thoughts on data storage for imaging facility

[Owens, Peter]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Dear all,

I have been searching for ideas regarding  what would be a suitable solution
for data storage for our microscopy facility (www.imaging.nuigalway.ie). The
first thought on this are to providing a data storage server that would be
networked to all microscopes (currently 10 standalone systems , scaling to
20 instruments in the future). We can cater for multi channel time lapse
live cell imaging so experiment  data sizes may go to the order of GB but
the norm is smaller data sizes.

I have a concern over upload download access speeds, security, what is the
right amount of storage that will cover our needs.

Does anyone out there have any suggestions / recommendations on how to go
about specifiying this system?

Thanks

Peter Owens, National University of Ireland Galway.

[Hugh Newman]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****


I would consider a NAS or Network Assisted Storage device such as the Drobo, found here. Make sure you get a model that can connect directly to a network and not just via USB. But even that will fill up with time. It might be handy to look back and see how much data is being written over the course of a couple of months to figure out what your needs are. I would suggest a NAS that once a year (or as needed) the data is removed and moved to a external server. So old data is still available, but probably via a slower connection, and your NAS is free to consume more fresh data.

Digital storage is always an issue. Hopefully you can find a solution that works for you.

Good luck!

Hugh Newman
 
Graduate Researcher

Dept. Physics and Physical Oceanography

Memorial University

St. Johns, Newfoundland, Canada

> Date: Tue, 14 Feb 2012 09:45:59 -0600
> From: [hidden email]
> Subject: thoughts on data storage for imaging facility
> To: [hidden email]
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I have been searching for ideas regarding  what would be a suitable solution
> for data storage for our microscopy facility (www.imaging.nuigalway.ie). The
> first thought on this are to providing a data storage server that would be
> networked to all microscopes (currently 10 standalone systems , scaling to
> 20 instruments in the future). We can cater for multi channel time lapse
> live cell imaging so experiment  data sizes may go to the order of GB but
> the norm is smaller data sizes.
>
> I have a concern over upload download access speeds, security, what is the
> right amount of storage that will cover our needs.
>
> Does anyone out there have any suggestions / recommendations on how to go
> about specifiying this system?
>
> Thanks
>
> Peter Owens, National University of Ireland Galway.
>
>  

     

[Kenton]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi
It may not be ideal but it does work having a storage drive as you have
described. A Gig at our university takes about 3min to transfer. The main
thing (and the reason for the post) is that you need a proper server, or at
least server network software as a shared drive on a normal PC can only have
10(?) simulations connections, and as all the microscopes/users may
automatically log in to it, it gets grumpy quite fast, and it deems the
wrong people the important ones! There are 3 microscopes for one PC shared
drive and it happens occasionally so with more it will be quite often.
Regards
Kenton

Dr Kenton Arkill
Honorary Research Fellow at the University of Bristol
Microvasular Research Laboratories, Physiology and Pharmacology
University of Bristol, Southwel St, BS2 8EJ

Research Ascociate at the University of Birmingham
Nanoscale Physics Research Laboratory, School of Physics and Astronomy
University of Birmingham, B15 2TT, UK

Location: Rm 106, Physics East, University of Birmingham
Tel: [phone]
Email: [hidden email]

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On
Behalf Of Peter Owens
Sent: 14 February 2012 15:46
To: [hidden email]
Subject: thoughts on data storage for imaging facility

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Dear all,

I have been searching for ideas regarding  what would be a suitable solution
for data storage for our microscopy facility (www.imaging.nuigalway.ie). The
first thought on this are to providing a data storage server that would be
networked to all microscopes (currently 10 standalone systems , scaling to
20 instruments in the future). We can cater for multi channel time lapse
live cell imaging so experiment  data sizes may go to the order of GB but
the norm is smaller data sizes.

I have a concern over upload download access speeds, security, what is the
right amount of storage that will cover our needs.

Does anyone out there have any suggestions / recommendations on how to go
about specifiying this system?

Thanks

Peter Owens, National University of Ireland Galway.

