Multiple TCP connections vs single connection - c++

I am designing a Data Distributor (say generating random number) where it will serve multiple clients.
The client C first sends the list of numbers in which it is interested to DD over TCP and listens for data on UDP. After some time (few minutes) the client may renew its subscription list by sending more numbers to DD.
I can design this in 2 ways.
FIRST:
New_Client_Connected_Thread(int sock_fd)
{
--Get Subscription
--Add to UDP Publisher List
--close(sock_fd)
}
Everytime client wants to subscribe to new set of data it will establish a new TCP connection.
SECOND:
New_Client_Connected_Thread(int sock_fd)
{
while(true)
{
--wait for new subscription list
--Get subscription
--Add to UDP Publisher List.
}
}
Here only 1 TCP connection per client would be required.
However if the client does not send new request, the Client_Thread would be waiting unnecessarily for long time.
Given that my Data Distributor would be serving lots of clients which of them seems to be the efficient way?

Libevent, or libev, which supports an event driven loop is probably more appropriate for the TCP portion of this.
You can avoid the threading, and have a single loop for the TCP portion to add your clients to the Publishers' list. Libevent is very efficient at managing lots and lots of connections and socket teardowns per second and is used by things like Tor (The onion router)
It seems like the TCP connection in your application is more of a "Control Plane" connection, and thus it's probably going to depend on how often your clients need to "control" your server thats going to make the decision whether to leave the socket open or close it after controlling. Keep in mind that keeping thousands of TCP connections open permanently does it kernel resource on the host, but on the other opening and closing connections the whole time introduces some latency due to the connection setup time.
See https://github.com/libevent/libevent/blob/master/sample/hello-world.c for an example of a libevent TCP server.
Since you're coding in C++, you will probably interested in the http://llucax.com.ar/proj/eventxx/ wrapper for libevent

Related

Maximum number of TCP connections

I am doing a TCP client - server simulation. In the simulation, I have created 2 clients and 2 servers. And I have programmed that read requests will go to server 1 and write requests will go to server 2. Thus, the client will always renew it's socket and make a new connection to the servers.
However, after the client has made 66561 times of connections to the server, instead of sending request packets, it will just simply send some empty ACK packets.
I expected both the clients to be able to send up to millions of requests, but currently, both the clients are only able to send up to 13k requests. Can anyone give me tips or advices?
Nagle's algorithm
Solutions:
Do not use small package in your app protocol
Use socket option TCP_NODELAY on both side client/server
Sounds like most previously created connections are still taking the resource (not released from system). From the information you give,
However, after the client has made 66561 times of connections to the server, instead of sending request packets, it will just simply send some empty ACK packets.
Looks like about 1000+ connections are released. Probably because of the 2msl time is due. If this is the case, suggest you explicitly release a connect before you create a new one.
Copy and paste your C/S part code would help the analyse.

TCP push-pull socket server design

I am designing a cross-platform messaging service as a learning exercise. I have programmed socket-based servers before, but always a "client-polls-server" design, like a web server. I want to be able to target mobile platforms, and I read that polling is a battery drain, so I would like to do push notification.
The server will be TCP-based, written in C++. What I'm having trouble getting my head around is how to manage the bi-directional nature of the design. I need a client to be able to send packets to the server as normal, but also listen for packets. How do I mitigate situations like, the client is sending data when the server is trying to send to it, or it's blocked listening for data but then needs to send something?
For example, consider the following crude diagram:
So, let's say client A is in the middle of sending a chunk of data (arrow 1). While this is happening, client B sends a message (arrow 2), which causes the server to attempt to send data back to client A (arrow 3), but client A hasn't finished sending arrow 1 yet. What happens in this instance? Should I setup 2 separate ports on each client, one for inbound, one for outbound? Do I need to keep track of the state of each connection?
Or is there a better approach to this altogether?
One socket port is inherently bidirectional. To handle both inbound and outbound traffic more or less concurrently you need to use nonblocking sockets.
I think the solution is pretty simple. The TCP server should have a list with connected clients. Since a TCP connection is bi-directional, the push mechanism is quite simple.
Another important thing, as long as your server isn't multithreaded, you can read from or write to one client at the same time.

