testnet.red-gate.com,7:Machine,1,4:Name,s0:,10 365548187:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s15:FavouriteThings,00 465547157:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s15:FavouriteThings,00 565546147:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s15:FavouriteThings,00 665545187:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s14:SqlMonitorData,00 765544157:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s14:SqlMonitorData,00 865543147:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s14:SqlMonitorData,00 965542187:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s4:msdb,00 1065541147:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s4:msdb,00 11…     So what are we seeing here, then? Well, AlertId is an auto-incrementing identity column, so ORDER BY AlertId DESC ensures that we see the most recent alerts first. AlertType indicates the type of each alert, such as Job failed (6), Backup overdue (14) or Long-running query (12). The TargetObject column indicates which monitored object the alert is associated with. The Read column acts as a flag to indicate whether or not the alert has been read. And finally the SubType column is used in the case of a Custom metric (40) alert, to indicate which custom metric the alert pertains to. Okay, now lets look at some of those columns in more detail. The AlertType column is an easy one to start with, and it brings use nicely to the next table, data.Alert_Type. Let’s have a look at what’s in this table: SELECT AlertType, Event, Monitoring, Name, Description FROM alert.Alert_Type ORDER BY AlertType;  AlertTypeEventMonitoringNameDescription 1100Processor utilizationProcessor utilization (CPU) on a host machine stays above a threshold percentage for longer than a specified duration 2210SQL Server error log entryAn error is written to the SQL Server error log with a severity level above a specified value. 3310Cluster failoverThe active cluster node fails, causing the SQL Server instance to switch nodes. 4410DeadlockSQL deadlock occurs. 5500Processor under-utilizationProcessor utilization (CPU) on a host machine remains below a threshold percentage for longer than a specified duration 6610Job failedA job does not complete successfully (the job returns an error code). 7700Machine unreachableHost machine (Windows server) cannot be contacted on the network. 8800SQL Server instance unreachableThe SQL Server instance is not running or cannot be contacted on the network. 9900Disk spaceDisk space used on a logical disk drive is above a defined threshold for longer than a specified duration. 101000Physical memoryPhysical memory (RAM) used on the host machine stays above a threshold percentage for longer than a specified duration. 111100Blocked processSQL process is blocked for longer than a specified duration. 121200Long-running queryA SQL query runs for longer than a specified duration. 131400Backup overdueNo full backup exists, or the last full backup is older than a specified time. 141500Log backup overdueNo log backup exists, or the last log backup is older than a specified time. 151600Database unavailableDatabase changes from Online to any other state. 161700Page verificationTorn Page Detection or Page Checksum is not enabled for a database. 171800Integrity check overdueNo entry for an integrity check (DBCC DBINFO returns no date for dbi_dbccLastKnownGood field), or the last check is older than a specified time. 181900Fragmented indexesFragmentation level of one or more indexes is above a threshold percentage. 192400Job duration unusualThe duration of a SQL job duration deviates from its baseline duration by more than a threshold percentage. 202501Clock skewSystem clock time on the Base Monitor computer differs from the system clock time on a monitored SQL Server host machine by a specified number of seconds. 212700SQL Server Agent Service statusThe SQL Server Agent Service status matches the status specified. 222800SQL Server Reporting Service statusThe SQL Server Reporting Service status matches the status specified. 232900SQL Server Full Text Search Service statusThe SQL Server Full Text Search Service status matches the status specified. 243000SQL Server Analysis Service statusThe SQL Server Analysis Service status matches the status specified. 253100SQL Server Integration Service statusThe SQL Server Integration Service status matches the status specified. 263300SQL Server Browser Service statusThe SQL Server Browser Service status matches the status specified. 273400SQL Server VSS Writer Service statusThe SQL Server VSS Writer status matches the status specified. 283501Deadlock trace flag disabledThe monitored SQL Server’s trace flag cannot be enabled. 