Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Build a Data Center or Contract Hosting?

Cliff posted more than 7 years ago | from the hire-or-pay dept.

Businesses 31

bbsguru asks: "Our Government agency has around 100 independent divisions that share a dozen national applications and a private WAN. We are working to consolidate some of these applications (e-mail, SQL databases, specialized web services), and are facing a familiar choice. One option is to contract out data hosting, e-mail server hosting, and so forth to various vendors (with negotiated SLA's and all the best guarantees, of course). We have already started doing this for our private WAN-to-World gateways, VPN management, and one major SQL application, each with a different vendor, so far. Others are advocating the creation of a national agency-owned facility, where employees would perform these functions instead of contractors. Network management, IDS, data replication and so forth, for all the consolidated applications under one umbrella. Is a series of contractors really the way to go, or are there real benefits to keeping it in house?"

The costs are always a factor, but the one-way nature of the contractor choice is also weighing in this decision. Some are concerned that if the expertise to create and manage these highly custom databases and services is farmed out to contractors, there will be no other choice in the future. Trouble is, as we evaluate our options, the process of contracting out bits of the whole is already underway. With each new contract, one more service to be brought into a datacenter is lost, making the whole thing less practical. Are we swimming upstream here?"

Sorry! There are no comments related to the filter you selected.

Bring it in house (5, Insightful)

Salvance (1014001) | more than 7 years ago | (#17374336)

When you're talking about that many divisions, I'd say bring it in house. It may cost a little more, but the level of control you need cannot easily be quantified with a simple price tag.

I used to work for a 100,000+ employee consulting company, and I saw SLAs and contracts broken time and time again ... or saw situations where companies had to spend millions (or even billions) to get out of contracts and unwind decisions that didn't make sense in the longterm. It was a nightmare to manage a few outsourcing contractors, I couldn't imagine trying to manage dozens or more.

And since you're working for a government agency, you probably won't even be able to achieve any significant cost savings by outsourcing (since most contractors save money by offshoring resources, which I believe is still a no-no for government work).

goverment = wasted money (1)

sinij (911942) | more than 7 years ago | (#17374364)

Contract it out, anything you will bring in-house to government will be drowned by bureaucracy, hidden agendas and general incompetence.

Re:goverment = wasted money (0)

Anonymous Coward | more than 7 years ago | (#17374672)

Oh, yes, because government contractors are such paragons of virtue


Re:goverment = wasted money (1)

smithbp (1002301) | more than 7 years ago | (#17375830)

Just because it's farmed out to a contractor doesn't mean it's going to save any money. I went from working for Dept of Treasury to working for one of their contractors, doing LESS work overall on the same project, and got about a ten thousand dollar raise. Point is, many contractors are billed out at insanely high hourly rates over that which their government counterparts are being paid. The government process and BS doesn't get removed when you bring in contractors either. It actually gets more convoluted and difficult to navigate. Once a contractor comes onto the project, there are more meetings on what should be done, who should do it, is it in the contract or the SLA to do that...blah blah blah. This just adds to more overall time being spent to get something done that someone working for the government and being paid on the proper GS scale could have done. The only problem in that is that government jobs aren't really on par with what someone who is skilled and motivated can make in the private sector. Working for a government agency isn't the great thing that it once was. This means that there may not be the bodies in place to throw into the positions that would be required of staffing a full datacenter on this scale.

We're doing both (4, Informative)

plover (150551) | more than 7 years ago | (#17374360)

Our company has gone both ways. We contract with a vendor to host our mainframes, but we've also built our own secure data center to house the many PC and Unix servers.

Contracting is attractive because the lawyers have this idea that you can sue the hosting service for failing to deliver services as promised. Of course if they fail to deliver, you have roughly three days before your company is permanently crippled, and seven before you are out of business, so that ultimately means only the vultures and the lawyers will get paid; but it sounds like a good idea to management. As a government agency you'll get yelled at and fired, and a few members of Congress won't get re-elected, but you won't go out of business.

One advantage to hosting is that they keep us current with hardware. Our contract stipulates an upgrade schedule for both hardware and operating systems, so we're constantly shuffling in the latest and greatest technology.

I don't know what the price difference is, as I don't ever see those kinds of numbers. But a new data center is mind-bogglingly expensive after you factor in generators, fuel tanks, chillers, security, alarms, power, fork-lifts, flooring, racks, cubes, offices, operators, guards, etc. With a hosting service you're sharing some of that overhead with the other customers of the host.

Running your own data center is good if you have a good team that knows what they're doing, and enough depth to survive the inevitable turnover. We do. But knowing how to successfully run a data center is different than knowing how to build one from scratch -- you need both kinds of knowledge before embarking down this path.

