DAVE YOUNG​ Head of Engineering T  +1 312 605 8900 THE MILL 1000 WEST FULTON MARKET, SUITE 250, CHICAGO, IL 60607 FOLLOW @MILLCHANNEL |  FACEBOOK  |  TWITTER  |  INSTAGRAM

The last two  years i have seen a lot of storage vendors pushing NVME upgrades or offering  explicitly  NVME storage. It appear that is becoming the norm these days. We are living in the era of technology advancements, new hard  drives, controllers , network cards, cpus becoming faster, bigger, lower latency every day. That is all good and companies are integrating  those in their workflows, well at least the companies  that can afford it. What i have seen in the the real world is that such solutions are still not enough for a M&E average production. Let me explain my thoughts. If you spec out a full NVME solution is probably in the 200k+  range and you may   get 40TB usable. The system will be super fast  and will have no issues keeping up with the iops and bandwidth for your ever demanding workflow.  Now that you have the system in, you have to figure out to to deliver that performance to the workstation. Find network cards that works with tour OS and compatible drivers.  Use the available  open source protocols  like NFS, SMB and figure out to to optimize them,  or write your own agent/protocol to actually take advantage of that performance. Ok well some of you have that figure out, now you have to see how your NLE will utilize such performance  and how to you maintain that setup with the inevitable upgrades  like OS versions, and application version.     What if i tell you that there is another way to get faster performance  than the NVME flash arrays at more affordable cost.     Well lets get a bit technical now. What is the fastest piece of hardware that access data in a computer? Is it the hard drive ? Is the CPU? Is it the bus (PCI slots) ? Is it the RAM? Which one is M&E industry  always in demand. Of course all those do matter in a traditional server but how do you get that performance? One piece of hardware that i have see not being utilize in the traditional  storage vendors is the RAM. Some of the hardware manufacturers will offer hardware controllers that  use some caching or they have some kind of cashing mechanism build in. Not large enough  to make a significant difference. They usually do not get into very details about how that works. You have to dig it up yourself.   RAM is the fastest  way to access data, large data, media and entertainment data. Load that  uncompressed movie into RAM, an average server these day can easily use 1TB of RAM. The challenge here is how to tell the filesystem and the OS to load that data into RAM. I have been working  with Solaris  ZFS for the last 5 years and no other OS/filesystem does it better  and more efficient. ZFS will load the most frequently accessed data in to RAM ARC. DDR4 at 2666MHz ECC low latency is utilized here. When the RAM is full the system will load data in the LARC which is the read cache. That is where the fast NVME drives  come handy. We used two 4TB Intel striped and we get 8TB read cache. I can tell you is quite amazing to see both of those RAM and READ CACHE max out on a production day when you have 60 edit bays hitting it. Lets do some quick math here, 1TB of RAM speed and 8TB of two NVME striped using PCI slots. I would say that is fast enough for a traditional  production workflow and also 9TB available  worth of media files ready to serve and any given time. I have posted some benchmarks ran a CLI  from a real production server  below so you can see some realistic numbers.    More over all the editing, compositing, finishing, motion graphics applications that are being developed are thirsty for ram. You can see it in system resources if you open  4K media  in your timeline. Somehow the application process sucks the entire ram available. They do that to load the data into ram for smoother playback. It does not matter how much ram you install, 64G, 124G, 256G  any modern release software will use it assuming that you have a heavy 4K timeline.        Here are some performance stats that we got from out latest  server ZFS-600 that we just put in production. These are results running CLI commands  not theoretical and accumulative bandwidth taken from data sheet products.    RAM sustainable bandwidth 197GB/sec  command line volume test run with dd   7.3GB/sec command line multithreaded using iozone 28GB/sec  that is upper case B means Bytes. The full data sheet and server specs  of the ZFS-600 can be found of the here http://ittemple.com/products/