293600Monitoring stopped (host machine credentials)SQL Monitor cannot contact the host machine because authentication failed. 303700Monitoring stopped (SQL Server credentials)SQL Monitor cannot contact the SQL Server instance because authentication failed. 313800Monitoring error (host machine data collection)SQL Monitor cannot collect data from the host machine. 323900Monitoring error (SQL Server data collection)SQL Monitor cannot collect data from the SQL Server instance. 334000Custom metricThe custom metric value has passed an alert threshold. 344100Custom metric collection errorSQL Monitor cannot collect custom metric data from the target object. Basically, alert.Alert_Type is just a big reference table containing information about the 34 different alert types supported by SQL Monitor (note that the largest id is 41, not 34 – some alert types have been retired since SQL Monitor was first developed). The Name and Description columns are self evident, and I’m going to skip over the Event and Monitoring columns as they’re not very interesting. The AlertId column is the primary key, and is referenced by AlertId in the alert.Alert table. As such, we can rewrite our earlier query to join these two tables, in order to provide a more readable view of the alerts: SELECT TOP 100 AlertId, Name, TargetObject, [Read], SubType FROM alert.Alert a JOIN alert.Alert_Type at ON a.AlertType = at.AlertType ORDER BY AlertId DESC;  AlertIdNameTargetObjectReadSubType 165550Monitoring error (SQL Server data collection)7:Cluster,1,4:Name,s29:srp-mr03.testnet.red-gate.com,9:SqlServer,1,4:Name,s0:,00 265549Monitoring error (host machine data collection)7:Cluster,1,4:Name,s29:srp-mr03.testnet.red-gate.com,7:Machine,1,4:Name,s0:,00 365548Integrity check overdue7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s15:FavouriteThings,00 465547Log backup overdue7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s15:FavouriteThings,00 565546Backup overdue7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s15:FavouriteThings,00 665545Integrity check overdue7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s14:SqlMonitorData,00 765544Log backup overdue7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s14:SqlMonitorData,00 865543Backup overdue7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s14:SqlMonitorData,00 965542Integrity check overdue7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s4:msdb,00 1065541Backup overdue7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s4:msdb,00 Okay, the next column to discuss in the alert.Alert table is TargetObject. Oh boy, this one’s a bit tricky! The TargetObject of an alert is a serialized string representation of the position in the monitored object hierarchy of the object to which the alert pertains. The serialization format is somewhat convenient for parsing in the C# source code of SQL Monitor, and has some helpful characteristics, but it’s probably very awkward to manipulate in T-SQL. I could document the serialization format here, but it would be very dry reading, so perhaps it’s best to consider an example from the table above. Have a look at the alert with an AlertID of 65543. It’s a Backup overdue alert for the SqlMonitorData database running on the default instance of granger, my laptop. Each different alert type is associated with a specific type of monitored object in the object hierarchy (I described the hierarchy in my previous post). The Backup overdue alert is associated with databases, whose position in the object hierarchy is root → Cluster → SqlServer → Database. The TargetObject value identifies the target object by specifying the key properties at each level in the hierarchy, thus: Cluster: Name = "granger" SqlServer: Name = "" (an empty string, denoting the default instance) Database: Name = "SqlMonitorData" Well, look at the actual TargetObject value for this alert: "7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s14:SqlMonitorData,". It is indeed composed of three parts, one for each level in the hierarchy: Cluster: "7:Cluster,1,4:Name,s7:granger," SqlServer: "9:SqlServer,1,4:Name,s0:," Database: "8:Database,1,4:Name,s14:SqlMonitorData," Each part is handled in exactly the same way, so let’s concentrate on the first part, "7:Cluster,1,4:Name,s7:granger,". It comprises the following: "7:Cluster," – This identifies the level in the hierarchy. "1," – This indicates how many different key properties there are to uniquely identify a cluster (we saw in my last post that each cluster is identified by a single property, its Name). "4:Name,s14:SqlMonitorData," – This represents the Name property, and its corresponding value, SqlMonitorData. It’s split up like this: "4:Name," – Indicates the name of the key property. "s" – Indicates the type of the key property, in this case, it’s a string. "14:SqlMonitorData," – Indicates the value of the property. At this point, you might be wondering about the format of some of these strings. Why is the string "Cluster" stored as "7:Cluster,"? Well an encoding scheme is used, which consists of the following: "7" – This is the length of the string "Cluster" ":" – This is a delimiter between the length of the string and the actual string’s contents. "Cluster" – This is the string itself. 