Fair enough. :) It also helps to read more than one post before replying, when you come into a thread with quite a few posts. LOL
-ASB: http://XeeSM.com/AndrewBaker On Thu, Mar 25, 2010 at 1:41 PM, Jonathan Link <[email protected]>wrote: > Independent corroboration is never superfluous! > > On Thu, Mar 25, 2010 at 1:33 PM, Andrew S. Baker <[email protected]>wrote: > >> Thanks for making my commentary superfluous, SC. :) >> >> -ASB: http://XeeSM.com/AndrewBaker <http://xeesm.com/AndrewBaker> >> >> >> On Thu, Mar 25, 2010 at 10:11 AM, Steven M. Caesare < >> [email protected]> wrote: >> >>> We have seem extended server life as well, for a variety of reasons, one >>> particular of which having been a significant number of applications >>> being deprecated, and the development effort is happening on new >>> platforms, so there's no incentive to upgrade the old systems , as they >>> will be EOL'ed once the user base is migrated. >>> >>> For boxes that may have been failing, we simply VM'ed many of them >>> (often bumping up the resources available to them in the process). >>> >>> With the advent of virtualization, I see us adding/upgrading VM servers >>> on a semi-regular basis, increasing the resources given to VM's and/or >>> migrating the heaviest ones to the new boxes, and slowly retiring the >>> old. >>> >>> The life cycle I suspect look similar to what it did for the physical >>> boxes (3-5 yrs with some maint. Costs), but the VM's they host will >>> likely be much more fluid... >>> >>> -sc >>> >>> > -----Original Message----- >>> > From: Charlie Kaiser [mailto:[email protected]] >>> > Sent: Thursday, March 25, 2010 10:01 AM >>> > To: NT System Admin Issues >>> > Subject: RE: How would you go about this? >>> > >>> > +1. >>> > While 5 or 6 years ago 3 year server replacements were the norm, >>> that's no >>> > longer the case. By the time you put together server cost, OS license, >>> and >>> > migration consulting costs, a small business is unwilling to pay $10 >>> or so to >>> > upgrade their SBS box or exchange server just because it's old. >>> > We're running into many more aged hardware issues than we used to, and >>> > some of them are ugly. >>> > >>> > *********************** >>> > Charlie Kaiser >>> > [email protected] >>> > Kingman, AZ >>> > *********************** >>> > >>> > > -----Original Message----- >>> > > From: Erik Goldoff [mailto:[email protected]] >>> > > Sent: Thursday, March 25, 2010 6:29 AM >>> > > To: NT System Admin Issues >>> > > Subject: RE: How would you go about this? >>> > > >>> > > "You get five years out of a server? I think you need the help." >>> > > >>> > > >>> > > >>> > > *or* YOU are luckily spoiled ! >>> > > >>> > > >>> > > >>> > > Yes, a 3 year lifecycle refresh is ideal, but not realistic >>> > > budget-wise for MANY out there in the real world. Especially in the >>> > > SMB market, I frequently run into aging servers with some of my >>> > > consulting clients. You'd be hard pressed to convince them to >>> replace >>> > > a server that is currently working as expected with new hardware >>> > > and/or new OS without proving any significant benefit in features >>> over >>> > > the existing systems. The biggest issue on aging servers that I see >>> > > is drive failures, and insufficient drive space/size due to data >>> > > growth. Data volumes can be replaced/upgraded without an entirely >>> new >>> > > server in many if not most cases. >>> > > >>> > > >>> > > >>> > > That said, we all know that Windows 2000 ( all flavors including >>> > > servers ) are dropping from Microsoft support July 12th this year. >>> So >>> > > the lack of support, service packs, and vulnerability fixes *will* >>> be >>> > > a driving factor for OS upgrades which work out well with hardware >>> > > upgrades >>> > > >>> > > Erik Goldoff >>> > > >>> > > IT Consultant >>> > > >>> > > Systems, Networks, & Security >>> > > >>> > > ' Security is an ongoing process, not a one time event ! ' >>> > > >>> > > From: Holstrom, Don [mailto:[email protected]] >>> > > Sent: Thursday, March 25, 2010 9:15 AM >>> > > To: NT System Admin Issues >>> > > Subject: RE: How would you go about this? >>> > > >>> > > >>> > > >>> > > You get five years out of a server? I think you need the help. I was >>> > > just looking for some help in picking up a file server. I replace >>> all >>> > > my workstations and servers every three years. But I only have 130 >>> > > workstations and servers. >>> > > >>> > > >>> > > >>> > > Your growth estimate is OK as it increases here at the Museum. That >>> is >>> > > why I am splitting the data onto several HDs. >>> > > Thanks for your help. >>> > > >>> > > >>> > > >>> > > From: Jonathan Link [mailto:[email protected]] >>> > > Sent: Wednesday, March 24, 2010 4:18 PM >>> > > To: NT System Admin Issues >>> > > Subject: Re: How would you go about this? >>> > > >>> > > >>> > > >>> > > I'm not going to answer your question, instead I'm going to pick >>> apart >>> > > your request. >>> > > >>> > > We really don't have any idea of what your rate of data growth is. >>> > > There are two estimates we can make from the data supplied, linear >>> > > growth or geometric growth. With linear, you're adding about 125 GB >>> > > of data per year. With geometric you're doubling your data every >>> ~19 >>> > > months. So, if you expect the same growth rate, in 5 years (assumed >>> > > life of a >>> > > server) you're at either +625 GB of data or over 8 TB of data. >>> > > >>> > > Just taking a step back and looking at it from 30,000 feet, a server >>> > > is the least of your storage concerns if you're doubling your data >>> > > every 19 months or so. >>> > > >>> > > On Wed, Mar 24, 2010 at 3:38 PM, Holstrom, Don <[email protected]> >>> > > wrote: >>> > > >>> > > I have a file server that has gone above 1 TB. When I first came >>> here >>> > > to the museum a few years ago (8), they had 33 gigs of data on one >>> > > server. I brought in file tape backups until last year when the >>> backup >>> > > went out of that range. >>> > > >>> > > I always used SCSI RAIDs but even now that is a bit high. >>> > > >>> > > So >>> > > >>> > > I have ordered a new file server with six HD openings. I am figuring >>> a >>> > > pair of 10,000-rpm 150 or 300 gig HDs for the OS, I can go Server 03 >>> > > or 08, figuring on 08. I would back up one with the other. Then for >>> > > data, two 2TBS backed up for the main data and two 1.5 or less for >>> > > other data, also backed up. >>> > > Then I could/would backup to external 2TB drives for longevity. >>> > > >>> > > What thinkist thee? Is there another way I should go? Data here will >>> > > continue to increase at the same rate... >>> > > >>> >> >> >> >> >> > > > > > ~ Finally, powerful endpoint security that ISN'T a resource hog! ~ ~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/> ~
