SAN / Epicor 9 Question

Yahoo Archive
1
As I said in the EUG, the information provided in that document is extremely questionable. The stats that are provided are basically the build for a company that can afford over $500,000 in server resources alone. Sorry, not trying to shoot the messenger, but rather informing the masses. It's still an interesting read, but highly unrealistic and the statistics are very hard to believe.

If you havent invested in a SAN/virtualization and are looking into it I put up a post that may have been missed at http://tech.groups.yahoo.com/group/vantage/message/97446

"Zac" Jason Woodward
Network Administrator
Intermountain Electronics, Inc.
O: 877-544-2291
M: 435-820-6515
F: 435-637-9601
www.ie-corp.com

Creating customer confidence through extraordinary service and experienced industry experts.


--- In vantage@yahoogroups.com, Scott Lepley <scott.lepley@...> wrote:
>
> Said document is availableÂ…
>
>
>
> https://epicweb.epicor.com/resources/MRCCustomers/VMware_and_Epicor_Deployment_Guide_FINAL.pdf
>
>
>
> Hope it helps!
>
>
> ------------------------------
>
> *From:* vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] *On Behalf
> Of *pbparker
> *Sent:* Tuesday, February 08, 2011 4:06 PM
> *To:* vantage@yahoogroups.com
> *Subject:* [Vantage] Re: SAN / Epicor 9 Question
>
>
>
> I can attest to the same resulting performance on pretty much an identical
> setup. We just purchased 2 EqualLogic PS6000's, 3 PowerEdge R710's all
> decked out, VMWare 4.1 and not sparing any expense in connectivity or
> anything else... and sadly it's all outperformed by a 5 year old PowerEdge
> physical box.
>
> We've tweaked everything and nothing seems to improve Epicor performance.
> Our entire company runs on these SAN's and EVERYTHING and I mean everything
> else runs better on this new setup except Epicor.
>
> So, I hope they get it figured out at some point. Epicor heralded at
> Perspectives last year this document that would reveal all the fixes for
> Epicor on VMWare and how to resolve performance, and said document is still
> not available as of last week of me trying to get it.
>
> Disappointed to say the least.
>
>
> --- In vantage@yahoogroups.com, "dcbmichaels" <dcbmichaels@> wrote:
> >
> > We were running our E9 production in VM using vsphere4.1 with Equallogic
> SAN on the backend, and citrix xenapp on the frontend. But had terrible
> speed issues, even after tweaking the system per the Dell/Equallogic/VMware
> tech support gurus. We have since setup a physical box and have gotten twice
> the performance out of the physical box as compared to the VM setup.(A
> report that takes 4min on the VM...takes less than 2min on the physical
> machine) So we are moving our production onto a physical box... while we use
> the VMs for our Development and testing E9 environment. The VM world is
> great...just not for production E9...
> >
>
>
> [Non-text portions of this message have been removed]
>
2
Hello,

We are currently running Vantage 406A and will be upgrading to Epicor 9 with a go live date of April 1 next year. As a part of the project we are also upgrading our network infrastructure, which includes placing Epicor 9 on VM Ware and upgrading our SAN. We are currently leaning toward the Dell Equillogic SAN solution with a Raid 50 configuration. I was wondering if anyone else was using a similar solution from Dell. We are also considering migrating from Progress to SQL Server, so the SQL Server database would be on the SAN.

Thanks,

David Pfiester
NuStep, Inc.
Ann Arbor, MI
734-769-3939 Ext 150
3
We already have VM Ware, SQL, and a NetApp SAN. We run 8.03. We plan on upgrading our SAN to either DELL Equallogic or NetApp. The performance of our SAN with SQL Server over running RAID on local disks was unbelievable. The end users definitely saw the performance increase. Running Vantage in a VM with 30 users, FRx, and Service Connect (all on the same server); works great as well.

From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf Of nustepvantage
Sent: Tuesday, August 24, 2010 9:49 AM
To: vantage@yahoogroups.com
Subject: [Vantage] SAN / Epicor 9 Question



Hello,

We are currently running Vantage 406A and will be upgrading to Epicor 9 with a go live date of April 1 next year. As a part of the project we are also upgrading our network infrastructure, which includes placing Epicor 9 on VM Ware and upgrading our SAN. We are currently leaning toward the Dell Equillogic SAN solution with a Raid 50 configuration. I was wondering if anyone else was using a similar solution from Dell. We are also considering migrating from Progress to SQL Server, so the SQL Server database would be on the SAN.

