Samba Beats Windows IT Week Labs Test Results

jmhowitt writes "Tests by IT Week Labs show the latest version of the open-source Samba file and print server software is 2.5 times faster than Windows Server 2003 in the same role. The news comes as many firms are grappling with the consequences of Microsoft ending support for NT4, coupled with uncertainty about when Microsoft will next update Windows. The performance difference between Windows Server 2003 and Samba 3 has increased dramatically compared with Samba 2 and Windows 2000 Server."

Comments

Best choice for the job?

[Sheetrock]

I've been using Samba for awhile, and despite some config difficulties it performed as advertised.

However, even if it's quicker than Windows Server 2003, NFS still seems to do a great deal better on my home network for the same things. For example, I typically get 10%-20% of the transfer with SMB as I do with NFS.

So I don't recommend using Samba at all unless you're looking for Windows compatibility.

Re:Best choice for the job?

[curious.corn]

NFS lives in the kernel, Samba in user space. So you're right but remember NFS is utterly insecurable, Samba not. For home NFS is the system of choice but in a larger environment... you want to run Samba (at least until NFSv4 becomes available)

Re:Best choice for the job?

[Make]

you mean, there is a kernel implementation of the NFS daemon. There is also one which runs purley in userspace, but on Linux, it is not used very much anymore nowadays.

There is smbclient in userspace (which is similar to an FTP client), but if you want to mount an SMB share into the linux VFS, you need the kernel module - like you need the NFS kernel module if you want to mount an NFS filesystem.

Re:Best choice for the job?

[__past__]

Excuse me?! FTP is an absolutely braindead protocol from todays point of view - even if you find an interoperable solution to get rid of the plain-text passwords, the multiple-tcp-connections design is a fucking pain for people who have to configure packet filters to make it work. The most popular FTP servers, like WU-FTP or ProFTPD are about as secure a code base as BIND or sendmail. If it were for me, FTP should take its friend telnet and get the fuck off the net, joining finger and rlogin in the nirvana of net services.

SFTP is a different matter however, but it's less an extension of FTP as an add-on to SSH to implement similar functionality in a completly different way. Not bad as a protocol, but it suffers from the lack of a robust SSH implementation.

Re:Best choice for the job?

[GOD_ALMIGHTY]

If it were for me, FTP should take its friend telnet and get the fuck off the net, joining finger and rlogin in the nirvana of net services.

Somehow I don't think these protocols were enlightened enough to reach Nirvana. I'd say that they are being reincarnated as bloated SOAP specifications, which will cause us to lose an entire CPU generations' gain in power to transform XML cause some developers were too lazy to learn CORBA.

Panther has it in Apple's Open Directory !

[mirko]

See article...

The best thing about Samba...

[_Hellfire_]

Apart from how bloody quick it is is the fact that you can log every transaction. This is immensely useful in a mission critical environment when you have to figure our exactly why one person in particular out of the entire network is having trouble. Check your Samba logs and 99.9% of the time your answer will be there.

As a system administrator I appreciate having that level of scrutiny on any network I take care of.

Re:The best thing about Samba...

[jschrod]

OK, flamebait, and not funny.

I have worked with several large organizations that use Samba to serve files to 10,000s of workstations. It works much better than W2K servers, not to speak of NT servers. Samba on a Sun HA cluster (e.g., F15K systems) is an appropriate choice for file sharing that is considered critical.

And please note that I don't say this because I'm anti-Microsoft. In fact, I'm the CEO of a company that is a MS partner, and I am very satisfied with MS' support for our company. (It's much better than that of most other proprietary vendors, though not in the league of Oracle and their ilk.) I regularily plan and deploy heterogenous infrastructures for very large installations (i.e., > 50,000 users) and can back up my opinion with real-world experience from several places. What are your credentials?

Nice advertising

[BiggerIsBetter]

Now where are the numbers to back it up?

Re:Nice advertising

[AstroDrabb]

Sadly the MS Empire does not allow you to release benchmark stats for their products. You agree to this when you use their products through their EULA. I am sure IT Labs doesn't want to get crap over it. Then agian, maybe they will find a way to post the numbers.

Knowledge of the protocol

[MadX]

I read a while ago about some of the SAMBA developers having a better grasp of how the services / protocol all tie together, than the M$ employees doing the development. Most of the current M$ team inherited code from the older versions of the OS, and they are merely building on top of this codebase. The SAMBA team have had to reverse engineer the protocol. So it seems to make sense therefore, that should you understand it better, you can sqeeze more out of the service on the whole. It therefore appears that it can only get better and better as they develop ..
I also don't know how many developers are on the samba team in total (contributors / developers), but I would almost start assuming more than the manpower assigned by M$ to this area of code for Windows .. And with it being opensource, bugs are easier to find ...