Send same packets to multiple clients

I have to develop a software to send same packets to multiple destination.
But i must not use multicast scheme.!!!! ( because my boss is a stupid man )
so, any way, the problem is that:
i have same packets and multiple IP address ( clients) and i can not use multicast
how can i do that in the best way?
i must use c++ as a language and Linux as a platform.
so please help me
Thanx
If your boss said you can't use multicast, maybe he/she has his/her reason. I guess broadcasting is out of the game too?
If these are the requisites, your only chance is to establish a TCP connection with every remote host you want to send packet to.
EDIT
UDP, conversely, would not provide much benefit over multicasting if your application will run over a LAN you are in charge for configuration of, that's the reason I specified TCP.
Maybe you have to describe your scenario a little better.
This could be done with either TCP or UDP depending on your reliability requirements. Can you tolerate lost or reordered packets? Are you prepared to handle timeouts and retransmission? If both answers are "yes", pick UDP. Otherwise stay with TCP. Then:
TCP case. Instead of single multicast UDP socket you would have a number of TCP sockets, one per destination. You will have to figure out the best scheme for connection establishment. Regular listening and accepting connecting clients works as usual. Then you just iterate over connected sockets and send your data to each one.
UDP case. This could be done with single UDP socket on the server side. If you know the IPs and ports of the clients (data receivers) use sendto(2) on the same data for each address/port. The clients would have to be recv(2)-ing at that time. If you don't know your clients upfront you'd need to devise a scheme for clients to request the data, or just register with the server. That's where recvfrom(2) is usefull - it gives you the address of the client.
You have restricted yourself by saying no to multicast. I guess sending packets to multiple clients is just a part of your requirement and unless you throw more light, it will be difficult to provide a complete solution.
Are you expecting two way communication between the client and the server ? in that case choosing multicast may prove complex. please clarify
You have to iterate through the clients and send packets one after another. You may want to persist the sessions if you are expecting response from the clients back.
Choice of UDP or TCP again depends on the nature of data being sent. with UDP you would need to handle out of sequence packets and also need to implement re-transmission.
You'll have to create a TCP Listerner on your server running at a particular port listening for incoming Tcp Client connections (Sockets).
Every time a client connects, you'll have to cache it in some kind of datastructre like a Name value pair (name being a unique name for the client amd value being the Network Stream of that client obtained as a result of the TCP socket).
Then when you are finally ready to transmit the data you could either iterate through this collection of name value pair connections and send them data as byte array one by one to each client or spawm off one thread per connected client and have it send the data concurrently.
TCP is a bulky protocol (due to its connection-oriented nature) and transmission of large data (like videos/images) can be quite slow.
UDP is definitely the choice for streaming large data packets but you'll have to trade-off with the delivery gurantee.

TCP/IP and designing networking application

i'm reading about way to implemnt client-server in the most efficient manner, and i bumped into that link :
http://msdn.microsoft.com/en-us/library/ms740550(VS.85).aspx
saying :
"Concurrent connections should not exceed two, except in special purpose applications. Exceeding two concurrent connections results in wasted resources. A good rule is to have up to four short lived connections, or two persistent connections per destination "
i can't quite get what they mean by 2... and what do they mean by persistent?
let's say i have a server who listens to many clients , whom suppose to do some work with the server, how can i keep just 2 connections open ?
what's the best way to implement it anyway ? i read a little about completion port , but couldn't find a good examples of code, or at least a decent explanation.
thanks
Did you read the last sentence:
A good rule is to have up to four
short lived connections, or two
persistent connections per
destination.
Hard to say from the article, but by destination I think they mean client. This isn't a very good article.
A persistent connection is where a client connects to the server and then performs all its actions without ever dropping the connection. Even if the client has periods of time when it does not need the server, it maintains its connection to the server ready for when it might need it again.
A short lived connection would be one where the client connects, performs its action and then disconnects. If it needs more help from the server it would re-connect to the server and perform another single action.
As the server implementing the listening end of the connection, you can set options in the listening TCP/IP socket to limit the number of connections that will be held at the socket level and decide how many of those connections you wish to accept - this would allow you to accept 2 persistent connections or 4 short lived connections as required.
What they mean by, "persistent," is a connection that is opened, and then held open. It's pretty common problem to determine whether it's more expensive to tie up resources with an "always on" connection, or suffer the overhead of opening and closing a connection every time you need it.
It may be worth taking a step back, though.
If you have a server that has to listen for requests from a bunch of clients, you may have a perfect use case for a message-based architecture. If you use tightly-coupled connections like those made with TCP/IP, your clients and servers are going to have to know a lot about each other, and you're going to have to write a lot of low-level connection code.
Under a message-based architecture, your clients could place messages on a queue. The server could then monitor that queue. It could take messages off the queue, perform work, and place the responses back on the queue, where the clients could pick them up.
With such a design, the clients and servers wouldn't have to know anything about each other. As long as they could place properly-formed messages on the queue, and connect to the queue, they could be implemented in totally different languages, and run on different OS's.
Messaging-oriented-middleware like Apache ActiveMQ and Weblogic offer API's you could use from C++ to manage and use queues, and other messaging objects. ActiveMQ is open source, and Weblogic is sold by Oracle (who bought BEA). There are many other great messaging servers out there, so use these as examples, to get you started, if messaging sounds like it's worth exploring.
I think key words are "per destination". Single tcp connection tries to accelerate up to available bandwidth. So if you allow more connections to same destination, they have to share same bandwidth.
This means that each transfer will be slower than it could be and server has to allocate more resources for longer time - data structures for each connection.
Because establishing tcp connection is "time consuming", it makes sense to allow establish second connection in time when you are serving first one, so they are overlapping each other. for short connections setup time could be same as for serving the connection itself (see poor performance example), so more connections are needed for filling all bandwidth effectively.
(sorry I cannot post hyperlinks yet)
here msdn.microsoft.com/en-us/library/ms738559%28VS.85%29.aspx you can see, what is poor performance.
here msdn.microsoft.com/en-us/magazine/cc300760.aspx is some example of threaded server what performs reasonably well.
you can limit number of open connections by limiting number of accept() calls. you can limit number of connections from same source just by canceling connection when you find out, that you allready have more then two connections from this location (just count them).
For example SMTP works in similar way. When there are too many connections, it returns 4xx code and closes your connection.
Also see this question:
What is the best epoll/kqueue/select equvalient on Windows?