Thanks,

David Pfiester
NuStep, Inc.
Ann Arbor, MI
734-769-3939 Ext 150



[Non-text portions of this message have been removed]
4
Hi Joe,



I guess I'm surprised you've seen such an increase in performance when going from local disks to a SAN. It almost seems counterintuitive.

How were the disks setup up (RAID Configuration) when using local disks? How much bandwidth do you have between your SAN and Host server?



From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf Of Joe Luster
Sent: Tuesday, August 24, 2010 10:25 AM
To: vantage@yahoogroups.com
Subject: RE: [Vantage] SAN / Epicor 9 Question





We already have VM Ware, SQL, and a NetApp SAN. We run 8.03. We plan on upgrading our SAN to either DELL Equallogic or NetApp. The performance of our SAN with SQL Server over running RAID on local disks was unbelievable. The end users definitely saw the performance increase. Running Vantage in a VM with 30 users, FRx, and Service Connect (all on the same server); works great as well.

From: vantage@yahoogroups.com <mailto:vantage%40yahoogroups.com> [mailto:vantage@yahoogroups.com <mailto:vantage%40yahoogroups.com> ] On Behalf Of nustepvantage
Sent: Tuesday, August 24, 2010 9:49 AM
To: vantage@yahoogroups.com <mailto:vantage%40yahoogroups.com>
Subject: [Vantage] SAN / Epicor 9 Question

Hello,

We are currently running Vantage 406A and will be upgrading to Epicor 9 with a go live date of April 1 next year. As a part of the project we are also upgrading our network infrastructure, which includes placing Epicor 9 on VM Ware and upgrading our SAN. We are currently leaning toward the Dell Equillogic SAN solution with a Raid 50 configuration. I was wondering if anyone else was using a similar solution from Dell. We are also considering migrating from Progress to SQL Server, so the SQL Server database would be on the SAN.

Thanks,

David Pfiester
NuStep, Inc.
Ann Arbor, MI
734-769-3939 Ext 150

[Non-text portions of this message have been removed]












Joe Rojas | Director of Information Technology | Mats Inc
dir: 781-573-0291 | cell: 781-408-9278 | fax: 781-232-5191
jrojas@... | www.matsinc.com Ask us about our clean, green and beautiful matting and flooring


This message is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake. Please note that any views or opinions presented in this email are solely those of the author and do not necessarily represent those of the company.


[Non-text portions of this message have been removed]
5
Hi Joe,

> I guess I'm surprised you've seen such an increase in performance when going from local disks
> to a SAN. It almost seems counterintuitive.

Spindles equals SPEED! There are several articles out there but here's
one I just Googled:

http://blogs.technet.com/b/pkcrm/archive/2009/04/25/think-spindles-not-space.aspx

(Tom's Hardware has some nice testing graphs as well.)

Mark W.
6
Thank Mark



I completely agree with the spindle concept but when comparing apples to
apples, I'd expect local storage to beat out a SAN every time.





From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf
Of Mark Wonsil
Sent: Tuesday, August 24, 2010 11:00 AM
To: vantage@yahoogroups.com
Subject: Re: [Vantage] SAN / Epicor 9 Question





Hi Joe,

> I guess I'm surprised you've seen such an increase in performance when
going from local disks
> to a SAN. It almost seems counterintuitive.

Spindles equals SPEED! There are several articles out there but here's
one I just Googled:

http://blogs.technet.com/b/pkcrm/archive/2009/04/25/think-spindles-not-s
pace.aspx

(Tom's Hardware has some nice testing graphs as well.)

Mark W.




[Non-text portions of this message have been removed]
7
RAID 10
RAID 10
RAID 10

Also, while it is ok to put Epicor in a VM, I would still recommend mapping
drives directly to the SAN and dedicating drives in the SAN to the Program
and the DB.

Also, check with Epicor about moving from Progress to SQL, it's not
something you can do yourself unless you are doing a re-implementation
versus an upgrade.