 

[Straatman, Kees (Dr.)]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Dear Peter,

After looking around and talking to several people I finally decided to ask our computer centre to take care for this. This means that the system is backed-up every night to a second server in a different building, security is taken care of, if the system is down they will sort out the problems (disk crashes!) and if I need more space I just ask for it. It also allows multiple users to access the server (from different microscopes or from behind their desk) without major speed problems as we use the university computer network, although for large datasets this is the bottle neck. We store data during image acquisition on the local computer and when the experiment is finished move it to the server. I agree, this comes with a price tag, but less headache for me.

Kees


Dr Ir K.R. Straatman
Senior Experimental Officer
Centre for Core Biotechnology Services
College of Medicine, Biological Sciences and Psychology
University of Leicester
http://www.le.ac.uk/biochem/microscopy/home.html


-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Peter Owens
Sent: 14 February 2012 15:46
To: [hidden email]
Subject: thoughts on data storage for imaging facility

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Dear all,

I have been searching for ideas regarding  what would be a suitable solution
for data storage for our microscopy facility (www.imaging.nuigalway.ie). The
first thought on this are to providing a data storage server that would be
networked to all microscopes (currently 10 standalone systems , scaling to
20 instruments in the future). We can cater for multi channel time lapse
live cell imaging so experiment  data sizes may go to the order of GB but
the norm is smaller data sizes.

I have a concern over upload download access speeds, security, what is the
right amount of storage that will cover our needs.

Does anyone out there have any suggestions / recommendations on how to go
about specifiying this system?

Thanks

Peter Owens, National University of Ireland Galway.

 

[Kurt Thorn]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

I maintain an 8 TB file server for this purpose.  It's a Ubuntu server
with  10 1 TB drives in a RAID 6 configuration.  It's mountable from any
of our microscopes and from most computers at the university.  It is not
designed for long term storage - in fact, because the data rate is a few
TB/year, we are implementing automatic deletion of files older than one
year, and requiring the end users to deal with long term data storage.  
It's mainly there to facilitate moving data around.

Once you have automated microscopes, you can generate pretty impressive
quantities of data - we have one user doing multiposition timelapse with
Z-stacks who generates ~200GB / night.

I'm happy to answer additional questions.

Best,
Kurt

On 2/14/2012 7:45 AM, Peter Owens wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I have been searching for ideas regarding  what would be a suitable solution
> for data storage for our microscopy facility (www.imaging.nuigalway.ie). The
> first thought on this are to providing a data storage server that would be
> networked to all microscopes (currently 10 standalone systems , scaling to
> 20 instruments in the future). We can cater for multi channel time lapse
> live cell imaging so experiment  data sizes may go to the order of GB but
> the norm is smaller data sizes.
>
> I have a concern over upload download access speeds, security, what is the
> right amount of storage that will cover our needs.
>
> Does anyone out there have any suggestions / recommendations on how to go
> about specifiying this system?
>
> Thanks
>
> Peter Owens, National University of Ireland Galway.
>
>
>

[Craig Brideau]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

We use a Mac Mini configured as a server (factory option) connected to a
Thunderbolt RAID box.  It is tied in to our university's Gigabit network
and thanks to some access control features it is fairly secure.  It did
take my boss some fiddling to get it up and running, but now that it's
working we haven't had to touch it (knock on wood!).  We originally had a
Drobo Pro but we found the access speed even using Firewire 800 was
underwhelming.  The Thunderbolt RAID (uses the new Thunderbolt connector on
the newest Macs) is quite a bit faster, to the point where you can see a
significant speed improvement whilst doing file copies over the network.