What is the best way to implement a heartbeat in C++ to check for socket connectivity?

Hey gang. I have just written a client and server in C++ using sys/socket. I need to handle a situation where the client is still active but the server is down. One suggested way to do this is to use a heartbeat to periodically assert connectivity. And if there is none to try to reconnect every X seconds for Y period of time, and then to time out.
Is this "heartbeat" the best way to check for connectivity?
The socket I am using might have information on it, is there a way to check that there is a connection without messing with the buffer?
If you're using TCP sockets over an IP network, you can use the TCP protocol's keepalive feature, which will periodically check the socket to make sure the other end is still there. (This also has the advantage of keeping the forwarding record for your socket valid in any NAT routers between your client and your server.)
Here's a TCP keepalive overview which outlines some of the reasons you might want to use TCP keepalive; this Linux-specific HOWTO describes how to configure your socket to use TCP keepalive at runtime.
It looks like you can enable TCP keepalive in Windows sockets by setting SIO_KEEPALIVE_VALS using the WSAIoctl() function.
If you're using UDP sockets over IP you'll need to build your own heartbeat into your protocol.
Yes, this heartbeat is the best way. You'll have to build it into the protocol the server and client use to communicate.
The simplest solution is to have the client send data periodically and the server close the connection if it hasn't received any data from the client in a particular period of time. This works perfectly for query/response protocols where the client sends queries and the server sends responses.
For example, you can use the following scheme:
The server responds to every query. If the server does not receive a query for two minutes, it closes the connection.
The client sends queries and keeps the connection open after each one.
If the client has not send a query for one minute, it sends an "are you there" query. The server responds with "yes I am". This resets the server's two minutes timer and confirms to the client that the connection is still available.
It may be simpler to just have the client close the connection if it hasn't needed to send a query for the past minute. Since all operations are initiated by the client, it can always just open a new connection if it needs to perform a new operation. That reduces it to just this:
The server closes the connection if it hasn't received a query in two minutes.
The client closes the connection if it hasn't needed to send a query in one minute.
However, this doesn't assure the client that the server is present and ready to accept a query at all times. If you need this capability, you will have to implement an "are you there" "yes I am" query/response into your protocol.
If the other side has gone away (i.e. the process has died, the machine has gone down, etc.), attempting to receive data from the socket should result in an error. However if the other side is merely hung, the socket will remain open. In this case, having a heartbeat is useful. Make sure that whatever protocol you are using (on top of TCP) supports some kind of "do-nothing" request or packet - each side can use this to keep track of the last time they received something from the other side, and can then close the connection if too much time elapses between packets.
Note that this is assuming you're using TCP/IP. If you're using UDP, then that's a whole other kettle of fish, since it's connectionless.
Ok, I don't know what your program does or anything, so maybe this isn't feasible, but I suggest that you avoid trying to always keep the socket open. It should only be open when you are using it, and should be closed when you are not.
If you are between reads and writes waiting on user input, close the socket. Design your client/server protocol (assuming you're doing this by hand and not using any standard protocols like http and/or SOAP) to handle this.
Sockets will error if the connection is dropped; write your program such that you don't lose any information in the case of such an error during a write to the socket and that you don't gain any information in the case of an error during a read from the socket. Transactionality and atomicity should be rolled into your client/server protocol (again, assuming you're designing it yourself).
maybe this will help you, TCP Keepalive HOWTO
or this SO_SOCKET