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

The last two  years i have seen a lot of storage vendors pushing NVME upgrades or offering  explicitly  NVME storage. It appear that is becoming the norm these days. We are living in the era of technology advancements, new hard  drives, controllers , network cards, cpus becoming faster, bigger, lower latency every day. That is all good and companies are integrating  those in their workflows, well at least the companies  that can afford it. What i have seen in the the real world is that such solutions are still not enough for a M&E average production. Let me explain my thoughts. If you spec out a full NVME solution is probably in the 200k+  range and you may   get 40TB usable. The system will be super fast  and will have no issues keeping up with the iops and bandwidth for your ever demanding workflow.  Now that you have the system in, you have to figure out to to deliver that performance to the workstation. Find network cards that works with tour OS and compatible drivers.  Use the available  open source protocols  like NFS, SMB and figure out to to optimize them,  or write your own agent/protocol to actually take advantage of that performance. Ok well some of you have that figure out, now you have to see how your NLE will utilize such performance  and how to you maintain that setup with the inevitable upgrades  like OS versions, and application version.     What if i tell you that there is another way to get faster performance  than the NVME flash arrays at more affordable cost.     Well lets get a bit technical now. What is the fastest piece of hardware that access data in a computer? Is it the hard drive ? Is the CPU? Is it the bus (PCI slots) ? Is it the RAM? Which one is M&E industry  always in demand. Of course all those do matter in a traditional server but how do you get that performance? One piece of hardware that i have see not being utilize in the traditional  storage vendors is the RAM. Some of the hardware manufacturers will offer hardware controllers that  use some caching or they have some kind of cashing mechanism build in. Not large enough  to make a significant difference. They usually do not get into very details about how that works. You have to dig it up yourself.   RAM is the fastest  way to access data, large data, media and entertainment data. Load that  uncompressed movie into RAM, an average server these day can easily use 1TB of RAM. The challenge here is how to tell the filesystem and the OS to load that data into RAM. I have been working  with Solaris  ZFS for the last 5 years and no other OS/filesystem does it better  and more efficient. ZFS will load the most frequently accessed data in to RAM ARC. DDR4 at 2666MHz ECC low latency is utilized here. When the RAM is full the system will load data in the LARC which is the read cache. That is where the fast NVME drives  come handy. We used two 4TB Intel striped and we get 8TB read cache. I can tell you is quite amazing to see both of those RAM and READ CACHE max out on a production day when you have 60 edit bays hitting it. Lets do some quick math here, 1TB of RAM speed and 8TB of two NVME striped using PCI slots. I would say that is fast enough for a traditional  production workflow and also 9TB available  worth of media files ready to serve and any given time. I have posted some benchmarks ran a CLI  from a real production server  below so you can see some realistic numbers.    More over all the editing, compositing, finishing, motion graphics applications that are being developed are thirsty for ram. You can see it in system resources if you open  4K media  in your timeline. Somehow the application process sucks the entire ram available. They do that to load the data into ram for smoother playback. It does not matter how much ram you install, 64G, 124G, 256G  any modern release software will use it assuming that you have a heavy 4K timeline.        Here are some performance stats that we got from out latest  server ZFS-600 that we just put in production. These are results running CLI commands  not theoretical and accumulative bandwidth taken from data sheet products.    RAM sustainable bandwidth 197GB/sec  command line volume test run with dd   7.3GB/sec command line multithreaded using iozone 28GB/sec  that is upper case B means Bytes. The full data sheet and server specs  of the ZFS-600 can be found of the here http://ittemple.com/products/

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Thanks everyone :) All good. :)

On 2017-07-21 15:00, Mike Owen wrote:

Sure, use coupon code: THINKBOX
On 21 July 2017 at 19:44, James Braid <jamesb@loreland.org> wrote:

On 21 July 2017 at 11:40, John Hickson <content@studiosysadmins.com> wrote:
>
> Hey a few people are still looking for free passes for the exhibition area at SIGGRAPH this year.
>
> Anyone still have active links available for registration?

Isotropix (Clarisse) have a free exhibits pass:
http://www.isotropix.com/siggraph-2017
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

On Jun 21, 2017, at 12:44 PM, Saker Klippsten <sakerk@gmail.com> wrote:
https://techcrunch.com/2017/06/21/facebook-open-sources-its-it-management-tool-for-distributing-adobes-creative-cloud-tools-to-employees/

So yeah TLDR; we spent engineering $$ and resources to make an app to deploy Adobe software cause your license scheme and deployment software that comes with it.. suckz
. 



To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Hi,
This news I like very much.
Thanks for the post.
- Brian
Every day is a gift Thats why they call it the present"

On Apr 4, 2017, at 10:04 AM, William Sandler <william.sandler@allthingsmedia.com> wrote:https://techcrunch.com/2017/04/04/apple-pushes-the-reset-button-on-the-mac-pro/


William Sandler All Things Media, LLCOffice: 201.818.1999 Ex 158.  william.sandler@allthingsmedia.comTo unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Hey SSA - came up on the Nuke list and has me curious:   is it still the
case with modern windows/samba etc that a //UNC path is slower than a
mapped drive?   Been any change to this behavior the last few years?

@Todd,
The issue we have currently is the READs are killing the performance on the server, we are writing the Writes to the local node, and then copying back to the server after. Too many large files needing to be read by nodes, scaled by 40 nodes. and then artists trying to use that same server to continue comping.
Really am considering just building a massive Flash storage box, and creating a symlink to the outputs directory for the comp and CG outputs.
On Fri, Sep 16, 2016 at 11:46 AM, Todd Smith <todd@sohovfx.com> wrote:

If the problem he is encountering is fast Nuke writes killing his NAS performance, then client side caching is going to have little to no effect. 
Since each render node is reading a unique set of frames, and successive renders of the same frame will rarely hit the same render node, client side caching will not be viable.Avere will fair slightly better mainly because of delayed write back which can help smooth out performance, and the chances of the data being in the cache already are higher, due to the artist working on the NAS based footage at their desk.