Depending on why you want to use SQL, you may be able to get the same
benefits by using the Progress replication to SQL

-----Original Message-----
From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf Of
nustepvantage
Sent: Tuesday, August 24, 2010 9:49 AM
To: vantage@yahoogroups.com
Subject: [Vantage] SAN / Epicor 9 Question

Hello,

We are currently running Vantage 406A and will be upgrading to Epicor 9 with
a go live date of April 1 next year. As a part of the project we are also
upgrading our network infrastructure, which includes placing Epicor 9 on VM
Ware and upgrading our SAN. We are currently leaning toward the Dell
Equillogic SAN solution with a Raid 50 configuration. I was wondering if
anyone else was using a similar solution from Dell. We are also considering
migrating from Progress to SQL Server, so the SQL Server database would be
on the SAN.

Thanks,

David Pfiester
NuStep, Inc.
Ann Arbor, MI
734-769-3939 Ext 150




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

Useful links for the Yahoo!Groups Vantage Board are: ( Note: You must have
already linked your email address to a yahoo id to enable access. )
(1) To access the Files Section of our Yahoo!Group for Report Builder and
Crystal Reports and other 'goodies', please goto:
http://groups.yahoo.com/group/vantage/files/.
(2) To search through old msg's goto:
http://groups.yahoo.com/group/vantage/messages
(3) To view links to Vendors that provide Vantage services goto:
http://groups.yahoo.com/group/vantage/linksYahoo! Groups Links
8
> I completely agree with the spindle concept but when comparing apples to
> apples, I'd expect local storage to beat out a SAN every time.

You're right of course, the raw throughput of a directly attached
drive is faster. However, depending on how you're using the storage,
there's other factors that do better under a SAN. Here's a good link
that explains both sides (read the comments too!)

http://www.sqlteam.com/article/which-is-faster-san-or-directly-attached-storage

Mark W.
9
Good Day:

Do not over look SSD's ( Solid State Drives) they are coming down in
price and increasing in size- potentially 15x faster.
Less energy - no heat, no noise.

There is also a unit ( I-Omega?) that connect directly to the mother
board that is even faster - not big.

All rapidly changing technology - still on the edge, but it is the
future.

Speed is a problem in E9, but do not overlook the WorkStation. We
have come to believe that the SQL DB is plenty fast enough, it is all
the layers that have to be navigated to produce the XML\Crystal based
displays\reports that slows things down.


len.hartka@...

________________________________

From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf
Of Mark Wonsil
Sent: Tuesday, August 24, 2010 12:07 PM
To: vantage@yahoogroups.com
Subject: Re: [Vantage] SAN / Epicor 9 Question




> I completely agree with the spindle concept but when comparing apples
to
> apples, I'd expect local storage to beat out a SAN every time.

You're right of course, the raw throughput of a directly attached
drive is faster. However, depending on how you're using the storage,
there's other factors that do better under a SAN. Here's a good link
that explains both sides (read the comments too!)

http://www.sqlteam.com/article/which-is-faster-san-or-directly-attached-
storage

Mark W.





This e-mail and any attachments may contain proprietary and/or confidential information. If you are not the intended recipient, please notify the sender immediately by reply e-mail or at 410-472-2900 and then delete the message without using, disseminating, or copying this message or any portion thereof. With e-mail communications you are urged to protect against viruses.


[Non-text portions of this message have been removed]
10
Good article.

The article kind of implied but didn't outright state that another
factor, besides speed, is cost.

SANs are expensive and you can usually build a DAS solution much
cheaper, especially if you don't buy HP or Dell branded drives.





From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf
Of Mark Wonsil
Sent: Tuesday, August 24, 2010 12:07 PM
To: vantage@yahoogroups.com
Subject: Re: [Vantage] SAN / Epicor 9 Question





> I completely agree with the spindle concept but when comparing apples
to
> apples, I'd expect local storage to beat out a SAN every time.

You're right of course, the raw throughput of a directly attached
drive is faster. However, depending on how you're using the storage,
there's other factors that do better under a SAN. Here's a good link
that explains both sides (read the comments too!)

http://www.sqlteam.com/article/which-is-faster-san-or-directly-attached-
storage

Mark W.




