I want to design a new distributed application, but I have a few queries that I need some genius advice on, hopefully from you people:
Scenario
I currently support a legacy application that is starting to fall between the cracks.
It is a distributed Client-Server app implemented using .Net Remoting. I can't explain exactly what it does, because I'm not allowed to.......But let's just say that it does LOTS of MATHS. I want to re-design and re-write the application using WCF.
Pre-requisites
The server side of the implementation will be hosted in a Windows Service.
The client side will be a windows forms application.
The server side will perform lots of memory-intensive processing.
The server will spit this data out to multiple thin clients (20-ish).
The majority of the time the server will be passing data to the clients, but occasionally the clients will be persisting data back to the server.
The speed at which the data is transmitted is highly-important, however I'm well aware that WCF can handle fast distribution of data.
Encryption/Security is not that important as the app will run on a highly protected local network.
Queries
Given the information above:
1)What sort of design pattern am I best going with? - Baring in mind I want the server to continually PUSH the newly calculated information immediately to the clients, as opposed to the current implementation that involves the client pulling from the server continuously.
2)What type of WCF binding should I use to ensure maximum speed of data transfer? (as close to real-time as possible is what I'm after)
3)Should I use a class library to share the common objects between the client and the server applications?
4)What is the best way in which to databind my objects on the client side in order to see live updates continually as data changes?
If I've forgotten anything then feel free to point this out
Help greatly appreciated.
1) What sort of design pattern am I best going with?
Based on your comments, you're wanting to transform the current polling mechanism to an event-based mechanism. That is, instead of the client constantly checking the server for results, have the server notify the client when a new calculation result is available.
I would recommend using Juval Lowy's Publish-Subscribe Framework for this.
(source: microsoft.com)
.
This framework is described in detail in this MSDN article. And you can download the framework's source code for free at Lowy's website, IDesign.net.
Basically, the server logic that performs the calculations inside the Windows service is the Publishing Client in the graphic, and the various WinForm applications are the Subscribing Clients. The Pub/Sub Service lives in your Windows service. It manages the list of subscribing clients and provides a single endpoint for your server to publish calculation results to. In this way, your server performs a calculation and publishes the result once to the Pub/Sub Service endpoint. The Pub/Sub Service is then responsible for publishing the result to the subscribed clients.
2) What type of WCF binding should I use to ensure maximum speed of data transfer?
If all of your WCF communication were on a single machine, you'd want to use the NetNamedPipeBinding. However, since you will be distributed, you want to use the NetTcpBinding.
For WCF binding decisions, I have found this chart useful.
3) Should I use a class library to share the common objects between the client and the server applications?
Since you are in control of both the client and server side, I would highly recommend sharing a class library instead of using Visual Studio's "Add Service Reference" feature. For a detailed discussion of this, refer to this SO question-and-answer.
4) What is the best way in which to databind my objects on the client side in order to see live updates continually as data changes?
I suspect this will depend on what controls you use to display the data. One way that immediately comes to mind would be to have your client fill an in-memory data table as each calculation result is received. This data table could then be bound to a ListBox control, for example, that shows the results in calculation order.
This to me looks like you need to implement the Observer pattern, but distributed. Whereby new calculations are made to the service, and WCF just happens to be the mechanism by which you push your notification back to the client.
Generally speaking, you have your business logic housed in a windows service, whereby a type is a Subject (Observable). You could publish an endpoint for clients to register for notifications. This would be a WCF service, with potentially two operations:
RegisterClient(...)
UnregisterClient(...)
When a client is registered with service, it can receive updates, broadly speaking, the when the service has finished calculating a result, it could iterate through all registered clients and initiate a push. The push being a communication through an endpoint on the client.
A client endpoint might typically by
Notify(Result...);
And your server simply calls that when it has new data...
Typically you'd use TCP to maximise throughput.
This is by no means exactly what you should do, but perhaps its a direction to start in?
