Filers can start out so fast and end up so slow. How is that possible? A lot of companies buy a new filer and run a few performance tests against it, and they’re happy with what they see. The problem with that is that performance tests of a greenfield environment aren’t really valid. It’s how your filer performs once you fill it up with files that matters. Then once the performance goes down below the threshold that you deem unsatisfactory, how do you figure out what caused that problem and fix it?
Last month we did a webinar where my colleague George Crump discussed these questions with John Gentry from Virtual Instruments. It looks like there are a number of things that can create performance challenges for your filer, and they talked about each of them.
One thing is bottlenecks that come from metadata and the overhead of writing the metadata for each file. They also discussed the idea of an NFS or SMB client going rogue. A shared filer only works when everybody plays along. If every client uses just a little bit of the available throughput and IOPs of a filer, everything will be just fine. However, as soon as you have one client that uses a significant portion of the power of your filer, all bets are off. In a multitenant environment, a similar problem is the noisy neighbor issue. The question is how do you identify a rogue client or noisy neighbor?
Sometimes the performance problems that you perceive your filer to have actually don’t come from the filer itself. Sometimes they come from the VM and the hypervisor it is running on. If the VM is short on resources, then it can make it look like you have an NFS performance problem.
Of course, sometimes the problem is the filer itself. Perhaps the disk drives are not fast enough to keep up with the number of IOPs you’re generating. Perhaps the way you have laid out the volume is not the optimum method for performance. Perhaps the processor of the filer is busy doing things other than serving NFS and SMB requests, such as backups via NDMP. And finally, perhaps you are experiencing a bottleneck created by a single node of a cluster.
George and John discussed all of these issues at length, as well as how to determine which of these problems is causing the performance issue. If you missed the webinar, you can watch it on demand any time.
Watch On Demand
W. Curtis Preston (aka Mr. Backup) is an expert in backup & recovery systems; a space he has been working in since 1993. He has written three books on the subject, Backup & Recovery, Using SANs and NAS, and Unix Backup & Recovery. Mr. Preston is a writer and has spoken at hundreds of seminars and conferences around the world. Preston’s mission is to arm today’s IT managers with truly unbiased information about today’s storage industry and its products.
StorageSwiss.com - The Home of Storage Switzerland 
File servers are often full or close to full. This does not help.
Rather than adding (again) more disk space, have a look at Z-OptimiZr for Cloud Storage (that works just fine with file servers as well).
This software frees 35% to 45% of the disk space currently used on your file servers.
More details here: http://www.smallerfaster.com/z-optimizr-for-cloud-storage
Bad form. Chris. Linking directly to your product directly from an analyst’s blog (or anybody else’s blog for that matter) is bad form. Maybe if I had asked “Does anyone know of any products that do XYZ?” And then you said “We do XYZ, and here’s a link to our product!” THAT would be OK. But your product isn’t event germane to the topic! Bad form, Chris.
Have you ever heard the phrase “to a hammer everything looks like a nail?” That’s definitely the case with your post. You obviously read neither the post nor watched the webinar, because both were talking about performance issues not capacity issues (although I do understand that a severe capacity issue could cause performance issues). This was clearly just your attempt to link to your product from our blog, not an attempt to actually add to the discussion.
Having said that, I completely disagree with your assertion. The product being discussed in the webinar is a performance analysis tool. It helps people find out if they have any performance issues, and if so, where they are coming from. It’s often difficult to do that in network storage.
So I’d say yes, it does help. Your product may help as well, but installing your product as a first step seems ill advised. It’s a bit like giving a guy penicillin as soon as he walks into the doctor, and THEN asking him what’s wrong. Virtual Instrument’s product will find out what’s wrong. If they find out the solution is penicillin, then maybe the customer could give you a call.
Dear W. Curtis, Thanks for your prompt reaction to my post!
Yes as you said, severe capacity issue can cause performance issues, we see this everyday at our customers. But I never said or advised that Z-OptimiZr should be installed as a first step. This would have been, as you said, ill advised. I just suggested to have a look in case performance issues would find their origin in capacity issues.
Regarding inserting links in posts, I can understand your remark. But these days, readers need quick and practical solutions to their problems and are generally happy to find relevant links in posts, not dirty spam of course. Things are a bit less rigid than they used to be few years ago. Well, of course, this is IMO.
Now what prompted me to drop this quick comment here? Well, I saw a tweet about your post and since storage is a topic of interest for me (the reason why I follow storagesuiss on Twitter), I read this post.
Since there was no comments (15 days after your post was published), I told to myself let’s drop few words that could initiate a discussion thread here. The author will probably appreciate to get some interest, the potential readers may find additional relevant/connected information and of course a link to our web site is always good to take. Additional to this I was also thinking that it could be an opportunity to let you know about our solutions and maybe get some feedback/opportunity to exchange few words with an expert in our domain. (Hey, at least this one worked!) For me it was a kind of small – author – reader – me – win win win “deal”.
So this was not at all as you said, only “your attempt to link to your product from our blog, not an attempt to actually add to the discussion”.
Browsing the web, readers often find topics of interest by ricochet on other topics, as we all do. So, again IMO, it is OK to post comments that may not be 110% on the exact topic of a post as long as they are really connected to it. I would say this is also what makes the value of comments, to open the discussion to connected topics.
Thanks for comparing our solutions to penicillin.It is a very powerful and efficient medication,
but as you said, not to be taken upfront without diagnostic.
Be assured I enjoyed exchanging these few words with you.
In case you would be curious to know more about what we do, feel free to follow-up on my email address (I dropped it in the subscription form when posting my first comment).
Chris
I’ll stand by my original comment that your reply did not address the blog post or the webinar. It was way off topic.
Posting off topic comments is one thing. Posting off topic comments that are nothing more than ad for your product — especially on a site where you would typically pay for some kind of advertising — is not cool at all.
The fact that I have retweeted the storageswiss tweet linking to this post few minutes after writing my first comment just confirms that is was not like you say and that I personally genuinely though that linking (both ways) our solutions and your post and blog could be interesting for our respective readers.
This said I can only respect your opinion and feelings, but I also regret the hyper-focused attitude that finally tend to make experts knowing everything about nothing.