[Non-text portions of this message have been removed]
11
We were RAID 10 on local server with 6 drives (10K SCSI) to the databases and a separate RAID 1 for logs and RAID 1 for system. The NetApp FAS 270 was configured for 14 disks. One disk was hot spare and the rest were RAID DP (special NetApp RAID 6). We created one aggregate with three VM volumes at 500 GIG, one 300 GIG SQL RAW drive and a 500 GIG NAS volume, all thin provisioned. The drives were 10K Fiber Channel drives. The SAN was connected to the VM Ware servers (3) and SQL server via 2Gb Fiber Channel connection with iSCSI 1Gb failover.

My reasoning for the faster SAN performance is the disks dedicated to the RAID array. The more disks, the more IOPs achieved. So say if one drive gives 100 IOPs and in RAID 10, 6 drives you get 300 IOPs; whereas, the RAID 6 with 13 drives gives you roughly 1100 IOPs. Now there is some overhead for RAID 6 but it is calculated according to one website less than 10%. Also, all those IOPs and not dedicated to SQL, we have VM Ware and the NAS portion of the SAN to think about as well. But, the only server really hitting the SAN other than SQL, is our Exchange server. Everything else is nominal. So this is why I believe the SQL server was much faster, and the end user noticed it! We also run several other databases on our SQL server, so this is not all Vantage IOPs. This is how I understand performance. The more drives, the more IOPS, the more performance.

I have just recently read the further comments on this matter. The more spindles, the better performance. Now say if I did use those same 14 drives in the local server and setup a DAS or whatever. Yes it would be faster. We actually had a DELL PowerVault 220S with 14 drives (DAS) connected to our Windows DC for file storage and for backup to disk. This thing was RAID 5 and slow. The con to a DAS solution is you only typically use it on one server. Why have 14 spindles committed to one server versus those same spindles connected to all your servers? Especially since the other servers can take advantage of the 14 spindles. And in a VM Ware environment, SAN is the only way to go.

NetApp has comes down on prices now and we could get a much better SAN for about $20K with double the capacity than the first SAN we bought 3 years ago for $42K. You get data de-duplication and the NAS built in. We are only looking at EqualLogic for the simplicity factor in configuration, replication, and future expansion. The EqualLogic is so point and click and it is just done. The NetApp requires more complexity.


Joe
From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf Of Joe Rojas
Sent: Tuesday, August 24, 2010 10:41 AM
To: vantage@yahoogroups.com
Subject: RE: [Vantage] SAN / Epicor 9 Question



Hi Joe,

I guess I'm surprised you've seen such an increase in performance when going from local disks to a SAN. It almost seems counterintuitive.

How were the disks setup up (RAID Configuration) when using local disks? How much bandwidth do you have between your SAN and Host server?

From: vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com> [mailto:vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com>] On Behalf Of Joe Luster
Sent: Tuesday, August 24, 2010 10:25 AM
To: vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com>
Subject: RE: [Vantage] SAN / Epicor 9 Question

We already have VM Ware, SQL, and a NetApp SAN. We run 8.03. We plan on upgrading our SAN to either DELL Equallogic or NetApp. The performance of our SAN with SQL Server over running RAID on local disks was unbelievable. The end users definitely saw the performance increase. Running Vantage in a VM with 30 users, FRx, and Service Connect (all on the same server); works great as well.

From: vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com> <mailto:vantage%40yahoogroups.com> [mailto:vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com> <mailto:vantage%40yahoogroups.com> ] On Behalf Of nustepvantage
Sent: Tuesday, August 24, 2010 9:49 AM
To: vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com> <mailto:vantage%40yahoogroups.com>
Subject: [Vantage] SAN / Epicor 9 Question

Hello,

We are currently running Vantage 406A and will be upgrading to Epicor 9 with a go live date of April 1 next year. As a part of the project we are also upgrading our network infrastructure, which includes placing Epicor 9 on VM Ware and upgrading our SAN. We are currently leaning toward the Dell Equillogic SAN solution with a Raid 50 configuration. I was wondering if anyone else was using a similar solution from Dell. We are also considering migrating from Progress to SQL Server, so the SQL Server database would be on the SAN.

Thanks,

David Pfiester
NuStep, Inc.
Ann Arbor, MI
734-769-3939 Ext 150

[Non-text portions of this message have been removed]

Joe Rojas | Director of Information Technology | Mats Inc
dir: 781-573-0291 | cell: 781-408-9278 | fax: 781-232-5191
jrojas@...<mailto:jrojas%40matsinc.com> | www.matsinc.com<http://www.matsinc.com> Ask us about our clean, green and beautiful matting and flooring

This message is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake. Please note that any views or opinions presented in this email are solely those of the author and do not necessarily represent those of the company.