Craig


On Tue, Feb 14, 2012 at 10:49 AM, Kurt Thorn <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/**wa?A0=confocalmicroscopy<http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy>
> *****
>
> I maintain an 8 TB file server for this purpose.  It's a Ubuntu server
> with  10 1 TB drives in a RAID 6 configuration.  It's mountable from any of
> our microscopes and from most computers at the university.  It is not
> designed for long term storage - in fact, because the data rate is a few
> TB/year, we are implementing automatic deletion of files older than one
> year, and requiring the end users to deal with long term data storage.
>  It's mainly there to facilitate moving data around.
>
> Once you have automated microscopes, you can generate pretty impressive
> quantities of data - we have one user doing multiposition timelapse with
> Z-stacks who generates ~200GB / night.
>
> I'm happy to answer additional questions.
>
> Best,
> Kurt
>
>
> On 2/14/2012 7:45 AM, Peter Owens wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/**wa?A0=confocalmicroscopy<http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy>
>> *****
>>
>> Dear all,
>>
>> I have been searching for ideas regarding  what would be a suitable
>> solution
>> for data storage for our microscopy facility (www.imaging.nuigalway.ie).
>> The
>> first thought on this are to providing a data storage server that would be
>> networked to all microscopes (currently 10 standalone systems , scaling to
>> 20 instruments in the future). We can cater for multi channel time lapse
>> live cell imaging so experiment  data sizes may go to the order of GB but
>> the norm is smaller data sizes.
>>
>> I have a concern over upload download access speeds, security, what is the
>> right amount of storage that will cover our needs.
>>
>> Does anyone out there have any suggestions / recommendations on how to go
>> about specifiying this system?
>>
>> Thanks
>>
>> Peter Owens, National University of Ireland Galway.
>>
>>
>>
>>

[Ryan Schreiner]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

For our microscope group we worked with GVS to build a custom system for
data storage.  We do a lot of confocal/live imaging and generate large data
sets.  A description from the company below:

-Ryan

GVS offer one of the most cost effective high speed and secure storage
solution for the microscopy listserver to allow user with different
requirement to have dedicated access to media at high speed, from single
Gigabit, to 10Gibit with well design storage and easy of management  to
allow OS X, Windows and Unix user to access to a centralize storage with
multiple 10Gbit speed.

The process begins with high speed SAS Nomadic 4U Drive's over Fibre
channel network direct attached to GVS9000 1XU Metro were Metro is able to
route Both 8GB Fibre and 10GBit as SAN and NAS storage, Nomadic 4U build
with intelligent web base Hardware RAID 0, 1, 3, 5, 6 redundant hot-swap
controller with 4x8GB Interface to each 4480 controllers to deliver
aggregated speed of 6.4Gbit/sec.

With this technology you  have complete security, redudince and closter
design of your entire storage media and ease of setting up different user
with deferent volume, folder and files access with each specific storage
capacity with built in time line for the backup if you have to move storage
to off site or backup, archive all the tools in place to do so with simple
web interface and and single push of button.

On Tue, Feb 14, 2012 at 10:45 AM, Peter Owens <[hidden email]>wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I have been searching for ideas regarding  what would be a suitable
> solution
> for data storage for our microscopy facility (www.imaging.nuigalway.ie).
> The
> first thought on this are to providing a data storage server that would be
> networked to all microscopes (currently 10 standalone systems , scaling to
> 20 instruments in the future). We can cater for multi channel time lapse
> live cell imaging so experiment  data sizes may go to the order of GB but
> the norm is smaller data sizes.
>
> I have a concern over upload download access speeds, security, what is the
> right amount of storage that will cover our needs.
>
> Does anyone out there have any suggestions / recommendations on how to go
> about specifiying this system?
>
> Thanks
>
> Peter Owens, National University of Ireland Galway.
>
>
>

[Sally Stowe]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Dear All,
I'm inclined to think that data storage is now so cheap,  the solutions so
varied, and the demand increasing so fast,  that the most practical and
cheapest option at the moment is just to put the biggest affordable hard
drive and a virus checker on each computer, institute a "delete after x
weeks"  policy, and let the users bring in a USB stick or a portable drive
to transport their data.  USB 3 and an SSD if speed is an issue.
Unless your central computing service can offer a good deal - but sometimes
its good to be off-network, particularly for older systems.

Decentralising technologies always seem to sneak up on us.
.
cheers
Sally

Sally Stowe


On 15 February 2012 02:45, Peter Owens <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I have been searching for ideas regarding  what would be a suitable
> solution
> for data storage for our microscopy facility (www.imaging.nuigalway.ie).
> The
> first thought on this are to providing a data storage server that would be
> networked to all microscopes (currently 10 standalone systems , scaling to
> 20 instruments in the future). We can cater for multi channel time lapse
> live cell imaging so experiment  data sizes may go to the order of GB but
> the norm is smaller data sizes.
>
> I have a concern over upload download access speeds, security, what is the
> right amount of storage that will cover our needs.
>
> Does anyone out there have any suggestions / recommendations on how to go
> about specifiying this system?
>
> Thanks
>
> Peter Owens, National University of Ireland Galway.
>
>
>

[Gregg Jarvis]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Peter,
I have found after 14 years of data storage for images, the best thing is
always using the back-up capabilities on any work supplied server and if in
doubt DVDs hold many images which can be readily available as a sort of
library and if the server or computer dos crash you can always have them.
Using a thumbnail drive with 8GB is a way for quick transfer files and
allows easier copying later on in editing software.
 Hope this helps,
Gregg Jarvis
Senior Spectroscopist.
Omega Optical Inc.
[hidden email]



On Tue, Feb 14, 2012 at 12:17 PM, Straatman, Kees R. (Dr.) <
[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear Peter,
>
> After looking around and talking to several people I finally decided to
> ask our computer centre to take care for this. This means that the system
> is backed-up every night to a second server in a different building,
> security is taken care of, if the system is down they will sort out the
> problems (disk crashes!) and if I need more space I just ask for it. It
> also allows multiple users to access the server (from different microscopes
> or from behind their desk) without major speed problems as we use the
> university computer network, although for large datasets this is the bottle
> neck. We store data during image acquisition on the local computer and when
> the experiment is finished move it to the server. I agree, this comes with
> a price tag, but less headache for me.
>
> Kees
>
>
> Dr Ir K.R. Straatman
> Senior Experimental Officer
> Centre for Core Biotechnology Services
> College of Medicine, Biological Sciences and Psychology
> University of Leicester
> http://www.le.ac.uk/biochem/microscopy/home.html
>
>
> -----Original Message-----
> From: Confocal Microscopy List [mailto:[hidden email]]
> On Behalf Of Peter Owens
> Sent: 14 February 2012 15:46
> To: [hidden email]
> Subject: thoughts on data storage for imaging facility
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I have been searching for ideas regarding  what would be a suitable
> solution
> for data storage for our microscopy facility (www.imaging.nuigalway.ie).
> The
> first thought on this are to providing a data storage server that would be
> networked to all microscopes (currently 10 standalone systems , scaling to
> 20 instruments in the future). We can cater for multi channel time lapse
> live cell imaging so experiment  data sizes may go to the order of GB but
> the norm is smaller data sizes.
>
> I have a concern over upload download access speeds, security, what is the
> right amount of storage that will cover our needs.
>
> Does anyone out there have any suggestions / recommendations on how to go
> about specifiying this system?
>
> Thanks
>
> Peter Owens, National University of Ireland Galway.
>
>
>

[mahogny]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

this is a really tough problem. these are the problems you are facing:
* you will need a whole lot of harddrives, and there is a limit to what you
can put into one machine
* backup is needed, but expensive. what is reasonable backup for this size?
and how can you make sure backup is not sagging down your system?
* how do you make this much data available over the internet?
* how do you organize the data? and keep it secure?

ok. so what you do not do:
* you do not use a windows computer. have fun with the viruses
* you do not store files just in the file system. this will ensure
organization chaos in a few years. no one will know what has been recorded
* do not buy tons of cheap external harddrives. this is almost guaranteed
data loss
* do not spread it over multiple computers. it is an administrative mess
* do not ignore the metadata. you really want to keep this for future
reference
* overall, avoid consumer products. these are wear, tear, throw away, and
no sophisticated technical support (you probably want it when you just lost
5TB data)
* put the data on a computer a user is using for a workstation

this is what we did:
we bought a Dell r510. 12+2 drive bays (12 hot swap, 2 internal). we bought
the trays and also 3TB SAS drives from 3rd party. this cut the price in
half (dell will not recommend it). we only bought 6 drives at first, we can
expand later. the computer runs ubuntu for technical reasons, but
redhat/centos might be more suitable for general users.

on each set of 6 drives, we run raid 6. on top of the raids, we use linux
LVM. this means we get one big contiguous area and can just expand the file
system. these rack servers also have solutions to add even more drives.
this makes it very easy to organize everything, and data is safe (against
hardware errors, not user errors). proper raid6 ensures no downtime for
broken harddrives.

the server is in a proper server hall. this makes it much more reliable
(environment matters). UPS and dual PSUs.

data is stored in OMERO whenever possible. this ensures organization and
searchability, but also makes it possible to serve the data to
collaborators over the internet.

additional functions will be run virtualized on the computer for security.
I have good experience with virtualbox but should look into KVM.

if you also want backup (this is only safe storage) then you need to buy
another computer like this one, and then run rsync between the two.

