Consider an Access Point with a packet queue that can hold up to 50 packets and a network whose bottleneck is a link with a datarate of 50 Mbit/s. There are quite a few nodes connecting wirelessly to the Access Point, one of them uploads a file (1.5 GB) via FTP to a remote server. … Read more
The scenario that I have is as such: I have a loopback TCP session and I want to improve its performance. I have my socket-like objects (System.Net.Sockets in .NET Framework, to be specific) sending messages of up to about 9MB each flowing in both directions of the TCP connection. I have observed typical delays of … Read more
Closed. This question is off-topic. It is not currently accepting answers. Want to improve this question? Update the question so it’s on-topic for Server Fault. Closed last year. Improve this question According to this website, as soon as round-trip time is increased, we lose a bit of our real-time capacity throughput to distant servers using … Read more
We’ve been trying to debug very slow transfer rates for remote clients hitting our web server (IIS on 2k12 R2). The RTT to the client is about 150ms, and we’re struggling to get more than about 500kbps out of our 100Mbps link. The client can speed-test download over 300MBit/s to a local speed test server, … Read more
I’ve got 2 windows 2012 servers that limit connection speed to anywhere from 1-3mBps and it seems to be related to tcp window scaling. Both servers have these settings TCP Global Parameters ———————————————- Receive-Side Scaling State : enabled Chimney Offload State : disabled NetDMA State : disabled Direct Cache Access (DCA) : disabled Receive Window … Read more
I have a typical long fat network — clients connect to servers with large bandwidth but also significant latency due to geographic distances of 30ms to 100ms, transferring large files. I’ve noticed the classic decay of throughput ~ which decays as 1 / latency. Typical throughputs based on tests are well below the theoretical bandwidths … Read more
I’m trying to debug a low tcp transfer rate on a host (“lowHost”). As a reference, i’m comparing it against a second host (“highHost”). I’m making two rate measurements: Downloading a big file using curl from a webserver. Using iperf lowHost goes through an extra network hop (“router”) compared to highHost. lowHost —(router)—–>(fw)—-> internet —-> … Read more
I think I may be having an issue with window scaling (RFC 1323) and am hoping that someone can enlighten me on what’s going on. Server: FreeBSD 9, apache22, serving a static 100MB zip file. 192.168.18.30 Client: Mac OS X 10.6, Firefox 192.168.17.47 Network: Only a switch between them – the subnet is 192.168.16/22 (In … Read more
I asked a similar question before, but I believe I was not clear with the details. I’m basically having some problems with my server related to high latency between the server and the clients due to the geographical distance. And making a research, it turned out that I can solve this issue by increasing the … Read more
I’ve tried changing the tcp’s window size by modifying net.ipv4.tcp_wmem and net.ipv4.tcp_rmem values, however the changes which I make have no effect on actual window size which I can check using a wireshark. I tried to do that on 2 different kernels (3.2.69 and 2.6.38) with no effect at all. I’ve even tried to change … Read more