Re:We're doing both (1)

walt-sjc (145127) | more than 7 years ago | (#17378138)

If you are building an ISP that does colocation, it's a WHOLE different ballgame. What you need for your enterprise is very different than what AT&T needs in a colo. Most likely, if you need guards at your building, you already have guards. If you don't need guards, just adding a data center isn't going to create the need.

It also depends on the SIZE of your data center. If it's small, being a room or 3 in a building, in-house management can be cost-effective. When you build it, you use suppliers that fully maintain what they sell - YOU don't need to know how to maintain a chiller - you just need to know who to call when it goes south (or have it automatically report back to the maintainer....) If you are large enough, chances are that you have a mechanical guy that can deal with all the minor stuff anyway. With modern servers, chances are you are not going to need a HUGE data center, although you will have to spend more time planning cooling and power.

I have built and managed smaller data centers (50 or so racks) in the past, and while challenging, they are not beyond the abilities of competent staff. Just make sure you do a LOT of planning, bring in the experts, and check references of everyone bidding on your jobs (don't just ASK for references, really check them...)

Re:We're doing both (1)

WinKing (1043446) | more than 7 years ago | (#17385380)

I think if you find a good combination of this then there is nothing like that. Complete outsourcing is not at all a good idea as you get no control over the quality after some time. And once outsourced you lost the ability to rearrange the things as per your conditions, you get totally dependent on the contractor. And total in housing not only costs you much but also get hectic to manage. So one should find out a great combination of both.

Why put all your eggs in one basket? (2, Informative)

dubl-u (51156) | more than 7 years ago | (#17374382)

If the choice is between screwing yourself by becoming dependent on a bunch of different companies or screwing yourself by turning everything over to one monstrous internal bureaucracy, I'd say go with the former. It might be hard to fire one vendor and turn a project over to another, but it will be completely impossible to fire the central organization.

Really, though, I suspect you've created a false dichotomy. Among the vast soup of tasks you're looking at, some are probably done best by vendors, some by distributed internal staff, and some by centralized internal staff. But even for the centralizable ones, there's no reason it has to be the same center for each one, any more than your phone provider and your electric provider need to be the same company.

The choice isn't about hosted or not (5, Insightful)

QuantumRiff (120817) | more than 7 years ago | (#17374394)

It is about Talent. Do you have (or can get) the people that are able to perform the duties well? Can you pay them competitively? and keep them? The cost of a data center is negligible vs. the cost of downtime due to mistakes, turnover, and bureaucracy. The hosted idea is great to get rid of mundane tasks that are not part of your "core competencies", or where you simply can't get or maintain enough work for a qualified person.

You mention contracts with Service level agreements. If you want to do this "in house" you will need to create these same contracts with the business units that you need. This will give the higher ups the same finger pointing trail that they would have with a hosted solution, as well as the same assurances of reliability. Quite honestly, you would basically have to treat this new "group" as a separate company within the company. I have yet to see a case when it is cheaper over the long run to have a hosted solution, but hosted is much faster to setup and get working. Not to mention, it is awfully hard to re-negotiate, or terminate a contract when a company is holding all of your crown jewels.

I manage a financial data center (4, Informative)

Travoltus (110240) | more than 7 years ago | (#17374634)

My boss's business was swirling around in the same toilet bowl that you are.

We found that for accountability reasons and, in related issues, reliability and reputation issues, we had to bite the bullet, build the data center, expire (domestically) outsourced (er, contractor) contracts, and take it all in-house.

If you're not a financial services company it might be a less dire necessity. If you're a public company of any type? Between you and me, I'd take the data center. For many reasons requiring about a megabyte sized post, SOX will inevitably bite you on the butt when your data is "elsewhere" - elsewhere being anywhere except right there in the data center where you can control its usage in a highly draconian matter. There was also a recent law that came into effect regarding keeping all internal emails.

Contractors don't necessarily screw up, but there's an old war term my pappy taught me that applies here... don't let your supply lines get too numerous or too thin. Too many pipes tend to spring one leak, and nowadays one leak is very bad news. Keep it all in-house and you're statistically guaranteed to have less drama.

Oh and before someone says it, yes, have two data centers. In case the first one becomes the real life setting for "Destroy All Humans" or something.

A couple of points. (1)

cheros (223479) | more than 7 years ago | (#17382252)

As the original architect of the key central government intranet network in one nation, the security manager of a trading platform for another and consulting performance auditor on a third, here are a few things I learned.

(0) Before all, ensure you have someone in charge with common sense, and who has enough power to take decisions (that is, if it's not yourself). If not, you're targeted as scapegoat before you even start. Finding a way to deal with the politics is crucial or you'll be fighting those battles as well as architecting - say hi to new shiny ulcers then.. A good approach is setting up an infrastructure board. Allows all stakeholders to give their view, but leaves the final decision with you or your team. The key is that you really listen to the needs out there - you're building this for a reason :-).

