similar to: Samba Variables and TCP/IP Throughput

Hello friends, I am trying to test NAS I/O performance over a network and trying to see the numbers for write and read speed. I have successfully configured and ran a lot of tests. However the numbers have not increased, I have reached a bottleneck. I tried playing around with the smb.conf file including all possible variations like, socket options = TCP_NODELAY SO_RCVBUF=(tried values like

I went to the site to subscribe again and ended up watching some of Jeremy's Google interviews. I particularly enjoyed the interview with James and the bow tie lesson at the end. :) So anyway, I recently upgraded my home network to end-to-end GbE. My clients are Windows XP SP3 w/hot fixes, and my Samba server is 3.5.6 atop vanilla kernel.org Linux 3.2.6 and Debian 6.0.6. With FDX fast

Yes, I get more than 30MB/s performance. The benchmark I use (NetBench) is essentially CPU bound, such that a faster processor = faster performance. With a very fast hardware config (dual 3.2GHz processors), I've been able to hit around 100MB/s. Changing the RAM or other attributes does not buy me much, it seems that processor power is the bottleneck (at least in my case). When doing your

Hi, may be there are some best practice suggestions for the "best mode" for channel bonding interface? Or in other words, when should/would I use which mode? E.g. I do have some fileservers connected to the users lan and to some ISCSI Storages. Or some Webservers only connected to the LAN. The switches are all new cisco models. I've read sone docs (1), (2) and (3) so the theory

Hi, Has anyone using CentOS 6 been able to successfully set an mtu larger than 9710 on an interface. I am seeing super jumbo frames with length > 10000. ... IP 10.79.4.53.64327 > 10.79.2.53.24294: Flags [.], seq 16060:29200, ack 1, win 32767, length 13140 ... CentOS release 6.7 (Final) Thanks, -- Stephen Clark *NetWolves Managed Services, LLC.* Director of Technology Phone:

Hi, I'm a happy Samba user. I've been working sucessfully in Linux for the past 6 months. I've built a Linux Server with very fast storage and I'm connecting it to many Windows XP "video editing workstations" via Gigabit Ethernet (all NICs using Jumbo Frames, by the way). Now I want to see if I can increase my data flow in and out of the Server so that more

I am working on setting up an NFS server, which will mainly serve files to web servers, and I want to setup two bonds. I have a question regarding *which* bonding mode to use. None of the documentation I have read suggests any mode is "better" than other with the exception of specific use cases (e.g. switch does not support 802.3ad, active-backup). Since my switch *does* support

Hello everyone, Quoting from Samba Team Blog #2 (25 Sept 2009): "Volker showed how to get more than 700MB/sec from Samba using smbclient and a modern Samba server, which shows what you can really do when you understand the protocol thoroughly and don't feel you have to invent a new one (SMB2 :-)." Would it be possible to get a complete accounting of how this was achieved? Thanks,

Does anyone know why, when transferring data via Gigabit Ethernet from OS X to Linux (or vice versa), you get about 50-60 MB/sec with Samba 3.0.X (all versions I've tried, the rate depending on whether you use Jumbo Frames) but you get about 105-110 MB/sec with Apple File Sharing Protocol and Netatalk 2.03? Is there any inherent reason why Samba goes at about half the speed? By the

Hello, are there any experiences with samba as a _really_ fast server? Assuming if the filesystem and network is fast enough, has anyone managed to get a throughput in samba of of let's say 300 MB/s ? Are there any benchmarks? regards, Martin

Hi! I do a rsync between 2 machines. The throughput is only 2 MByte/Sec. Each machine is a Supermicro server with 2 x 8 Core Opteron 6128 64 GByte of ECC RAM 1 LSI MegaRAID SAS 9280-24i4e 24 x 2TByte SATA Disks as a RAID6 2 Intel Corporation 82599EB 10-Gigabit SFI/SFP+ Network-cards Both run Ubuntu 11.04 64Bit. Both use rsync version 3.0.7 protocol version 30 There are no

I'm trying to test out an Equallogic PS5500 with a server running CentOS 4.7 I can create a volume and mount it fine using the standard iscsi-initiator-utils tools. The Equallogic box has 3 Gigabit interfaces and I would like to try to set up things so I can read/write from/to the volume using multiple NICs on the server i.e. get 200+ Mbyte/s access to the volume - I've had some

Hey folks, I had some general questions and when reading through the list archives I came across an iSCSI discussion back in February where a couple of individuals were going back and forth about drafting up a "best practices" doc and putting it into a wiki. Did that ever happen? And if so, where is it? Now my questions : We are not using iSCIS yet at work but I see a few places

I hope I'm in the right place for this question. I am in the process of creating a large scale production fileserver architecture for the purpose of serving very large files (~4.5GB) to a small number of clients (~200). My assumptions to this point have been that my throughput would be constrained by hardware and not software, but I would like some validation of this thinking. I am hoping

Hi, I am having some difficulties with matrix operations. It is a little hard to explain it so please bear with me. I have a very large data set, large enough that it needs to be split in parts in order to deal with. I can work things on these "parts" but the problem lies in adding together these parts for the final answer. So that been said, let's say that i split the data in 2

Hi all, I am trying to get some decent speed out of a vanilla Debian (Proxmox) box and am running out of ideas. Basically I can exchange data at 110MB/sec (raw, nfs, ftp) but not with Samba shares. I tried almost every samba tuning tip I could find to no avail. The only optimisation left to smb.conf at the moment is debug level = 0. As one can see the robocopy throughput is merely half of what

I have a Linux Server (3.06 Xeon) with a very fast RAID array -- reads at around 500 MB/sec as clocked by Bonnie++. I have 6 GigE nics on my machine -- on two 133Mhz PCI-x bus segments (not on the same bus as the RAID drives) I have noticed two puzzling things and I'm wondering if anybody has any ideas about why I'm seeing these: 1) transfer speeds over a single NIC from a single

>Yes, jumbo frames, no irq coalescence, blockio and see if >you can get Backup Exec to use large io request sizes when >reading and writing the data. The larger the better. Ok, Jumbo's enabled on the switch and media server. For the sake of our sanity jumping back and forth, I am trying to enable jumbo's on the bonded pair in the target, # ifconfig bond0 mtu 9014 and it errors

Hi, We are looking into replacing our current storage solution and are evaluating gluster for this purpose. Our current solution uses a SAN with two servers attached that serve samba and NFS 4. Clients connect to those servers using NFS or SMB. All users' home directories live on this server. I would like to have some insight in who else is using gluster for home directories for about 500

Greetings ... I hate to ask these questions, but I dought I will get a straight answer from an Salesmen ... I am looking at putting gigabit in as a back bone for a few Linux servers, but I have used an Accton Gigabit ethernet card with RedHat 8.0/9 and found it a little unstable ... do and RedHat users have a suggestion on a good, but not expensive Gigabit card, basicly for a Samba