Posted by Rob Pelmas on Wed, Jun 03, 2009 @ 08:56 AM
We're a bunch of performance geeks here. We've been tweaking blocksizes, stripe, and interleave settings on disk since SGI first gave you access to 'em. Tuning and re-tuning SWAP size, location, type is in our blood. A few percentage points here, double digit gains there, all without more capex. Gotta love it.
Now, anytime a paradigm shift in technology comes out there's a steep cost differential to it, right? 10Gb networking had only a tiny little blip of time when it was out of reach of the masses, which is a refreshing change. You can kit out most servers with a card, an acceptable managed switch with a 10Gb port or two, for a very reasonably cost.
Why go to 10? Our desktops have had Gb cards for what seems like forever, and very fast CPUs. With just a couple 'power' users you could swamp the networking capabilities of a server. Of course, a handful of years ago disks could only cough up 150Mb/sec or so of sustained data, so network tended to not be the gating factor in server performance. Modern disk starts at well over 300Mb/sec, and if you stripe or otherwise use some common sense design principles you can achieve multiples of that.
Xinet and NAPC both use the 1 to 6 rule for users and performance: with 6 retouchers (or 'power' users), you can assume 1 of them will be accessing the server at one time. 12=2, 18=3. It's a rough rule of thumb, but one that seems to stand up over time. 12 heavy hitters can thus drain 120Mb/sec out of a server, which is the better part of 2 1Gb cards bonded together. Add in the other users, doing layout, OPI printing (yep, some folks still use an OPI workflow), and Portal access, you've got a saturated pipe. 10gb gives you a good 800mb/sec of access speed, which will sate all but the most demanding organizations needs for data.
Next of course, we can talk about teaming 10Gb interfaces! (insert evil chortle of delight here).
Posted by Brian Dolan on Wed, Apr 01, 2009 @ 10:27 AM
Being in the position that I’m in affords me the opportunity to travel
and see many of our clients around the country. I love working with
all of you and you all have your own unique personalities as well as
challenges in your respective environments. Something I want to share
with you all is a question I’m often asked, “How can I make Xinet do
more for us? I know it can do X, Y or Z but I only use it for (fill in
the blank) and want to do more with it!” Usually the next questions
is, “How do your other clients do it and how are they using Xinet?”
Well, let’s start with the second one. . .how DO other clients work
with Xinet. This is a hard one since all of you use it for different
reasons and have unique needs even though many of you are in the same
business.
So, how do I make it do the trick your asking? Get to it already would ya!
A couple of things to think about before I can suggest anything:
1)
Listen to your clients (internal and external)- You know your company
and how it ticks better than us. Yes, we at NAPC all have been around
the industry for a long time and bring plenty of knowledge to the table
but. . .you're the one hearing the conversations in meetings or through
the hallways with questions like, "How can we share out assets to
client X but not allow them to do or see Y", or "How can I automate the
process of creating multiple PDF's
from one print command?", or "How can I customize the interface so it
looks like my clients brand?" or "Can I do . . ." You get the point
right???. . .if not, the point is you know your world better than we do so listen
to your user community and then start thinking about how to solve their
challenges with the tool set at hand. And if the current tool set
doesn't accomplish the needed task, then there is most likely a plugin or a solution to make it do the trick like Creative Banks or Elegant or nTransit. Xinet is an open product so customizations can be done and probably already have been. Ask us-we'll be happy to help.
2)
Be creative yourself- it's easiest to hear from someone that says yeah,
we did this thing and it really rocks or ask us to replicate something
that was done before but think about how to do "it" yourself. You may be in IT or
work in some IT capacity but that doesn't mean you're not creative!
You are-you just have to find the time to sit down and think about it.
I know, I know, easier said than done as we're all super busy but if
you want to really do something, you'll make the time.
3) Is
my idea even possible? Look, all technologies have their limits so if
you want Xinet to make your coffee (not too strong of course!) AND fold
your laundry, you might be pushing it. So, this is a good time to ask
NAPC as well as look at the Xinet manuals. Seriously, look at the
manuals. I know a lot of you depend on NAPC for the knowledge to be
handed to you and we don't mind that at all. That's what we're here for! Although, you might be
better served by reading up on the technology you manage.