(1) Start with defining your needs as precise as you can make them. I'm talking about matters like sensitivity to downtime and maintenance windows, scalability, interoperability (don't forget 'who with' as a factor :-), standards, etc etc.

(2) Do not assume for one moment that you've captured them all, and under NO circumstance sign a contract on a fixed delivery. One of the most abused tools to completely screw your budget is change control. The vendors or consultants "help" you along in defining what you need, only to then throw it later in your face because something obvious was left out. And real life is not stable either. So, go for incremental upgrades where possible..

(3) KEEP IT SIMPLE. Once again: KEEP IT SIMPLE. Until you get the fundamentals right like budget, overall architecture, platforms, hosting and comms stay well clear of features. You're doing this for a government, so depending on what approach they take to grabbing credit you may face all sorts of demands before you have as much as a working platform. Create a 'feature queue' where you can park all that shiny stuff until the grunt is up and stable.

(4) Make sure you're kept abreast of what (a) politicians get briefed on and (b) what they say in public because they can commit your project to insane deliverables because of , for instance, a lack of understanding. The problem is that if it doesn't get delivered YOU will take the blame, so manage knowledge carefully or get some sort of briefing process going.

(5) Pilot the service, and there are a few caveats here. Make sure you CAN switch if off - define shutdown and very wide maintenance windows, but keep a grip on it - otherwise you may not be allowed to shut down if you're successful which is a pain. Also, don't skimp on the pilot as you need real data and info, not mickey mouse stuff. What you're building has to last a couple of years.

(6) When in pilot, be prepared to change things. That's the whole idea of a pilot. Leave your ego at the door, and that of others too. The pride in creating something that will stand up for years instead of being another headline of failure.

That's off the cuff (no time otherwise). Good luck!

Re:A couple of points. (1)

Travoltus (110240) | more than 7 years ago | (#17385064)

(7) Make sure you're paid in US currency when you're done!

PS: Well said!

Outsource definitely (0, Interesting)

Anonymous Coward | more than 7 years ago | (#17374800)

Especially if you have a lot of sensitive data. In which case you should do what a lot of companies in your boat do (banks, health insurance companies, telcos, etc.)... outsource to some third world country like India, China, Egypt, or a nice state mafia controlled country like Russia. They will definitely give you that sound piece of mind that all your most sensitive data is being taken care of, since we know for sure that they have the laws to ensure that and the resources and ethics to enforce them.

Perfect Market vs. Real World (3, Insightful)

dbarclay10 (70443) | more than 7 years ago | (#17374876)

Something you should consider is "perfect market" vs. "real world".

In a perfect market, outsourcing is the main way of taking advantage of economies of scale. You don't run your own national telecommunications network, you outsource it to the national network. You end up paying (cost - economies of scale + profit). The trick is, if you can reach those economies of scale with your datacenter, and you're a competent bunch, you end up paying (cost - economies of scale). So you can save money. There are obvious security and accountability advantages too.

That's the perfect market. In the real world, these folk charge far more than (cost - economies of scale + profit). They cater to inept organisations who couldn't collaboratively tie their shoes up without a contractor to show them how to do it. So you end up paying (cost_of_incompetents_doing_the_job - economies of scale + profit). The profit part of the equation is miniscule compared to the differences between "cost" and "cost of incompetents doing the job". If the home-grown data center would be big enough (I don't know that it would be, given the brief description in the post), and if it was competently-run, then you can save huge amounts of money by doing it in-house (again, aside from all other benefits).

This post has dealt exclusively with cost. Personally I would consider the other factors (security, accountability) to be the deciding factors, assuming that both options implemented services competently.

Re:Perfect Market vs. Real World (1)

DavidNWelton (142216) | more than 7 years ago | (#17374970)

Good points. Also, while I don't think the question is about outsourced hosting at the cheap end like I discuss, keep in mind that there is something of a "market for lemons" effect: ns.html []

Security? (1)

Krishna Dagli (768334) | more than 7 years ago | (#17375360)

What about security of your data when you do Contract Hosting? Have you classified what you can give away to out side agencty?

My professional opinion (1)

guruevi (827432) | more than 7 years ago | (#17375748)

as an IT consultant and I helped building 2 data centers so far, third coming up. How much are you filling the data centers and at what rate. If you have lots and lots of servers and fill up about a half rack each month, it's usually cheaper to do it in-house.

Calculate what you pay up-front and the general cost of maintaining. With a decent crew, you should be able to manage the data center with 3-5 people. Also calculate in the cost of the ground, power, cooling, ... and of course insurances and see what a co-locator would charge you. You know that co-locators do charge you the costs and a percentage of profit and their prices can change anytime and once your locked into a datacenter, you usually can't get out for cheap.

