Messages in this thread | | | Date | Wed, 20 May 1998 16:03:03 -0700 (PDT) | From | Natchu Vishnu Priya <> | Subject | Re: skb size |
| |
> On Thu, 21 May 1998, Oommen Thomas wrote: > > > Testing showed that this may be mainly due to the OS (kernel) limitations. > > > > UDP services like tftp/rcp as well as SMTP/UUCP give very good throughput, > > though. > > > > The network consultant is even suggesting of replacing the 'free' linux > > with a better & powerful 'real' unix with a 'full' commercial > > implementation of TCP, which will not drop packets on heavy load. > > The advice we received (which we find hard to buy) is that a free OS > > like Linux must not be used for Production servers as the kernel lacks lot > > of TCP functionalities. > > Any hints (to help save linux's as well as our prestige) will be greatly > > appreciated.
Ok.. There is no real problem in the implementation of TCP in linux.
I gather you are in india, and it looks like you have a line through vsnl I have done that before.. the vsnl link has very high loss rates of the order of 1 in 10-100 at peak time.
TCP is bound to suffer with such high loss rates, and any implementation ("conformant") will perform badly.
In fact linux 2.1.x (to my knowledge) is the only one that implements sack.
Even with 2.0.x you should be seeing much better performance than you would get with any other production OS.
Solution: Check out your packet drop rates.. complain to vsnl.. increase bandwidth on your leased line to vsnl.
vishnu
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
| |