Todd SmithHead of Information Technology
soho vfx | 99 Atlantic Ave. Suite 303, Toronto, Ontario M6K 3J8office: (416) 516-7863 fax: (416) 516-9682 web: sohovfx.com

---

@SahamSorry, totally missed this reply last week. How about this as a $free option. Please do get in touch with Thinkbox support. They have a couple of scripts for Maya to cache locally all/some(filtering system) asset data and dynamically rewire the scene file up, before proceeding with render/sim/bake/script/bifrost/export job. This code will see the light of day for everyone, but we have much bigger plans in this space. Of course, using Avere or PerfAccel will really help here, assuming the budget is available.
On 16 September 2016 at 16:10, Rob Tomson <content@studiosysadmins.com> wrote:

If you have a Linux farm with local SSDs accessing storage over NFS, I would check out PerfAccel (http://datagres.com/perfaccel_caching.html). It's a persistent client-side NFS cache and works very well to reduce filer load and bring the data closer to the compute nodes.   Rob

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

--
         ,\|//,
        ( o - )
-oOO--(_)--Ooo-----------------
Saham Ali
Founder/Systems EngineerDvNT Technologies
saham@dvnttechnologies.com
407.729.3584 - Direct

  .ooO
  (    )     Ooo.
--\  ( ------(    )-------------------
    \_)       )  /
              (_/
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

As far as I understand, the blades are more dense than current blades, and you can fit a lot more in than shown in some of the diagrams. How many, exactly, I'm not sure, but as Dave Goodbourn said, I think they're looking at putting them in back to back.

More compute, less noise/power/cost.

My concerns are about upgrades, and having water _anywhere_ near my server room.

Piqued my interest though. Bringing the "Power Usage Effectiveness" down from an industry average of 1.7 to 1.1 or below is a step change.
--


On 04/08/16 18:00, Rob LaRose wrote:

9 servers in 24RU, though.  Is that a bit space-wasteful?


--Rob
==== This message was composed entirely with my right thumb. Please excuse any typos.
On Aug 4, 2016, at 11:10 AM, Brian Krusic <brian@krusic.com> wrote:

WOW, I like whats Im seeing.
Havent read everything yet though.
Perhaps good timing as Im reviewing our current space heater farm.
- Brian
A good day is when no one shows up... and you don't have to go anywhere."
On Aug 4, 2016, at 8:00 AM, Peter Smith <lists@kush-t.co.uk> wrote:

Hola!

Anyone looked at this technology? http:/www.iceotope.com/

Just had a quick tour myself. Looks VERY interesting for our industry needs.

What do you think?
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Dan Young Lead Systems Engineer

London New York Los Angeles Montral T  +1 212 775 0600 135 Spring Street, New York NY 10012 Twitter Facebook framestore.com

Oops, that's good information isn't it? Sorry, first day back from a two week vacation.

Los Angeles, specifically Santa Monica.

Thanks,

Ryan

On Thu, Jul 21, 2016 at 4:00 PM, Saker Klippsten <sakerk@gmail.com> wrote:

where art thou?
On Thu, Jul 21, 2016 at 3:48 PM, Ryan Lynch <76.lynch.ryan@gmail.com> wrote:

Looking for a projection screen vendor / installer. I'm looking at a specific Stewart Film Screen but the vendor Stewart recommended seems to only have experience with churches and schools. I'd like to find someone has has dealt with real screening rooms.

Anyone got a solid recommend?

Thanks,

Ryan

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Come see G-Tech's Evolution Series of products in action.  From the field, to desktop, to .....  Ooops...  Can't mention that yet...  G-Tech will be showing its newest products at NAB.  Come see why G-Tech is the only end to end solution for post-production that starts by securing your footage in the field, easily transferring to your desktop and working storage, then easy daily creation and transport, as well as backup and sharing. Hope to see you there! Proudly representing G-Technology in Canada. Peter Reitmeier +1 647 226-2355 peter@up-market.ca www.up-market.ca - Manufacturers Repreentatives

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Yes, please refrain from using the forums area for sales pitches - this is for discussions, questions and answers.

If someone asks a specific question and you are answering it that is great.

Thanks you./

-john

On 2015-06-19 10:49, Todd Smith wrote:

Hey John, Any chance we can mute this constant Oracle sales pitch? This guys isn't adding any value here. Thanks, Todd Smith Head of Information Technology soho vfx |  99 Atlantic Ave. Suite 303, Toronto, Ontario M6K 3J8 office: (416) 516-7863 fax: (416) 516-9682 web: sohovfx.com

---

Ivan, After a short search, I came up with nothing, just as you have. I would suggest thinking about the time and effort it would take, even if you could find the exact drives you were looking for, and then try to integrate, test, troubleshoot and then put them into production. My suggestion would be taking a look at a refresh of h/w and then spending your time migrating your data to the new architecture. There are a lot of great choices out there but we have found that the Oracle ZFS products have the highest perfomance and lowest TCO of any of the NAS offerigs and as a plus the product can also have a co-existing SAN personality along with the NAS simplicity, performance and lower cost. LMK if you woud like some additional information or one of our vendors that have this solution and they can telll you whathe they think about going that direction. Mark

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe