I'm building a WPF, C#, .NET solution that needs to rapidly change data connections.
All the connections will eventually sync back to a parent Oracle database, but in the meantime, I might be pulling data from an access database, or a local sql server compact database, or an xml file, or even a web service, or sharepoint. Problem is, I might even need to add providers, and need to be able to keep the providers in sync with each other, and do it real time, with no loss of connection/seamless to the users. This is all dependent on what type of machine, which domain, and what kind of network connectivity we have, and is a client requirement, not something I can change.
Does anyone have a good recommendation for what the best way to accomplish this would be?
Has the client explained why they need this functionality? Often they ask for things to solve a problem that they forsee, without adequate knowledge of how best to solve it.
If you're coding something like an application for a travelling salesman to use, which will have intermittent connectivity to the Oracle database, then maybe you should look at using some means other than a direct database connection for synchronising the databases.
Say using a WCF/SOAP service to pass serialized data objects back and forth or you could look at using MSMQ to transfer changes back and forth between the intermittently connected mobile application and the Oracle database server. It would, of course mean that you'll need to run a server side application/service to handle this data and pass it into the Oracle database, but it would allow for intermittent connections to be handled more easily without having to handle database connection error logic.
In the meantime if your client code should look at layering the code to use a factory Repository type pattern. As business logic just calls an interface it is then possible to use database specific code within your data layer that was decided upon at run time (say through a config setting).
you can create one or two tables in any of your server whcih will include all connection strings and the condtions on which data you need to use which connection .
and your business layer will be independent of connections .
other options is to use sofware facotry pattern.
there you can repository independent of connections and on run time decided which data repository will connect to whcih DB.
Related
I am developing a windows application for agricultural purpose. This application will be used by multiple users to maintain the data. The main issue is there won't be network connectivity on the work location. But however by end of the day they can go and synchronize if there are any option.
I just want to know how can we import and store all the data locally and update the data to database when there is network.
The options that i thought is to have SQL on every machine that runs this application. Store the data to local database when there is no network.
Having a separate button to export the local data to the centralized database when there is network.
Looks like this is complicated. Is there any better and easier option.
I prefer using c#, Visual studio.
Thanks.
You can use SQLite for storing data locally. It's fast, lightweight, and public domain.
You can use whatever the database of choice for the centralized server.
Well, this a quite broad question, as it has many options and scenarios. The questions you should ask yourself are:
Does user handle new information only or any information from any other user from the previous syncing?
Do you have to handle update conflicts?
Do you handle text information only or you have complex types and binary files?
As for the solution, the easiest way, from my point of view, would be using SQL Lite on portable devices, is a lightweight SQL client that will allow you to handle information easily. On the server you can use whatever you want, SQL Server, MySQL or any other SQL flavor you may like. Just make sure there is a connector for your portable device OS.
If you keep thinking of using SQL server on the portable device, it's a battery hogger!!!, you might want to check Microsoft Sync framework, as it provides almost all possible scenarios for handling data syncing, manage conflicts, etc.
Thanks for the answers. Please find the below solution that we implemented.
1) Installed SQL express on all the local machines
2) Used Microsoft Sync framework to sync the data. The sync is configured on demand.
Issues faced:
1) We were using geometry datatype on few tables and this was not supported by sync framework.
2) Any change in the database schema will not reflect on the client machine. We will have to delete all the system generated procedures used to track the table change and regenerate it. I am sure there will be a much better way to do this.
Cheers,
Jebli
I am current building (in C#) a fairly basic point-of-sale program for a local community in Uganda to use in tracking business at their sunflower seed press. I was thinking that I would need some sort of database (like a SQL database), but I've never set up a database before, so I'm wondering what the best way to do this is. Maybe a database isn't the best way. The program will not have internet access, so everything will have to be done locally on the machine.
I think your first step should be designing out what data you need to store. Build an Entity Relationship Model and decide what your domain model is going to be. There are many different Database Engines out there that you can use that have different features, installation requirements, etc. A database engine can be installed locally, or on a remote machine to connect to. If you're writing a C# app, you'll probably want to use the System.Data namespace. You can use plain ADO .NET, or use something like Linq To Enttiies to help create proxy classes for your data tables.
You can access a SQL database using the same API for queries / record extraction regardless of the DB Engine uses. In some caess, you may need to use a seperate library that provides an implementation (or a better one), as in the case of an Oracle Database and the Oracle Data Access Components. Right out of the gate, .NET works very well with Microsoft SQL Server, but other options would work.
The details of what database engine are not as important as defining a good set of data tables to represent your data.
Yes. If it has lots of data you have to consider using database. Whether you have internet or not, as long as you have local network, you can easily do database.
Set up a database server ( maybe sql)
Do your database and install it on the database server
Do your application and connect to your database through connection string.
You are on the right track to use a database to store data. It is pretty easy to accomplish. Your computer does not need to be connected to the internet.
SQL Server Express Edition is free with a limit of 10 gigs of data. This will probably be much, much more space than you will need.
From C#, use ADO.NET. It is very simple if you know some SQL. Code samples here.
Is there a standard messaging protocol(s) / API(s) available to keep databases in sync. Or alternatively API(s) for creating and parsing messages.
Our company is working with another company to provide two different software packages to two different kinds of users. The data sits in two separate databases but parts of it have to remain in sync.
Their system is pretty much a black box to us. And vice versa.
So what would be required would be to track updates, and turn these into messages and send them to a web service, map these back to the destination database fields, and commit them.
The database schemas do not match.