7 characters. "," – This is a final terminating character that indicates the end of the encoded string. You can see that "4:Name,", "8:Database," and "14:SqlMonitorData," also conform to the same encoding scheme. In the example above, the "s" character is used to indicate that the value of the Name property is a string. If you explore the TargetObject property of alerts in your own SQL Monitor data repository, you might find other characters used for other non-string key property values. The different value types you might possibly encounter are as follows: "I" – Denotes a bigint value. For example, "I65432,". "g" – Denotes a GUID value. For example, "g32116732-63ae-4ab5-bd34-7dfdfb084c18,". "d" – Denotes a datetime value. For example, "d634815384796832438,". The value is stored as a bigint, rather than a native SQL datetime value. I’ll describe how datetime values are handled in the SQL Monitor data repostory in a future post. I suggest you have a look at the alerts in your own SQL Monitor data repository for further examples, so you can see how the TargetObject values are composed for each of the different types of alert. Let me give one further example, though, that represents a Custom metric alert, as this will help in describing the final column of interest in the alert.Alert table, SubType. Let me show you the alert I’m interested in: SELECT AlertId, a.AlertType, Name, TargetObject, [Read], SubType FROM alert.Alert a JOIN alert.Alert_Type at ON a.AlertType = at.AlertType WHERE AlertId = 65769;  AlertIdAlertTypeNameTargetObjectReadSubType 16576940Custom metric7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s6:master,12:CustomMetric,1,8:MetricId,I2,02 An AlertType value of 40 corresponds to the Custom metric alert type. The Name taken from the alert.Alert_Type table is simply Custom metric, but this doesn’t tell us anything about the specific custom metric that this alert pertains to. That’s where the SubType value comes in. For custom metric alerts, this provides us with the Id of the specific custom alert definition that can be found in the settings.CustomAlertDefinitions table. I don’t really want to delve into custom alert definitions yet (maybe in a later post), but an extra join in the previous query shows us that this alert pertains to the CPU pressure (avg runnable task count) custom metric alert. SELECT AlertId, a.AlertType, at.Name, cad.Name AS CustomAlertName, TargetObject, [Read], SubType FROM alert.Alert a JOIN alert.Alert_Type at ON a.AlertType = at.AlertType JOIN settings.CustomAlertDefinitions cad ON a.SubType = cad.Id WHERE AlertId = 65769;  AlertIdAlertTypeNameCustomAlertNameTargetObjectReadSubType 16576940Custom metricCPU pressure (avg runnable task count)7:Cluster,1,4:Name,s7:granger,9:SqlServer,1,4:Name,s0:,8:Database,1,4:Name,s6:master,12:CustomMetric,1,8:MetricId,I2,02 The TargetObject value in this case breaks down like this: "7:Cluster,1,4:Name,s7:granger," – Cluster named "granger". "9:SqlServer,1,4:Name,s0:," – SqlServer named "" (the default instance). "8:Database,1,4:Name,s6:master," – Database named "master". "12:CustomMetric,1,8:MetricId,I2," – Custom metric with an Id of 2. Note that the hierarchy for a custom metric is slightly different compared to the earlier Backup overdue alert. It’s root → Cluster → SqlServer → Database → CustomMetric. Also notice that, unlike Cluster, SqlServer and Database, the key property for CustomMetric is called MetricId (not Name), and the value is a bigint (not a string). Finally, delving into the custom metric tables is beyond the scope of this post, but for the sake of avoiding any future confusion, I’d like to point out that whilst the SubType references a custom alert definition, the MetricID value embedded in the TargetObject value references a custom metric definition. Although in this case both the custom metric definition and custom alert definition share the same Id value of 2, this is not generally the case. Okay, that’s enough for now, not least because as I’m typing this, it’s almost 2am, I have to go to work tomorrow, and my alarm is set for 6am – eek! In my next post, I’ll either cover the remaining three tables in the alert schema, or I’ll delve into the way SQL Monitor stores its monitoring data, as I’d originally planned to cover in this post.