Right!?!?!?!?!? You've all heard RTFM or to be politically correct I
should say RTM
but whatever, you get the point. I've been to plenty of training
classes, received lots of great advice from others in and out of the
industry but my best resource to date has been the manuals. Read up
everyone!
So
really, those ARE my suggestions. Listen to your clients, be creative
yourself in coming up with ways to solve the challenge and do your
research by speaking with us and reading up on the Xinet manuals.
Seriously, you all will be waaaay
better off in the end if you put the time into it. Again, I know
you're all busy but this is important stuff here right?!?! Just like
working out, which I do all the time! :), it takes dedication and Xinet
is no different.
Also, after working the three steps above
first, you'll be able to better answer your first question yourself,
"How can I make Xinet do more for us". And if not, again, we're here
to help with suggestions and industry knowledge to get you to where you
need to be.
Bottom line, be proactive with learning this stuff . . .it'll really help you in the long run.
Oh
yeah, one more thing on this subject, work closely with the people that
have a hand in Xinet. If you're more on the creative side of things,
create the relationship needed to work together harmoniously with your
IT staff. Contrary to popular belief, they are truly there to help
you, not hold you back even though sometimes it may seem that way.
And, if you're in IT, be open about this stuff and the ideas that may
come your way. Don't start with "No", think about it and be creative
in helping solve the problem or challenge at hand by working closely
with your clients. Can I get a "Kum ba ya!"
Any way. . .two quick things before I get off my high horse. . .
1) Dialogue ES is around the corner. If you're familiar with how Dalim's Dialogue currently works then you'll probably be happy to hear how it's evolving. This week Dalim
is releasing the internal beta so I'll get my hands on it and write
another blog entry just on that subject later but some quick things to
mention:
The interface has changed quite a bit.- this is good stuff
guys and gals. . .totally revamped and much more slick. Again, more to
come later. As far as functionality, it's totally rewritten from the
ground up and now has a database behind it. This can open up all kinds
of possibilities, think about it. Linear versus non-linear workflows.
So now, instead of having user a, b, then c approve or reject a
document, it can be more of fluid approval process and not so much in a
linear fashion as it is now. Also, once a user approves or rejects the
document, that action doesn't have to stop the process as it does now
in a multi-user approval process. In other words, if user a rejects
the document, user b or c can still approve or reject it themselves
versus how it is now. In the current version, if a user in an approval
workflow
rejects the document, it's done, that's it. No one else can approve or
reject it using the built in approval tools. That changes with ES.
New icons for statuses, new list views to easily see all users status,
new interface, etc. Lots more to come there.
Any
way, thanks for taking the time to read our blog and I hope you all stay
tuned for more from NAPC. We're dedicated to making you successful!
Brian
For more info go to www.napc.com
Posted by Robert Sullivan on Mon, Mar 23, 2009 @ 09:08 PM
So here I go a blogging away. I'm not too sure of how to start because usually I'm following an agenda of some kind, or a question to start with. Someone has a problem or a challenging workflow to configure. Blogging is more free flowing, I guess... We'll see.
I do a lot of training for new clients and there is so much information to absorb that they can easily become overwhelmed & overloaded. As they get use to the system they'll Venture further (pun intended) and start looking for cleaner ways to use their new system. But not many will make that leap and really push the limits of what they can get out of it early on. There's way more power under the hood.
The problem I find so often is that people have an idea of what the server can do but aren't sure how to get there. And that stops them. At that point, the thing that most often drives them is a client request, or the boss (every one has a boss...) saw a Webinar about some cool widget and wants it created. Is it done yet?
When I was in Printing, I worked for a guy that always said,
"If I wanted it tomorrow, I'd be asking for it then. I want it now!"
Then the calls start coming in in earnest. Which is great for me because now I have something to dig into. A new challenge. But I wonder how do we get people to push the envelope before they get the push themselves? It's training, and it's knowledge. NAPC runs Webinars' all the time on different applications, from FullPress, the Venture database, to Dalim Dialogue or the Xinet Uploader. Sharing the knowledge is driving the train here!
Do you have an idea for a Webinar you'd like to see. Tell us. Got an idea for any Trigger automations, let me know. If you can conceive of it, we can build it... well, I'm looking for the stuff we can build out-of-the-box. Custom stuff comes later.
-Sully