[Non-text portions of this message have been removed]



[Non-text portions of this message have been removed]
12
Weird, I sent a response earlier but it never showed up...

Using a VM and a SAN can be great and work well, but you need to
appropriately segment the drives within the SAN. RAID 10 is still the way to
go, and I wouldn't put the DB or program on drives that are inside the VM
files. I would still give direct access to the SAN Drives individually.

I recently was working on this with a customer and they had initially setup
their SAN with 16 drives, all 16 setup in RAID 10 and then the 7 different
VMs on the drives, all sharing all of them.

Performance on 9.04 was terrible.

I wasn't able to get them to set it up the way I wanted to, but I was
eventually able to get them to dedicate 8 drives to the Epicor 9.04 VM
alone, and leave the remaining 8 drives for the other VMs for Frx, APM, etc.

Personally, I would have setup the Epicor VM on 2 drives with only the OS on
it, and them mapped 6 drives outside the VM for Epicor and the DB, that way
the program and the DB have actual access to the spindles for the best
performance.

Yes, I know this sort of takes the "snapshot" aspect away from VMs for
backups, which I know is one of the perks that people like, however you can
do a backup to the VM drive, so that the most recent DB data is backed up
within that "snapshot".

VMs are great, but they still have a ways to go.

-----Original Message-----
From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf Of
Joe Rojas
Sent: Tuesday, August 24, 2010 1:57 PM
To: vantage@yahoogroups.com
Subject: RE: [Vantage] SAN / Epicor 9 Question

Good article.

The article kind of implied but didn't outright state that another factor,
besides speed, is cost.

SANs are expensive and you can usually build a DAS solution much cheaper,
especially if you don't buy HP or Dell branded drives.





From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf Of
Mark Wonsil
Sent: Tuesday, August 24, 2010 12:07 PM
To: vantage@yahoogroups.com
Subject: Re: [Vantage] SAN / Epicor 9 Question





> I completely agree with the spindle concept but when comparing apples
to
> apples, I'd expect local storage to beat out a SAN every time.

You're right of course, the raw throughput of a directly attached drive is
faster. However, depending on how you're using the storage, there's other
factors that do better under a SAN. Here's a good link that explains both
sides (read the comments too!)

http://www.sqlteam.com/article/which-is-faster-san-or-directly-attached-
storage

Mark W.




[Non-text portions of this message have been removed]



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

Useful links for the Yahoo!Groups Vantage Board are: ( Note: You must have
already linked your email address to a yahoo id to enable access. )
(1) To access the Files Section of our Yahoo!Group for Report Builder and
Crystal Reports and other 'goodies', please goto:
http://groups.yahoo.com/group/vantage/files/.
(2) To search through old msg's goto:
http://groups.yahoo.com/group/vantage/messages
(3) To view links to Vendors that provide Vantage services goto:
http://groups.yahoo.com/group/vantage/linksYahoo! Groups Links
13
We did not segment our SAN. We created on big aggregate so the volumes can use all the spindles in the SAN. I do not know what SAN you are referring to, but ours allowed us to create one RAID DP aggregate with separate volumes inside the aggregate; so all volumes share the spindles. I would have configured one RAID 10 aggregate, then your VM Ware volumes, then your SQL RAW drive back to the SAN (separate volume). So the VM Ware system disk is vmdk but the databases would be RAW on the SQL server. (We have separate Vantage and SQL server). If I was using Progress on the same server as Vantage, I would set it up the same way as a VM SQL server.

Our SQL server databases are not in the vmdk but our Vantage server is. With the new SANs you can snapshot your vmdks and the RAW drives at the same time and restore to that same point in time.

From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf Of Ned
Sent: Tuesday, August 24, 2010 2:14 PM
To: vantage@yahoogroups.com
Subject: RE: [Vantage] SAN / Epicor 9 Question



Weird, I sent a response earlier but it never showed up...

Using a VM and a SAN can be great and work well, but you need to
appropriately segment the drives within the SAN. RAID 10 is still the way to
go, and I wouldn't put the DB or program on drives that are inside the VM
files. I would still give direct access to the SAN Drives individually.