    Read the article

  • Fin de LINQ to HPC : Microsoft abandonne sa plateforme de traitement de gros volumes de données pour son concurrent Hadoop

    Fin de LINQ to HPC : Microsoft abandonne sa plateforme de traitement de gros volumes de données pour se concentrer sur le support de son concurrent Hadoop Microsoft abandonne LINQ to HPC (High performance computing), nom de code Dryad, sa propre plateforme haute performance pour des calculs distribués et la gestion intensive des données, pour se concentrer sur le support de son concurrent Hadoop dans ses produits. L'éditeur avait récemment manifesté son intérêt pour la plateforme Java de stockage et de traitement par lot de très grandes quantités de données (Big Data) Hadoop, en publiant notamment deux connecteurs

    Read the article

  • PostgreSQL lots of large Arrays and Writes

    - by strife911
    Hi, I am running a python program that spawns 8 threads and as each thread launch its own postmaster process via psycopg2. This is to maximize the use of my CPU-cores (8). Each thread call a series of SQL Functions. Most of these functions go through many thousands of rows each associated to a large FLOAT8[] Array (250-300) values by using unnest() and multiplying each FLOAT8 by an another FLOAT8 associated to each row. This Array approach minimized the size of the Indexes and the Tables. The Function ends with an Insert into another Table of a row of the same form (pk INT4, array FLOAT8[]). Some SQL Functions called by python will Update a row of these kind of Tables (with large Arrays). Now I currently have configured PostgreSQL to use most of the memory for cache (effective_cache_size of 57 GB I think) and only a small amount of it for shared memory (1GB I think). First, I was wondering what the difference between Cache and Shared memory was in regards to PostgreSQL (and my application). What I have noticed is that only about 20-40% of my total CPU processing power is used during the most Read intensive parts of the application (Select unnest(array) etc). So secondly, I was wondering what I could do to improve this so that 100% of the CPU is used. Based on my observations, it does not seem to have anything to do with python or its GIL. Thanks

    Read the article

  • osx bash grep - finding search terms in a large file with one single line

    - by unsynchronized
    Is there simple unix command line i can enter which lets me isolate say 512 bytes either side of a search term, even if there is only one "line" in a very large text file? Ok, this should be easy. Famous last words. I'm not that familiar with grep, but it seems it is mainly used to filter out lines in the input that contain search terms. I have a very large json file that I downloaded that i want to search for a particular term. before you click the link - it's over 244MB so be warned - it is from the internet wayback machine and contains lists of zip files of archived photos. i am trying to find mine. Their web interface is broken, so i found the json file that they make public here - it's the last one on the list. when i grep looking for my username, it finds it, but proceeds to dump that line to the console. the problem is that line is 244MB long, and it's the only line in the file. i tried using less, but could not get that to do much - it's very slow, and seems to have the same issue. is there simple unix command line i can enter which lets me isolate say 512 bytes either side of a search term?

    Read the article

  • iis7 large worker process request queue creating process blocking aspnet.config & machine.config amended (bottleneck)

    - by scott_lotus
    ASP.net 2.0 app .net 2.0 framework IIS7 I am seeing a large queue of "requests" appear under the "worker process" option. State recorded appear to be Authenticate Request and Execute Request Handles more than anything else. I have amended aspnet.config in C:\Windows\Microsoft.NET\Framework64\v2.0.50727 (32 bit path and 64 bit path) to include: maxConcurrentRequestsPerCPU="50000" maxConcurrentThreadsPerCPU="0" requestQueueLimit="50000" I have amended machine.config in C:\Windows\Microsoft.NET\Framework64\v2.0.50727\CONFIG (32 bit and 64 bit path) to include: autoConfig="true" maxIoThreads="100" maxWorkerThreads="100" minIoThreads="50" minWorkerThreads="50" minFreeThreads="176" minLocalRequestFreeThreads="152" Still i get the issue. The issue manifestes itself as a large number of processes in the Worker Process queue. Number of current connections to the website display 500 when this issue occurs. I dont think i have seen concurrent connections over 500 without this issue occurring. Web application slows as the request block. Refreshing the app pool resolves for a while (as expected) as the load is spread between the two pools. Application pool in question FIXED REQUEST have been set to refresh on 50000. Thank you for any help. Scott quick edit to say hmm, my develeopers are telling me the project was built with .net 3.5 framework. Looking at C:\Windows\Microsoft.NET\Framework64\v3.5 there does not appear to be a ASPNET.CONFIG or a MACHINE.CONFIG .... is there a 3.5 equivalent ? after a little searching apparenetly 3.5 uses the 2.0 framework files that 3.5 is missing. So back to the original question , where is my bottleneck ?

    Read the article

  • Load Sharing Regarding Large Websites

    - by JHarley1
    Hello, I have a question regarding Load Sharing for large websites. My Understanding: So if you have a website that has millions of fits a day you will need to have an architecture that can support this sort of pressure. You can either do one or two things: Invest in a single large server that has huge amounts of processing power, memory and storage (such as Microsoft's TerraServer). Spread the load of your website across a number of machines. Let me tackle the second approach, so you have a collection of machines all running Web Server Software and all having access to identical copies of the websites pages. You can either spread the load across these machines using a cyclic pattern in a DNS or you can use a Load Ballancing Switch. The advantages of this approach is: - Redundancy - servers can fail and the others would "pick up the slack" - Incremental - the ability to easily add new machines to this set-up. My Question's Is there a Virtual approach to this issue of load balancing now? If the website runs from a database - is there still only a single copy of the database? If a user had a session running on one Server (e.g. they had gone to www.example.org and had been assigned to Server 2 - were they had created a session) if they refreshed the website (and were allocated Server 3) would they still have their session? What are the other disadvantages associated with Load Balancing? Many Thanks, J