Re:Knowledge of the protocol

[NewbieProgrammerMan]

Good points. Here's an additional one: the Samba team doesn't have PHBs to get in the way. In my limited experience, if you're given an existing codebase and told to improve on it, that's exactly what you're expected to do - and it's all you're expected to do. You can't discover that "wow, this legacy code is crap," throw the offending chunks away and write something that works correctly and is more stable and/or secure.

The Samba team has complete freedom with their code, while the Microsoft developers do not.

Re:Knowledge of the protocol

[Zathrus]

You can't discover that "wow, this legacy code is crap," throw the offending chunks away and write something that works correctly and is more stable and/or secure.

That's because there are tradeoffs in everything... if you've been told to "clean up the codebase", take a bit to look at the codebase, and tell your manager that it's going to take X amount of time to do that the manager has to decide whether or not it's worth the time to do so -- since otherwise your time could be spent doing other things. And odds are the cleaning up isn't going to show an immediate return to the company. Of course, there are other plusses to cleaning up code -- like doing it right may mean that you can implement future features in less time -- but those are harder to quantify.

Any large project -- be it OSS or closed source -- has to deal with these issues in one way or another. Sure... in OSS anyone (in theory) can decide to go off and clean up the code base. But unless it's done with the input from the team then that effort may be for naught -- unless you're communicating structural changes then merging the two code bases may prove impossible (since new features/bugfixes will have diverged the codebase), the rest of the team may not feel comfortable with the new structure as they are with the old (which is part of a larger issue -- if anyone feels like they "own" parts of the code then they may get offended if you say it's crap and rewrite it entirely -- which is one reason why code ownership is bad), or other issues. If you do do it with the blessing of the team, it still has to be done in a reasonable amount of time for it to be worthwhile -- otherwise the code will either diverge too far or the project will stagnate while waiting on the rewrite.

And, of course, any time you rewrite you run the risk (read: certainty) that you'll introduce new bugs in known, working code.

Open source projects are freed from the time == money constraint if they have no commercial interests whatsoever, but that isn't to say that time becomes free. It's just that it's not necessarily an overriding factor. (Oh, and it's not one at all companies either -- that's entirely up to your manager and the structure of project management; but the more rigorous the framework of management the more likely it is be one).

Uh, where are the benchmarks?

[Fefe]

Where are the numbers?

Where are the graphs?

The article basically quotes some guy (who is actually selling Samba and thus has a vested interest) saying that Samba is 2.5 times faster than Windows 2003.

Now I have no reason not to believe him, but I was expecting a little more. And I'd wager the suits considering switching to Samba also expect more.

I just love the per client license fees

[hughk]

that Samba/Linux charge. The performance boost is just a nice plus.

A frequent rule in the Windows business is to split systems up over many machines. Which is great for Microsoft because essentially, you pay per client connection. With Linux/Samba, you pay according to the support that you want.

The really good thing in 3.0 was allowing the participation in ADS networks (Win 2K) as well as NT4.0. Domain controller support could be better for ADS, but otherwise it is fine.

Easy Way of Handling Printers

[fuzzybunny]

I serve printers from samba boxes to WinXP and W2k clients. I do not like dealing with setting up print queues on unix (unix printing and modem handling are evil, created by spawns of satan to make systems administrators miserable for all eternity), and I don't like Samba's way of dealing with them. It's still a bit too black-magic-swing-a-cat-over-your-head-at-midnight -y for my tastes when I need it to work in a hurry.

I've found CUPS to be a magnificent way of dealing with this; the combination of Samba, Unix, and WinXP/2k actually deals with printers very nicely over IPP.

The numbers....

[Bill_Mische]

There are a lot of people asking about the numbers.

Well, I get the print edition of IT Week and the numbers are there on page 19, in the form of a lovely little graph. The (print) article says they used a HP ProLiant BL10 eCLass Server (900MHz PIII, 40Gb ATA, 512Mb Ram) and goes in to a little detail about the benchmarking software used.

I couldn't see a copy of the article on their website but you can download an electronic copy, in some god-forsaken windows only ebook format, from www.itweek.co.uk/ebook.

Samba starter question?

[WuphonsReach]

I know this is more of a AskSlashdot question...

My impression of Linux/Unix systems has always been that each host has it's own set of user accounts and if I have 3 hosts it means that I have to maintain 3 sets of passwords. With NT4/Win2000, my servers share a common userspace so that you only have to maintain a single user account. Is there something under Linux/Unix that does this?

How easy is it to drop a Samba server into an existing Win2000 network? Our Novell 5 server is starting to show it's age (file/printing only) and I'm starting to wonder whether to move to a later version of Novell, switch to Linux/Samba, use a NAS device, or just load up another Win2000 server.

(With the security issues this year with Windows, however, I'm not sure I want to make Windows our main file server.)

We just decided to use Samba

[SailFly]

I'm a networking, sysadmin, programmer (mostly programmer) consultant for small businesses in Sarasota, Florida. Most of my customers are small businesses (less than 12 people) and are looking for ways to keep costs down.