Related
I'm at the investigative stage for Workflow Service and WPF.
Having State Machine WF Service hosted in IIS and one or more WPF clients talking to the WF Service sounds reasonable choice so far.
However, although days of reading and research it isn't clear to me what would be the best strategy for tracking the transfer between states from WPF app.
There're numerous samples of tracking participants but most of them are based on One process scenario.
So I am thinking of a structure as below.
A server-side WCF operation that any client calls to register its client-side endpoint
A custom tracking participant that goes through all registered client-side endpoints and sends TrackingRecord at it's Track() method.
Advantage of this approach is that it allows real time update of the states without extra layers like ETW. Another advantage is that it allows decoupling of the logic (or maybe model layer) from the presentation layer.
Can anyone share the opinion over the above structure?
I would also welcome any suggestions for achieving the goal.
[EDIT]
To make my idea above more detailed and clear, below steps would be a typical usage.
1) (WPF client) contains and opens a WCF endpoint for receiving TrackRecords.
2) (WF Service) opens a WCF operation (or a simple WF instance with a Receive message) that registers client-side address to an internal store.
3) (WF Service) a custom tracking participant is created and added that will send TrackingRecord to the registered clients' endpoints.
4) (client) connects to the above service and hands out client-side endpoint mentioned at step 1 and consequently receives TrackingRecords.
[ EDIT 2 ]
To put my goal in simple terms, I'd like to know
1) the most efficient way of tracking the StateMachine's state on a WF Service (IIS) + WPF or any types of client app through TrackingParticipant.
2) if my suggestion can be improved
Meanwhile, I have implemented this and works good so far. I also added MvvM Light framework's messaging feature at the client-side so that it propagate the received message to the models easily.
You might take a look at SignalR instead of trying to force WCF to become a pub/sub platform which is not it's strength. There is an example on my blog with the visual tracking example with the tracking participant split out from the tracking application so it's not all in one process. That blog also has links to two other blogs where similar things were done, but all using a messaging architecture more suitable to events like this.
http://panmanphil.wordpress.com/2012/11/05/slides-and-sample-from-the-chippewa-valley-code-camp/
There is an existing mechanism that wraps a lot of the functionality that you are suggesting (If I understand your needs correctly). If you need to utilize a WCF service to communicate in a bidirectional way (i.e. PUSH data to connected clients) I would suggest leveraging the PollingDuplex Binding.
I have used PollingDuplex in the past with various Silverlight clients to exchange data, and I have read articles like this one describing the steps to produce the same behavior in WPF space.
This approach will automate much of the endpoint registration and tracking logic that you apparently are thinking of doing manually.
I hope this helps.
We have a number of different old school client-server C# WinForm client-side apps that are essentially front-ends for the database. Then there is a C# server-side windows service that waits on the client apps to submit orders and then it processes them.
The way the server-side service finds out whether there is work to do is that it polls the database. Over the years the logic of polling for waiting orders has gotten a lot more complicated due to the myriad of business rules. So because of this, the polling stored proc itself uses quite a bit of SQL Server resources even if there is nothing to do. Add to this the requirement that the orders be processed the moment they are submitted and you got yourself a performance problem, as the database is being polled constantly.
The setup actually works fine right now, but the load is about to go through the roof and, it is obvious, that it won't hold up.
What are some effective ways to communicate between a bunch of different client-side apps and a server-side windows service, that will be more future-proof than the current method?
The database server is SQL Server 2005. I can probably get the powers that be to pony up for latest SQL Server if it really comes to that, but I'd rather not fight that battle.
There are numerous options ways you can notify the clients.
You can use a ready-made solution like NServiceBus, to publish information from the server to the clients or other servers. NServiceBus uses MSMQ to publish one message to multiple subscribers in a very easy and durable way.
You can use MSMQ or another queuing product to publish messages from the server that will be delivered to the clients.