    Read the article

  • How do I import large sql file to local LAMP (xampp) environment

    - by mraslton
    I have used Linux to import a large mysql dump file (into a new database), but am new to how the process works in a local LAMP environment using xampp, as xampp does not support SSH. I've dowloaded the large_dump_file.sql from the Linux server to my local system. I'm using Windows XP and have used xampp to setup LAMP. I am able to access the local_database via phpMyAdmin, but the dump file is too large to import using that app. I'm trying to import the file via the command prompt, but so far with no success. At the prompt: cd .. cd .. cd xampp cd mysql cd bin I've found that mysqlimport is used to import .csv and .txt files, and mysql is used to import .sql files, but can't find documentation as to whether or not to use the -u -p options so I've tried many variations of the command with no luck. What would be the proper command? I've modified the hosts, virtual-hosts conf, and apache config files. Do I need to change any other config files on my local system? Thanks.

    Read the article

  • Error importing large MySQL dump file which includes binary BLOBs in Windows

    - by Daniel Magliola
    I'm trying to import a MySQL dump file, which I got from my hosting company, into my Windows dev machine, and i'm running into problems. I'm importing this from the command line, and i'm getting a very weird error: ERROR 2005 (HY000) at line 3118: Unknown MySQL server host '+?*á±dÆ-N+Æ·h^ye"p-i+ Z+-$?P+Y.8+|?+l8/l¦¦î7æ¦X¦XE.ºG[ ;-ï?éµ?º+¦¦].?+f9d릦'+ÿG?-0à¡úè?-?ù??¥'+NÑ' (11004) I'm attaching the screenshot because i'm assuming the binary data will get lost... I'm not exactly sure what the problem is, but two potential issues are the size of the file (2 Gb) which is not insanely large, but it's not trivially small either, and the other is the fact that many of these tables have JPG images in them (which is why the file is 2Gb large, for the most part). Also, the dump was taken in a Linux machine and I'm importing this into Windows, not sure if that could add to the problems (I understand it shouldn't) Now, that binary garbage is why I think the images in the file might be a problem, but i've been able to import similar dumps from the same hosting company in the past, so i'm not sure what might be the issue. Also, trying to look into this file (and line 3118 in particular) is kind of impossible given its size (i'm not really handy with Linux command line tools like grep, sed, etc). The file might be corrupted, but i'm not exactly sure how to check it. What I downloaded was a .gz file, which I "tested" with WinRar and it says it looks OK (i'm assuming gz has some kind of CRC). If you can think of a better way to test it, I'd love to try that. Any ideas what could be going on / how to get past this error? I'm not very attached to the data in particular, since I just want this as a copy for dev, so if I have to lose a few records, i'm fine with that, as long as the schema remains perfectly sound. Thanks! Daniel

    Read the article

  • Fast distributed filesystem for a large amounts of data with metadata in database

    - by undefined hero
    My project uses several processing machines and one storage machine. Currently storage organized with a MSSQL filetable shared folder. Every file in storage have some metadata in database. Processing machines executes tasks for which they needed files from storage and their metadata. After completing task, processing machine puts resulting data back in storage. From there its taken by another processing machine, which also generates some file and put it back in storage. And etc. Everything was fine, but as number of processing machines increases, I found myself bottlenecked myself with storage machines hard drive performance. So I want processing machines to put files in distributed FS. to lift load from storage machines, from which they can take data from each other, not only storage machine. Can You suggest a particular distributed FS which meets my needs? Or there is another way to solve this problem, without it? Amounts of data in FS in one time are like several terabytes. (storage can handle this, but processors cannot). Data consistence is critical. Read write policy is: once file is written - its constant and may be only removed, but not modified. My current platform is Windows, but I'm ready to switch it, if there is a substantially more convenient solution on another one.