I recently was working on this with a customer and they had initially setup
their SAN with 16 drives, all 16 setup in RAID 10 and then the 7 different
VMs on the drives, all sharing all of them.

Performance on 9.04 was terrible.

I wasn't able to get them to set it up the way I wanted to, but I was
eventually able to get them to dedicate 8 drives to the Epicor 9.04 VM
alone, and leave the remaining 8 drives for the other VMs for Frx, APM, etc.

Personally, I would have setup the Epicor VM on 2 drives with only the OS on
it, and them mapped 6 drives outside the VM for Epicor and the DB, that way
the program and the DB have actual access to the spindles for the best
performance.

Yes, I know this sort of takes the "snapshot" aspect away from VMs for
backups, which I know is one of the perks that people like, however you can
do a backup to the VM drive, so that the most recent DB data is backed up
within that "snapshot".

VMs are great, but they still have a ways to go.

-----Original Message-----
From: vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com> [mailto:vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com>] On Behalf Of
Joe Rojas
Sent: Tuesday, August 24, 2010 1:57 PM
To: vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com>
Subject: RE: [Vantage] SAN / Epicor 9 Question

Good article.

The article kind of implied but didn't outright state that another factor,
besides speed, is cost.

SANs are expensive and you can usually build a DAS solution much cheaper,
especially if you don't buy HP or Dell branded drives.

From: vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com> [mailto:vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com>] On Behalf Of
Mark Wonsil
Sent: Tuesday, August 24, 2010 12:07 PM
To: vantage@yahoogroups.com<mailto:vantage%40yahoogroups.com>
Subject: Re: [Vantage] SAN / Epicor 9 Question

> I completely agree with the spindle concept but when comparing apples
to
> apples, I'd expect local storage to beat out a SAN every time.

You're right of course, the raw throughput of a directly attached drive is
faster. However, depending on how you're using the storage, there's other
factors that do better under a SAN. Here's a good link that explains both
sides (read the comments too!)

http://www.sqlteam.com/article/which-is-faster-san-or-directly-attached-
storage

Mark W.

[Non-text portions of this message have been removed]

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

Useful links for the Yahoo!Groups Vantage Board are: ( Note: You must have
already linked your email address to a yahoo id to enable access. )
(1) To access the Files Section of our Yahoo!Group for Report Builder and
Crystal Reports and other 'goodies', please goto:
http://groups.yahoo.com/group/vantage/files/.<http://groups.yahoo.com/group/vantage/files/>
(2) To search through old msg's goto:
http://groups.yahoo.com/group/vantage/messages
(3) To view links to Vendors that provide Vantage services goto:
http://groups.yahoo.com/group/vantage/linksYahoo! Groups Links



[Non-text portions of this message have been removed]
14
We were running our E9 production in VM using vsphere4.1 with Equallogic SAN on the backend, and citrix xenapp on the frontend. But had terrible speed issues, even after tweaking the system per the Dell/Equallogic/VMware tech support gurus. We have since setup a physical box and have gotten twice the performance out of the physical box as compared to the VM setup.(A report that takes 4min on the VM...takes less than 2min on the physical machine) So we are moving our production onto a physical box... while we use the VMs for our Development and testing E9 environment. The VM world is great...just not for production E9...
15
Hey guys - I've been watching this thread and others like it for some
time and I'm struck by the wildly divergent experiences different users
are having. There are may variables as well all know that could affect
the performance but I'm rarely seeing enough detail to quantify why some
experience success and others less so.



I was disappointed by the most recent hardware guide from Epicor
regarding the subject of virtualization and SAN's. I'd love for this
group to perhaps develop a set of guidelines that we could put up as a
help to us all. Are there folks game out there? We would need to
develop first a mutually agreed upon list of specifications.
Participants would need to provide their outcomes and info regarding
their server hardware, O/S, virtual environment and SAN as applicable.



For our part, we are planning on testing a dedicated server using a SAN
via fiber channel. We would dedicate high performance 15k SAS spindles
on the SAN to the Epicor system.



Anyway folks - please sound off. I'm curious if there is resonance (and
frustration!) on this issue felt as strongly by others.



John A. Hatcher

Manager of IS

Versa Products Co., Inc.