I am aware that we are going to have to roll most of this ourself, but some ideas around messaging or techniques would be good.
One solution : SQL Server Integration Service. It appears from SQL Server 2005. This is exactly what you need. It was called DTS in SQL Server 2000 for Data Transformation Service. This was created to import/export/transform data from one point to an other. This is really easy to use from SQL Server 2005 (DTS is quite horrible).
So basically, you will have to write packages to import data from their database, transform, filter, etc. it exactly how you need it to insert it into your database. And vice versa.
Regarding the black box fact, you should generate the database relational design to make it easier.
EDIT
Just in case of you need to install it, I remember bugs from the SQL Server 2005 installer not installing SSIS at all. I had to satisfy all warnings in the installer system requirements step to obtain it.
You have two problems:
track the changes that have to be synced
apply the changes to the peer
There is a solution that combines a solution to both issues and I'm sure you are aware of it: replication. Merge Replication would allow both sites to update the data and would also provide merge conflict resolution. But replication only works when the table schema is similar and puts a big constraint on development as schema changes have to be carefully coordinated between the sites. In practice, when the sites are operated by independent companies, is quite difficult to maintain for a long term.
If you want to roll your own the change tracking part has built in support in SQL Server:
Change Tracking
Change Data Capture
Both can be used for a sync solution as a mean to detect what changed.
Applying the changes can be resolved by a web service, but there are also built-in solutions in SQL Server that allow for far higher scalability and throughput: Service Broker. Relying on a message defined API for sync allows the two sites to evolve at their own pace and change the schema almost at will, as long as the communication API (the message protocol)remains unchanged.
The answers provided give me some good ideas, but I think we are going to end up doing something a bit different.
We are using MSMQ, and defining a standard messaging system which we will roll ourselves.
As to how we will know what things have changed I am not sure at the moment.
I am designing a program that will build and maintain a database, and act as a central server. This is the 'first stage' of a grander plan. Coming later will be 3-5 remote programs built around the information put into this database.
The requirements are:
The remote programs must be able to access the information in the database.
The remote programs must be able to set alerts when information in the database changes.
The remote programs must be able to request the central server to go out and fetch new / different data.
So, the question is this: how do I expose this data and events to the outside world? My two choices are:
Have them communicate directly with my 'server' application. This seems easier to:
do event notifications (although I suppose I'm probably missing something in SQL).
It also seems like this is more 'upgradeable' - that is I don't need to worry about the database updating and crashing all my remote programs because something changed. I can account for this and transform it the data to a version the child program will understand.
Just go ahead and let them connect directly to the database.
This nice thing about this is that it's solved. I can use LINQ for SQL. The only thing the main server application needs to do is let the remote programs know where the database is.
I'm unsure how to trigger / relay 'events' for field changes in a database over different programs that may or may not be on the same computer.
Forgive my ignorance on this question. I feel woefully unprepared to ask it, but I'm having a hard time figuring out where to get started with this. It is my first real DB project :-/
Thanks!
If the other programs are going to need to know about updates to the database, then the best solution is to manage all db updates through your server application so it can alert clients of the changes. Otherwise it will be tough for the clients to be aware of changes to the db. This also has the advantage of hiding the implementation details of your storage solution from the clients, so you are free to change databases, etc...
My suggestion would be to go with option 1. Build out a web service that can provide the information they all need. This will be the most flexible and allow you to reduce duplicate backend code that would happen with direct communication with the database.
I would recommend looking at some Data Source design patterns first. This types of patterns will help you come up with solutions about how to manage the states of your data. Otherwise I think that I would require some more information about your requirements for the clients to make any further useful suggestions.
I recommend you learn about SQL Server and/or databases first. You don't appear to realize that most of what you want from your "central server" can all be done by SQL Server itself.
A central databse is the simplest option and the cheapest to both build and maintain.
There are however a few scenarios where a central database could cause problems:
High load on one of the systems: A high load on one of the systems could reduce performance on the other systems. For example someone running an internal report stops you being able to take orders on your eCommerce site.
With several systems writing to the same database there is a greater chance of locking.
With several systems dependent on the same database schema, how do you upgrade? All systems at the same time?
If you need to take down the database all systems stop.
I am creating a simple project management system which uses NHibernate for object storage. The underlying database is SQL express (at least currently for development).
The client runs on either the desktop or laptop. I know I could use web-services and store the DB only on the desktop, but this would force the desktop to be available all the time.
I am currently thinking about duplicating the DB, having two instances with "different data". To clarify, we are not talking about a productive app here, its a prototype.
One way to achieve this very simple would be the following process:
Client: Check if desktop DB is available (through web service)
Client: If yes, use desktop storage, no problem here
Client: If not, use own DB as storage
Client: Poll desktop regulary, as soon as it comes on, sync
Client: Switch to desktop storage
...
Desktop: Do not attempt any DB operation before checking for required sync
Desktop: If sync needed, do it...
My question is now, how would you sync? Assume 4 or 5 types of objects, all have GUID as identifiers. Would you always manually "lazy load" all objects of a certain type and feed them to the DB. Would you always drop the whole desktop DB in case the client DB may be newer and out of sync?
Again, I want to stress out, I am not assuming any conflicts or stale data, I basically just want to "copy the whole DB from the client". Would you use NHibernate for this? Or would you separate the copy process?
When I think about it, my questions comes down to this:
Is there any function from NHibernate:
SyncDBs_SourceWins_(SourceDB, TargetDB)
Thanks for help,
Chris
Not NHibernate, but how about Sync Services for ADO.NET 2.0?
-sa