    Read the article

  • How to configure a large mtu (linux)

    - by Somejan
    I have a gigabit ethernet connection from my laptop to my router, and a working ipv6 connection to the internet. I can receive very large packets from sites on the internet, with sizes up to at least 10000 bytes (according to wireshark). (edit: turns out to be linux's 'generic receive offload') However, when trying to send anything, my local computer fragments at just below 1500 bytes for ipv6. (On ipv4, I can send tcp packets to the internet of at least 1514 bytes, I can ping with packets up to the configured mtu of 6128 but they are blackholed.) I'm on ubuntu 12.04. I have configured an mtu for my eth0 of 6128 (the maximum it accepts), both using ip link set dev eth0 mtu 6128 and in the NetworkManager applet gui, and restarted the connection. ip link show eth0 shows the 6128 mtu is indeed set. ip -6 route shows that none of the paths the kernel knows about have an mtu set. I can ping over ipv4 with packets up to 6128 bytes (though I don't get responses), but when I do ping6 myrouter -c3 -s1500 -Mdo I get error replies from my own computer saying that the packets are too large and the mtu is 1480. I have confirmed with Wireshark that nothing is put on the wire, and the replies are indeed generated by my own computer. So, how do I get my computer to use the larger mtu?

    Read the article

  • Export-Mailbox - fails with large folders

    - by grojo
    I am trying to move messages from a rather large mailbox to an archive mailbox. However I run into errors all the time. the command I am executing is Export-Mailbox -Identity MAILBOX_FROM -TargetMailbox ARCHIVE -TargetFolder ARCHIVE_FOLDER -StartDate 2009-02-01 -EndDate 2009-02-28 -DeleteContent -Confirm:$false I can copy/move some messages, but run into frequent "an unknown error has occurred" (statuscode -1056749164) I run the console as administrative user, and all permissions are set right, as far as I can tell. I've restricted the start and end dates in case the number of messages moved/deleted should create problems. Anything I am missing in my setup? Corrupted messages? Over-limit message sizes? Update: What I've learnt so far, is that folder with more than approx 3000 messages will generate errors. If mail retention is set (default 30 days), Export-Mailbox will scan all messages whether these were deleted in previous runs or not, and date restriction to limit number of messages will not work. To avoid errors, I've switched off deleted message retention for the mailbox, and moved the messages from one large folder to multiple folders, and moved these one by one...

    Read the article

  • Large, high performance object or key/value store for HTTP serving on Linux

    - by Tommy
    I have a service that serves images to end users at a very high rate using plain HTTP. The images vary between 4 and 64kbytes, and there are 1.300.000.000 of them in total. The dataset is about 30TiB in size and changes (new objects, updates, deletes) make out less than 1% of the requests. The number of requests pr. second vary from 240 to 9000 and is dispersed pretty much all over, with few objects being especially "hot". As of now, these images are files on a ext3 filesystem distributed read only across a large amount of mid range servers. This poses several problems: Using a fileysystem is very inefficient since the metadata size is large, the inode/dentry cache is volatile on linux and some daemons tend to stat()/readdir() it's way through the directory structure, which in my case becomes very expensive. Updating the dataset is very time consuming and requires remounting between set A and B. The only reasonable handling is operating on the block device for backup, copying, etc. What I would like is a deamon that: speaks HTTP (get, put, delete and perhaps update) stores data it in an efficient structure. The index should remain in memory, and considering the amount of objects, the overhead must be small. The software should be able to handle massive connections with slow (if any) time needed to ramp up. Index should be read in memory at startup. Statistics would be nice, but not mandatory. I have experimented a bit with riak, redis, mongodb, kyoto and varnish with persistent storage, but I haven't had the chance to dig in really deep yet.

    Read the article

  • File corruption (bad checksums) in large files copied to VMware guest

    - by AllanA
    In setting up a development lab, I've got a desktop system running ESXi 4.1.0 (free license) on SATA RAID 0 (already purchased and configured when I started this job; I'm open to hardware input as it pertains to my problem.) Its guests so far include two Win2008 Server R2 64-bit VMs and on Ubuntu 10.04 64-bit VM. I'm installing onto the Windows servers. We've been copying off some fairly large files (over a gigabyte) for an installation, hoping to install more quickly from a (virtual) hard drive than from the network for from BD-ROM. The problem is that they keep coming up with different checksums from the originals. The file sizes are the same, but md5sum reports different numbers (and so does the installer, as it refuses to continue when the checksums don't match.) I've tried copying directly from the BD-ROM (attaching the OS drive to the host system's physical drive). I've tried copying the large files onto a co-worker's Windows machine from his Blu-Ray drive; when I do that, the checksums match. But when I copy from his machine to the VM guest over a network share, the checksums no longer match. Thinking this meant a corrupt destination drive, I deleted it in vSphere and added another freshly created drive. The problem persists. I'm not sure what to try next.