After proposing a new 2.4GHz server with Win2k3, they were sticker shocked and decided to not hire me for the job. Then one of THEM mentioned Linux (which I love and hav used for 5 years). I told them that I use Linux in my software development practice, and we could consider this as an alternative for File Server (Samba), centralized security (ldap) and backups (Mandrake backup utility). We're also using VNC (realvnc.org) for remote desktop. I can also easily SSH and do remote X session from my office, or use VNC.

It's been up for a week now, and they LOVE IT! It's fast, flexible, and you cant beat the price. And I've learned my Lesson to be mention Linux even when they specifically ask for Windows (I'm not a pushy sales person, but I do believe an presenting choices to my customers)

They wanted to outsource their IT department (the owner doesn't ever want to worry or think about their IT issues), so we made a deal that allows me to keep their systems updated, but doesn't force him to hire an on-site IT person.

Speed was NOT an issue for the Samba server, since they mostly use MS Office (win xp pro workstations) documents. However, this was a great step for them to embrace and support open source software (I donate to several projects in turn).

I hope this story might help somebody who is considering doing something similar. I'm happy to answer any questions about our experiences.

-Scott James

Apples and oranges, Moriarty.

[Medievalist]

Samba 3 is a very different beast from 2.2, your remark is equivalent to basing an opinion of Windows 2K on your experiences with DOS 5.0.

The numbers are in the dead-tree edition, I'm told. I don't know if they actually show any real information, because I haven't seen them.

Samba had a 2x speed advantage over Windows NT 3.51 when that was the current MS offering, though, so I don't find this completely unbelievable.

Re:Excuse me...

[Frobnicator]

A 2.5 x speed advantage can only mean one thing: misconfigured Windows machines.

You might be interested in this article describing the implementations of SMB and Samba, as well as the documentation list at samba.org. Additionally, Microsoft has had many issues with the LMB, DMB, and backup browser implementations, partly due to historical network decisions and old networking bugs and highly visible on volitle networks. Some of the difficulties with SMB include methods of caching the data and cache consistancy, the inconsistant rate of refereshes done within the network, (on windows) the auto-detection and auto-creation of LMB and LMB-backup nodes, and the presumption of the underlying network.

I'll focus on the network cache consistancy problem since that's the one I've had problems with. I don't know about the general speed issue (what speed are you referring to? throughput? Resource availablility? Master Browser updates? connection speed and concurrency under a heavy user load? ) I have experienced all kinds of problems with a highly volitile network, with programmers running multiple OS's inside of virtual machines. These virtual OS's need to be frequently restarted, meaning the network is constantly gaining and droping objects.

A prime example of Microsoft's bad cache coherency problem is that if an object is deleted or removed from the network, the information can take over an hour to propogate through the entire network. The worst case isn't nearly as bad in the pure-Samba implementation, but the difficulty remains. This failure means that newly added resources aren't immediately visible on the network, or recently removed resources take a long time to be removed, and show up as errors when you try to access them. Or the object can be visible on some machines, but not available on others.

When there is a high level of volitility on the network (machines being frequently rebooted or shut down, network re-wiring, etc.) this can really plague any SMB or CIFS network, but is especially hard on Windows boxes, and more so the older your Windows implementation. Problems are exacerbated if either the LMB or LMB-backup system is the one going back up and down, because the Windows boxes will respond less-quickly to the problem; this results in further instability for the SMB network, since critical nodes are not available, propogate incorrect data, and take longer to reconfigure.

As you mentioned, the Samba boxes are faster than the Windows boxes, but not as big of a difference as you experience. You said you have "a LAN full of Win2000/XP boxes", which probably means they are on most or all of the time. Is it unreasonable to assume that the author has a more volitle network, or is otherwise more prone to speed impairment issues?

frob

Until M$ breaks compatibility.. then start over

[HighOrbit]

MicroSoft has a history of maintaining its monopoly by breaking compatibility with competitor's products by subtily changing (or they claim its extending and enhancing) the protocol. The most famous example were DrDOS and Java. If Samba gets too close, I wouldn't be suprised if MS didn't come up with an "enchancement" to Active Directory or SMB/CIFS or the NT-authentication protocols that will break Samba. The up-coming service pack will be the perfect oportunity for a "security fix" that will wall out Samba for a while.

(Related but slightly off-topic) A few days ago, there was an article about IE having broken support for standards, especailly CSS. I don't think that is an acident. I strongly suspect that MS won't fix IE because the "problem" helps them maintain a monopoly in browsers. If you want to get your stuff to render properly in 95% of people's browsers, you have to code to IE, not the "standard". This means your stuff won't render properly in the other 5% of browsers unless you go through lots of trouble to do browser dectection, alternate pages, or take lots of care for cross-browser compatibility.