/Johan


On Tue, Feb 14, 2012 at 4:45 PM, Peter Owens <[hidden email]>wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I have been searching for ideas regarding  what would be a suitable
> solution
> for data storage for our microscopy facility (www.imaging.nuigalway.ie).
> The
> first thought on this are to providing a data storage server that would be
> networked to all microscopes (currently 10 standalone systems , scaling to
> 20 instruments in the future). We can cater for multi channel time lapse
> live cell imaging so experiment  data sizes may go to the order of GB but
> the norm is smaller data sizes.
>
> I have a concern over upload download access speeds, security, what is the
> right amount of storage that will cover our needs.
>
> Does anyone out there have any suggestions / recommendations on how to go
> about specifiying this system?
>
> Thanks
>
> Peter Owens, National University of Ireland Galway.
>
>
>

--
-----------------------------------------------------------
Johan Henriksson
PhD student, Karolinska Institutet
http://mahogny.areta.org  http://www.endrov.net

[Chris Guerin]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi everyone:

I agree with Johan's post, and in fact it is very similar to the solution we
use at the VIB in Gent. I wanted to give a few more facts and figures to
help anyone who is thinking about this issue.

We have a number of conventional widefield systems, a dedicated system for
multi-position time lapse imaging and a single confocal microscope for a
department of about 180.

We implemented our system about 4 years ago and installed 14 usable TB of
storage with a complete mirrored backup server. To date we have had to
archive all data pre-2010 to have adequate space for new data. Shortly we
will have to add a new rack and backup which we estimate at about 35K Euros.

All of our core workstations are wired to a 1GB switch for fast(er) data
transfer, but it could still be improved so in a new space we are building
we are installing a 10GB fiberoptic backbone.

Storage isn't cheap, nor is administering it trivial. However I disagree
with those who feel it is a users only problem. I feel that it is part of
the essential service of a microscopy core to first protect the data we
produce and also to ensure that one copy of the unchanged raw data is
preserved in the case of any challenges to scientific results. When we argue
for more money to safeguard data we always remind the admin/PI's that every
specimen brought to the microscope represents a significant capital
investment and loss of the data is the equivalent of losing money-this
usually works.

I agree with all of Johan's do nots and would add one more. Do not allow
users to have write access to your archived data or transfer from your
workstations using USB sticks or portable hard drives even with virus
protection software.

Best to all,

Chris

[Turnbull, Lon]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi All,

What to do about microscope data storage can depend on how your university regulates access to the network.

At our university all computers that have access to the network are controlled by the university and so it is unwise to have a data acquisition computer attached to the network.  Computers that are hooked to the network can be rebooted or slowed down at any time by some automated program!  This is not desirable for long term data acquisition.  We have had some users acquire data from a confocal microscope over a 4 day time frame.  This is next to impossible to do with a computer that is hooked up to the network. Some users do gather many GB of data, but are responsible for removing it for analysis elsewhere.  Usually this is done with flash drives or portable disk drives since optical media does not store enough bytes.  These devices are relatively inexpensive.  I believe that it is the responsibility of the researchers to find a mechanism for long term storage of their data.

The problem of viruses is minimized on this campus, because most of the computers that students use are on the network and any devices that are plugged into one of these computers are automatically checked and cleaned.  We have had only one virus problem on our confocal system in five years.