    Read the article

  • Copy past speed very slow for a large number of files on Windows [closed]

    - by Arno2501
    I've run the following test I've created a folder containing 15'000 files of 400 bytes using this batch : @ECHO off SET times=15000 FOR /L %%i IN (1,1,%times%) DO ( fsutil file createnew filename%%i.txt 400 ) then I copy past it on my Windows Computer using this command : robocopy LargeNumberOfFiles\ LargeNumberOfFiles2\ After it has completed I can see that the transfer rate was 915810 Bytes/sec this is less than 1 MB/s. It took me several seconds to copy 7 MBytes Please note that this is very slow. I've tried the same with a folder with a single file of 50 Mbytes and the transfer rate is 1219512195 Bytes/sec. (yeah GB/s) instantaneous. Why copying large number of files take so much time - ressources on a windows filesystem ? Please note that I've tried to do the same on a linux system which runs on the same computer in a virtual machine (vmware player) with ext3 filesystem. I use the cp command and the copy is instantaneous ! Please also note the following : no antivirus I've tested that behaviour on multiple windows computers (always ntfs) i always get comparable results (transfer rate under 1MB/s avg 7-8 seconds to copy 7 MBytes) I've tested on multiple linux ext3 system the copy is always instantaneous for that amount (15000 files of 400 bytes) The question is about understanding what makes windows filesystem so slow to copy large number of files compared to a linux one for instance.

    Read the article

  • Why "Algorithms" and "Data Structures" are treated as separate disciplines?

    - by Pavel Shved
    This question was the last straw; and I've been wondering for a long time about it, Why do people think about "Algorithms" and "Data structures" as about something that can be separated from each other? I see a lot of evidence that they're separated in programmers' minds. they request "Data Structures & Algorithms" books they refer to "Data Structures" and "Algorithms" as separate university courses they "know Algorithms", but are "weak in Data Structures" (can't find the link, sorry). etc. In my opinion "Data Structures" are algorithms, since the concept of "Data Structure" is about Algorithms to operate data that go in and out of the structures. But the opinion seems not a mainstream. What do I miss?

    Read the article

  • Ideal way to deliver large data over Web Services

    - by zengr
    We are trying to design 6 web services, which will serve another client component. The client component requires data from the web service we are implementing. Now, the problem is, there is not 1 WS we are implementing, there is one WS which the client component hits, this initiates a series (5 more) of WSs which gather data from their respective data stores and finally provide the data back to the original WS, which then delivers the data back to the client component. So, if the requested data becomes huge, then, this will be a serious problem for our internal communication channel. So, what do you guys suggest? What can be done to avoid overloading of the communication channel between the internal WS and at the same time, also delivering the data to the client component.

    Read the article

  • What are the lesser known but cool data structures ?

    - by f3lix
    There a some data structures around that are really cool but are unknown to most programmers. Which are they? Everybody knows linked lists, binary trees, and hashes, but what about Skip lists, Bloom filters for example. I would like to know more data structures that are not so common, but are worth knowing because they rely on great ideas and enrich a programmer's tool box. PS: I am also interested on techniques like Dancing links which make interesting use of the properties of a common data structure. EDIT: Please try to include links to pages describing the data structures in more detail. Also, try to add a couple of words on why a data structures is cool (as Jonas Kölker already pointed out). Also, try to provide one data-structure per answer. This will allow the better data structures to float to the top based on their votes alone.

    Read the article

  • Is there a standard practice for storing default application data?

    - by Rox Wen
    Our application includes a default set of data. The default data includes coefficients and other factors that are unlikely to ever change but still need to be update-able by the user. Currently, the original default data is stored as a populated class within the application. Data updates are stored to an external XML file. This design allows us to include a "reset" feature to restore the original default data. Our rationale for not storing defaults externally [e.g. XML file] was to minimize the risk of being altered. The overall volume of data doesn't warrant a database. Is there a standard practice for storing "default" application data?

    Read the article

  • bitshift large strings for encoding QR Codes

    - by icekreaman
    As an example, suppose a QR Code data stream contains 55 data words (each one byte in length) and 15 error correction words (again one byte). The data stream begins with a 12 bit header and ends with four 0 bits. So, 12 + 4 bits of header/footer and 15 bytes of error correction, leaves me 53 bytes to hold 53 alphanumeric characters. The 53 bytes of data and 15 bytes of ec are supplied in a string of length 68 (str68). The problem seems simple enough - concatenate 2 bytes of (right-shifted) header data with str68 and then left shift the entire 70 bytes by 4 bits. This is the first time in many years of programming that I have ever needed to do something like this, I am a c and bit shifting noob, so please be gentle... I have done a little investigation and so far have not been able to figure out how to bitshift 70 bytes of data; any help would be greatly appreciated. Larger QR codes can hold 2000 bytes of data...