(201) 843-2400 x4148

(201) 843-2931 fax





[Non-text portions of this message have been removed]
16
In my experience, a couple of the right SSDs in a RAID1 array as local storage blows away any entry level SAN. It works well if you don't need a huge amount of storage space, or otherwise put only your database on the SSDs. I can't really compare with your setup though because we run Vantage 6.

For a great discussion on this topic (local storage vs. SAN storage) see: http://community.spiceworks.com/topic/101928

Andy moss

--- In vantage@yahoogroups.com, "nustepvantage" <dpfiester@...> wrote:
>
> Hello,
>
> We are currently running Vantage 406A and will be upgrading to Epicor 9 with a go live date of April 1 next year. As a part of the project we are also upgrading our network infrastructure, which includes placing Epicor 9 on VM Ware and upgrading our SAN. We are currently leaning toward the Dell Equillogic SAN solution with a Raid 50 configuration. I was wondering if anyone else was using a similar solution from Dell. We are also considering migrating from Progress to SQL Server, so the SQL Server database would be on the SAN.
>
> Thanks,
>
> David Pfiester
> NuStep, Inc.
> Ann Arbor, MI
> 734-769-3939 Ext 150
>
17
I know I dont post much in here. I keep mostly to the EUG, but this and virtualization is an extremely hot topic right now.

Few notes on VMWare:
Currently it sucks with epicor. Most SMB's can only afford the basics or just esxi. Getting OpenEdge to play nice in VMWare is going to set you back between $12,000 - 30,000 for VMWare licensing alone. I highly encourage people to look into other virtualization technologies. RHEV, KVM, Citrix, and others.

I personally am running RHEV and overall I believe our performance to be great.

Notes on SANs:
Technically you will always get better performance on a physical server. this is in large part because you can setup the disc in a manner that can appropriately fit your needs. I understand the argument that more is better, but this only goes up to a point. Equalogic and most other mainstream SANs have you set them up with one large volume that you then carve up for your needs. This is fine so long as you dont exceed the IOPS that the san can provide.

In relation to virtualization it all comes down to how much RAM you can dedicate to the VM. The only time that you will see a lag in performance with a SAN is if you are doing something that is both CPU and Hard Drive Intensive. Very few application can do this unless you start getting into geological survey stuff where it needs to be reading large amounts of data, processing and then writing the results. In situations like this you should be using DAS anyhow.

When you are looking at SANs and their performance each company has its own ways of improving performance. NetAPP specifically has a high amount of RAM in their SAN's that try to cache what they believe will be the next data set to be used thus making it so it never touchs the hard drives. Each vendor has their own strengths and weeknesses. To pick on NetApp again if you choose to deploy your hard disk in their recommended configuration you loose about 50% of your disk to parity.

When it comes down to the disk, faster is always better. I love SSD, the biggest problem is that there is nothing out there that is target to enterprises. When a vendor such as Dell finally chooses to make this available then it will really take off. The problem you encounter here is you have to look into alternative ways of creating your SAN.

Should you choose to go with SATA over SAS the only acceptable way you can safely deploy it is in RAID 10. RAID 5 and 50 both have inherent problems that can result in complete data corruption. I admit it is a very rare bug, but it can still occur. In addition when you go with SATA its very easy to bet into the 1 tb and 2 tb drives. The problem that occurs here is when a driver fails. Its not a matter of if it will, its when. When setup with RAID 50 or 5 and you start the rebuild on a 1 TB - 2 TB size disk you are not looking at hours to rebuild, but rather days to weeks depending on the amount of regular business activity. There is just too large of a margin for problem here.

SAS and FC are king. In my career I have only had 3 SAS drives fail during their warranty period. And all 3 of those were tagged as failed due to a faulty backplane of an HP MSA. So truthfully I have never had a SAS drive fail. Being able to get the 15k rpm is a massive improvement over the 7200 rpm of SATA.

When it comes to connection mediums such as choosing FC, iSCSI, or NFS there have been alot of advancements. If you have a smaller environment having the hypervisor with between 6 and 8 gig nics will meet any and all needs you may have. However I strongly suggest looking into 10 gigE as the price is very affordable, especially when compared to FC. I watch the performance of my 10 gigE and I am only tapping about 5 - 10% of its capacity. Purchasing 2 of them for the hypervisor and teaming/bonding them for redundancy and performance is very affordable.