You can host a WCF service on the Windows service and connect to it from each client using a Duplex channel. Each time there is a change the service will notify the appropriate clients or even all of them. This is more complex to code but also much more flexible. You could probably send enough information back to the clients that they wouldn't need to poll the database at all.
You can have the service broadcast a UDP packet to all clients to notify them there are changes they need to pull. You can probably add enough information in the packet to allow the clients to decide whether they need to pull data from the server or not. This is a very lightweight for the server and the network, but it assumes that all clients are in the same LAN.
Perhaps you can leverage SqlDependency to receive notifications only when the data actually changes.
You can use any messaging middleware like MSMQ, JMS or TIBCO to communicate between your client and the service.
By far the easiest, and most likely the cheapest, answer is to simply buy a bigger server.
Barring that, you are in for a development effort that has a high probability of early failure. By failure I don't mean that you end up scraping whatever it is you end up building. Rather, I mean you launch the changes and orders will be screwed up while you are debugging your myriad of business rules.
Quite frankly, I wouldn't consider approaching a communications change under pressure; presuming your statement about load going "through the roof" in the near term.
If your risk exposure is such that it has to be 100% functional day one (which is normal when you are expecting a large increase in orders), with no hiccups then just upsize the DB server. Heck, I wouldn't even install the latest sql server on it. Instead, just buy a larger machine, install the exact same OS and DB server (and patch levels) and move your database.
Then look at your architecture to determine what needs to go away and what can be salvaged.
If everybody connects to SQL Server then there is also the option of Service Broker. Unlike other messaging/queueing solution recommended so far it is entirely contained in your database (no separate product to deploy, administer and configure), it offers a single story vis-a-vis your backup/recovery and high availability needs ( no separate backup for message store, no separate DR/HA, whatever is your DB solution is also your messaging solution) and overs a uniform programming API (SQL).
Even when everything is within one single SQL Server instance (ie. there is no need to communicate over network between multiple SQL Service instances) Service Broker still has an ace that no one can match: activation. With activation you eliminate completely the need to poll because the system itself will launch your processing code (will 'activate') when there are events to process. The processing code can be internal (T-SQL procedure or SQLCLR .Net procedure) or external (see external activator).
This is the scenario.
I have multiple clients on our application, and one server.
The server is itself disconnected from the clients, it just downloads some data from the web via a windows service (web services and FTP), processes the data and updates a database to which all the clients are connected and draw data from.
I would like to be able to actively notify the clients, and with a certain degree of granularity, when some downloading occurs (i.e. only the Traders when a price/trade update occurs, or only the Engineers when there's something for them) without polling.
The server should fire up a notification to all the connected clients instead of having them continuously "ask" if there is an update, because in this case I would have to maintain state on all the clients.
I thought about XMPP, with Matrix.
To do so each client has to open a persistent connection with the windows service, but I lack the exact details on how to implement this. MAybe with nodes!
For what I understand XMPP is perfect for what I want to accomplish and gives me the extensibility to grow to some more functionality if I have the need to.
I don't know if to implement my own server or use one of the existing one (I hear jabberd2 has an excellent windows server).
But most important: I need suggestions on A) an XMPP server to run on Windows and B) a C# library. Besides Matrix I have found very few, and above all I need notifications support (pubsub).
For simplicity, I'd consider using a WCF service that implements a long polling technique. This article gives some details on scaling the WCF service efficiently.
For notifications that there is new data in the database, if you are using SQL Server, try SqlDependency. It allows you to set up an event that fires in your code whenever the result of a given query changes. I've used it effectively for just this sort of thing.
I have developed a windows service which reads data from a database, the database is populated via a ASP.net MVC application.
I have a requirement to make the service re-load the data in memory by issuing a select query to the database. This re-load will be triggered by the web app. I have thought of a few ways to accomplish this e.g. Remoting, MSMQ, or simply making the service listen on a socket for the reload command.
I am just looking for suggestions as to what would be the best approach to this.