    Read the article

  • Can a variable like 'int' be considered a primitive/fundamental data structure?

    - by Ravi Gupta
    A rough definition of a data structure is that it allows you to store data and apply a set of operations on that data while preserving consistency of data before and after the operation. However some people insist that a primitive variable like 'int' can also be considered as a data structure. I get that part where it allows you to store data but I guess the operation part is missing. Primitive variables don't have operations attached to them. So I feel that unless you have a set of operations defined and attached to it you cannot call it a data structure. 'int' doesn't have any operation attached to it, it can be operated upon with a set of generic operators. Please advise if I got something wrong here.

    Read the article

  • Ideal way/architecture to deliver large data over Web Services

    - by zengr
    We are trying to design 6 web services, which will serve another client component. The client component requires data from the web service we are implementing. Now, the problem is, there is not 1 WS we are implementing, there is one WS which the client component hits, this initiates a series (5 more) of WSs which gather data from their respective data stores and finally provide the data back to the original WS, which then delivers the data back to the client component. So, if the requested data becomes huge, then, this will be a serious problem for our internal communication channel. So, what do you guys suggest? What can be done to avoid overloading of the communication channel between the internal WS and at the same time, also delivering the data to the client component.

    Read the article

  • EMEA Analytics & Data Integration Oracle Partner Forum

    - by milomir.vojvodic
    MONDAY 12TH NOVEMBER, 2012 IN LONDON (UK) For Oracle Partners across Europe, Middle East and Africa: come to hear the latest news from Oracle OpenWorld about Oracle BI & Data Integration, and propel your business growth as an Oracle partner. This event should appeal to BI or Data Integration specialized partners, Executives, Sales, Pre-sales and Solution architects: with a choice of participation in the plenary day and then a set of special interest (technical) sessions. The follow on breakout sessions from the 13th November provide deeper dives and technical training for those of you who wish to stay for more detailed and hands-on workshops. Keynote: Andrew Sutherland, SVP Oracle Technology Hot agenda items will include: The Fusion Middleware Stack: Engineered to work together A complete Analytics and Data Integration Solution Architecture: Big Data and Little Data combined In-Memory Analytics for Extreme Insight Latest Product Development Roadmap for Data Integration and Analytics Venue: Oracles London CITY Moorgate Offices Places are limited, Register from this Link Note: Registration for the conference and the deeper dives and technical training is free of charge to OPN member Partners, but you will be responsible for your own travel and hotel expenses. Event Schedule During this event you can learn about partner success stories, participate in an array of break-out sessions, exchange information with other partners and enjoy a vibrant panel discussion. Nov. 12th  : Day 1 Main Plenary Session : Full day, starting 10.30 am.  Oracle Hosted Dinner in the Evening Nov. 13th  onwards Architecture Masterclass : IM Reference Architecture – Big Data and Little Data combined (1 day) BI-Apps Bootcamp  (4-days) Oracle GoldenGate workshop (1 day) Oracle Data Integrator and Oracle Enterprise Data Quality workshop (1 day) For further information and detail download the Agenda (pdf) or contact Michael Hallett at [email protected] and Milomir Vojvodic at [email protected] v\:* {behavior:url(#default#VML);} o\:* {behavior:url(#default#VML);} w\:* {behavior:url(#default#VML);} .shape {behavior:url(#default#VML);} Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;}

    Read the article

  • Very large database, very small portion most being retrieved in real time

    - by mingyeow
    Hi folks, I have an interesting database problem. I have a DB that is 150GB in size. My memory buffer is 8GB. Most of my data is rarely being retrieved, or mainly being retrieved by backend processes. I would very much prefer to keep them around because some features require them. Some of it (namely some tables, and some identifiable parts of certain tables) are used very often in a user facing manner How can I make sure that the latter is always being kept in memory? (there is more than enough space for these) More info: We are on Ruby on rails. The database is MYSQL, our tables are stored using INNODB. We are sharding the data across 2 partitions. Because we are sharding it, we store most of our data using JSON blobs, while indexing only the primary keys

    Read the article

< Previous Page | 83 84 85 86 87 88 89 90 91 92 93 94  | Next Page >