As for SAN choices I personally with with a custom SAN that I built with Dell Hardware for around $15,000. I am willing to put money down that it will outperform the cheapest Dell Equalogic and pretty much every other solution out there that is under the $100,000 range.

As for running Epicor on a SAN I beleive that it comes down to a combination of factors that need to be looked at. You can't have one great piece of equipment and then play it cheap everywhere else. I am not saying that you have to buy the best, but rather you need to make sure that all the pieces play well together to maximize your investment. I run Epicor 9.05.602 virtualized on a SAN. For us this is the first step towards fully virtualizing our environment. The startup cost set us back around $32,000 and I am extremely satisfied with the results that I am getting.

Please feel free to email me directly if you have any further questions. zac@... ..Otherwise I will try and do better to keep up with the Yahoo groups.

Sorry if I hijacked this post.

"Zac" Jason Woodward
Network Administrator
Intermountain Electronics, Inc.
O: 877-544-2291
M: 435-820-6515
F: 435-637-9601
www.ie-corp.com

Creating customer confidence through extraordinary service and experienced industry experts.
18
I can attest to the same resulting performance on pretty much an identical setup. We just purchased 2 EqualLogic PS6000's, 3 PowerEdge R710's all decked out, VMWare 4.1 and not sparing any expense in connectivity or anything else... and sadly it's all outperformed by a 5 year old PowerEdge physical box.

We've tweaked everything and nothing seems to improve Epicor performance. Our entire company runs on these SAN's and EVERYTHING and I mean everything else runs better on this new setup except Epicor.

So, I hope they get it figured out at some point. Epicor heralded at Perspectives last year this document that would reveal all the fixes for Epicor on VMWare and how to resolve performance, and said document is still not available as of last week of me trying to get it.

Disappointed to say the least.

--- In vantage@yahoogroups.com, "dcbmichaels" <dcbmichaels@...> wrote:
>
> We were running our E9 production in VM using vsphere4.1 with Equallogic SAN on the backend, and citrix xenapp on the frontend. But had terrible speed issues, even after tweaking the system per the Dell/Equallogic/VMware tech support gurus. We have since setup a physical box and have gotten twice the performance out of the physical box as compared to the VM setup.(A report that takes 4min on the VM...takes less than 2min on the physical machine) So we are moving our production onto a physical box... while we use the VMs for our Development and testing E9 environment. The VM world is great...just not for production E9...
>
19
Said document is available�



https://epicweb.epicor.com/resources/MRCCustomers/VMware_and_Epicor_Deployment_Guide_FINAL.pdf



Hope it helps!


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

*From:* vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] *On Behalf
Of *pbparker
*Sent:* Tuesday, February 08, 2011 4:06 PM
*To:* vantage@yahoogroups.com
*Subject:* [Vantage] Re: SAN / Epicor 9 Question



I can attest to the same resulting performance on pretty much an identical
setup. We just purchased 2 EqualLogic PS6000's, 3 PowerEdge R710's all
decked out, VMWare 4.1 and not sparing any expense in connectivity or
anything else... and sadly it's all outperformed by a 5 year old PowerEdge
physical box.

We've tweaked everything and nothing seems to improve Epicor performance.
Our entire company runs on these SAN's and EVERYTHING and I mean everything
else runs better on this new setup except Epicor.

So, I hope they get it figured out at some point. Epicor heralded at
Perspectives last year this document that would reveal all the fixes for
Epicor on VMWare and how to resolve performance, and said document is still
not available as of last week of me trying to get it.

Disappointed to say the least.


--- In vantage@yahoogroups.com, "dcbmichaels" <dcbmichaels@...> wrote:
>
> We were running our E9 production in VM using vsphere4.1 with Equallogic
SAN on the backend, and citrix xenapp on the frontend. But had terrible
speed issues, even after tweaking the system per the Dell/Equallogic/VMware
tech support gurus. We have since setup a physical box and have gotten twice
the performance out of the physical box as compared to the VM setup.(A
report that takes 4min on the VM...takes less than 2min on the physical
machine) So we are moving our production onto a physical box... while we use
the VMs for our Development and testing E9 environment. The VM world is
great...just not for production E9...
>


[Non-text portions of this message have been removed]