How reliable does the notification has to be? If a notification is lost (lets say the communication pipe has a hickup in a router and drops the socket), will the world end come or is business as usual? If the service is down, do notifications from the web site ned to be queued up for when it starts up, or they can e safely dropped?
The more reliable you need it to be, the more you have to go toward a queued solution (MSMQ). If reliability is not an issue, then you can choose from the mirirad of non-queued solutions (remoting, TCP, UDP broadcast, HTTP call etc).
Do you care at all about security? Do you fear an attacker my ping your 'refresh' to death, causing at least a DoS if not worse? Do you want to authenticate the web site making the 'refresh' call? Do you need privacy of the notifications (ie. encryption)? UDP is more difficult to secure (no session).
Does the solution has to allow for easy deployment, configuration and management on the field (ie. is a standalone, packaged, product) or is a one time deployment that can be fixed 'just-in-time' if something changes?
Withous knowing the details of all these factors, is dififcult to say 'use X'. At least one thing is sure: remoting is sort of obsolete by now.
My recommendation would be to use WCF, because of the ease of changing bindings on-the-fly, so you can test various configurations (TCP, net pipe, http) w/o any code change.
BTW, have you considered using Query Notifications to detect data changes, instead of active notifications from the web site? I reckon this is a shot in the dark, but equivalent active cache support exists on many databases.
Simply host a WCF service inside the Windows Service. You can use netTcpBinding for the binding, which will use binary over TCP/IP. This will be much simpler than sockets, yet easier to develop and maintain.
I'd use standard TCP sockets - this will survive all sorts of moving of components, and minimize configuration issues IMHO.
I'm writing a simple accounting program consists of several C# winform clients and a java server app that read/write data into a database. One of the requirement is that all C# clients should receive updates from the server. For example, if user a create a new invoice from his C# client, other users should see this new invoice from their client.
My experience is mainly on web development and I don't know what's the best way to fulfill this requirement with C#s client and Java servlet server.
My initial though is to run ActiveMQ with Glassfish and use messaging pub/sub method so that updates can be pushed to C# client. I will create different topics like newInvoice, cancelInvoice, etc in order to differentiate the message type. Each message will simply contains the object encoded in JSON.
But it seems to me that this involves quite a lot of work. Given that my user base is very small ( just 3 or 4 concurrent user), it seems to me that there should be some simpler solutions. (I'm not familiar socket programming :) )
I know this is a client-server programming 101 questions but would be great if any experienced programmer can point me to some simple solutions.
The simplest approach here is often to simply use a poll - i.e. have the clients query for data every (your time interval). That avoids a whole family of issues (firewalls, security, line-of-sight, resolution, client-tracking, etc).
With WCF, you can have callbacks on duplex channels (allowing the server to actively send a message to clients), but this is more complex. I value simplicity, so I usually just poll.
Tricks that help here are designing the system to have an inbuilt mechanism for querying "changes since x" - for example, an audit table, perhaps fed by database triggers. The exact details vary per project, of course.
Another option that you might want to look at is ADO.NET Sync Services; this does much of what you ask for, for keeping a local copy of the database up to date with the server - but has a few complexities of its own. This is available (IIRC) in the "Local Database Cache" VS template.
Rather than pushing information from the server to 1:N Clients, would it not be easier to have the clients Poll the server for updates every so often ? Or when the client launches and creates a connection to the server, the server could dynamically generate a new Message Queue for that Client Connection, which the client could then poll for updates?
There are several push technologies available to you, like ActiveMQ (as you mentioned), or XMPP. But if you only have 3 or 4 clients to concern yourself with, polling would be the simplest solution. It doesn't scale well, but that isn't really a concern in your case, unless your server is an 8086 or something 8-)
You may want to take a look at StreamHub Push Server - its a popular Comet server written in Java that has a .NET Client SDK for receiving updates from the server in C#. It also has a Java Client SDK and the usual Ajax/Comet web browser support giving you more flexibility in the future to push data to web, Java and C# clients.