-----------------------------------------------------------------

Lon Turnbull, Ph.D.
Microscope Supervisor
Department of Biological Sciences
University of North Texas
940-369-8721
[hidden email]

> Date: Tue, 14 Feb 2012 09:45:59 -0600
> From: [hidden email]
> Subject: thoughts on data storage for imaging facility
> To: [hidden email]
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I have been searching for ideas regarding  what would be a suitable solution
> for data storage for our microscopy facility (www.imaging.nuigalway.ie). The
> first thought on this are to providing a data storage server that would be
> networked to all microscopes (currently 10 standalone systems , scaling to
> 20 instruments in the future). We can cater for multi channel time lapse
> live cell imaging so experiment  data sizes may go to the order of GB but
> the norm is smaller data sizes.
>
> I have a concern over upload download access speeds, security, what is the
> right amount of storage that will cover our needs.
>
> Does anyone out there have any suggestions / recommendations on how to go
> about specifiying this system?
>
> Thanks
>
> Peter Owens, National University of Ireland Galway.
>
>  

     

[Sylvie Le Guyader-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hello

We have a NAS (9Tbx2, mirrored every night) connected to our 5 confocals via a 1Gb network cable that is independent from the intranet. Our offline stations access the NAS from the intranet so the bandwidth is shared by far many more people and upload/downloads are slower. We routinely acquire 50-100Gb files.

In any case I would recommend autosaving data on the local internal HD of the microscope PC and transferring them to the server only after acquisition.

Med vänlig hälsning / Best regards
 
Sylvie
 
@@@@@@@@@@@@@@@@@@@@@@@@
Sylvie Le Guyader
Live Cell Imaging Unit
Dept of Biosciences and Nutrition
Karolinska Institutet
Novum
14183 Huddinge
Sweden
office: [phone]
LCI room: [phone]
mobile: [phone]

> -----Original Message-----
> From: Confocal Microscopy List
> [mailto:[hidden email]] On Behalf Of Turnbull, Lon
> Sent: 15 February 2012 21:08
> To: [hidden email]
> Subject: Re: thoughts on data storage for imaging facility
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hi All,
>
> What to do about microscope data storage can depend on how your university
> regulates access to the network.
>
> At our university all computers that have access to the network are controlled by
> the university and so it is unwise to have a data acquisition computer attached to
> the network.  Computers that are hooked to the network can be rebooted or slowed
> down at any time by some automated program!  This is not desirable for long term
> data acquisition.  We have had some users acquire data from a confocal
> microscope over a 4 day time frame.  This is next to impossible to do with a
> computer that is hooked up to the network. Some users do gather many GB of
> data, but are responsible for removing it for analysis elsewhere.  Usually this is
> done with flash drives or portable disk drives since optical media does not store
> enough bytes.  These devices are relatively inexpensive.  I believe that it is the
> responsibility of the researchers to find a mechanism for long term storage of their
> data.
>
> The problem of viruses is minimized on this campus, because most of the
> computers that students use are on the network and any devices that are plugged
> into one of these computers are automatically checked and cleaned.  We have had
> only one virus problem on our confocal system in five years.
>
> -----------------------------------------------------------------
>
> Lon Turnbull, Ph.D.
> Microscope Supervisor
> Department of Biological Sciences
> University of North Texas
> 940-369-8721
> [hidden email]
>
>
>
> > Date: Tue, 14 Feb 2012 09:45:59 -0600
> > From: [hidden email]
> > Subject: thoughts on data storage for imaging facility
> > To: [hidden email]
> >
> > *****
> > To join, leave or search the confocal microscopy listserv, go to:
> > http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> > *****
> >
> > Dear all,
> >
> > I have been searching for ideas regarding  what would be a suitable solution
> > for data storage for our microscopy facility (www.imaging.nuigalway.ie). The
> > first thought on this are to providing a data storage server that would be
> > networked to all microscopes (currently 10 standalone systems , scaling to
> > 20 instruments in the future). We can cater for multi channel time lapse
> > live cell imaging so experiment  data sizes may go to the order of GB but
> > the norm is smaller data sizes.
> >
> > I have a concern over upload download access speeds, security, what is the
> > right amount of storage that will cover our needs.
> >
> > Does anyone out there have any suggestions / recommendations on how to go
> > about specifiying this system?
> >
> > Thanks
> >
> > Peter Owens, National University of Ireland Galway.
> >
> >
>