Re:My professional opinion (1)

walt-sjc (145127) | more than 7 years ago | (#17378224)

You know that co-locators do charge you the costs and a percentage of profit and their prices can change anytime and once your locked into a datacenter, you usually can't get out for cheap.

Your costs can only change at any time if you have a really shitty contract. Contract negotiations can take longer than implementation in some cases. There are so many colo companies out there that you really can dictate your terms if your are persistent. Yes, moving from one data center to another is VERY expensive and a total pain in the ass, so you craft a contract and working relationship with your colo that doesn't require that move.

other gov't nets (1)

tsrimovsky (60498) | more than 7 years ago | (#17375854)

There are several examples of government agencies running their own networks between labs or facilities spread out nationally. The Deparatment of Energy is the first one that comes to mind. The NSF also has experience with this. There are also a lot of regional optical networks at major cities which could be used to help get to where you want to go instead of costly custom last mile solutions from your network vendor of choice.

It might be beneficial for you to
a) See if an umbrella agency has a national net you can leverage.
b) Talk to DOE (ESnet), NLR, Internet2 (Abilene) and get some information that could help you do this more cheaply than going to Qwest and having them price the whole party.

A datacenter? (1)

afidel (530433) | more than 7 years ago | (#17376008)

Remember, you will need two datacenters, whether you build them yourself or outsource you need two. One to be your main site and the other for DR. Depending on your requirements and the available technology in your current/planned solutions you might need as much space in the DR site, or quite a bit less, than in your main site. Failing to plan for DR now will cause you no end of headaches down the road when someone realizes this important piece of the puzzle was left out.

Build your own (1)

thogard (43403) | more than 7 years ago | (#17376924)

It will be cheaper to build your own data centres in the long run than to contract them out.
For a large gov't project, you have to have two for disaster recovery. Selecting the sites will be a political mess but the real issues are "1) where is power and air conditioning cheap?" and "2) where can we get staff"
For this I would propose you put your main data center near a major hydro dam and have your secondary site near where most of your existing staff is. That way when your main site goes down, your people are where they need to fix the most problems. It also helps build up the staff into proper procedures of working remotely which can lead to all sorts of advantages that the federal gov't isn't too happy about even considering.
There are also a number of nice military data centers that the US gov't already owns.

3rd Option (2, Informative)

wired_scribe (199831) | more than 7 years ago | (#17377318)

I think there is a third option that you should consider. Use a co-lo facility. Instead of trying to build your own data center (which is outrageously expensive,) or have someone else manage everything (which is unreliable,) put the servers in an existing data center and manage them in house. I am part of a hosting initiative at my company (we host environments for some of our customers,) and we've either priced out or tried the first two options. We are in the process of spending millions to move from a managed hosting center to a co-lo facility. We have found a 3rd party organization that can handle the hardware portion (if a drive fails they change it out at the data center, they change tapes out during backups etc.) We decided how much of the system they manage, and we take care of the rest. That way I don't spend my time dealing with updating windows and creating users. I spend it managing the databases and applications which I specialize in.

Re:3rd Option (1)

walt-sjc (145127) | more than 7 years ago | (#17378382)

When you look at the long term costs (over 5 years for example,) building your own data center isn't as outrageously expensive as you think... But it depends on a number of factors such as size, competency of employees, how much you need to physically touch your stuff, etc.

Think about pay rates (1)

Amezick (102131) | more than 7 years ago | (#17377380)

Do you want to pay your operators Gov salaries (2.?% increase this year) or do you want to pay them a competitive wage? Think hard about the quality you get for each of these...

consider EC2 (1)

aminorex (141494) | more than 7 years ago | (#17378348)

You could operate your whole network on the EC2 cloud at lower cost. It's not viable if you are dealing with SCI, however.

Re:consider EC2 (1)

eluusive (642298) | more than 7 years ago | (#17380490)

Strange, I've never seen a green woman. What country do they come from? Maybe they just eat a lot of spinach?

The obvious (1)

Marxist Hacker 42 (638312) | more than 7 years ago | (#17378928)

You're working for the government. India cannot be trusted, and even American owned companies cannot be trusted not to outsource to India. At least some part of your data is going to be data about your constituent citzens. Do you really want the front page of whatever paper your Bureaucrats and Legislative body respects to have a headline about you losing data to identity thieves in India?

That alone should argue for you keeping all data in-house and relatively under control.

First step - what do you really need? (2, Informative)

avronius (689343) | more than 7 years ago | (#17381008)

100 independent divisions that share a dozen national applications and a private WAN
  1. Identify your business, and what the basic requirements are.
    • 100 different divisions / private wan, virtual lans, vpns / disaster recover, failover
  2. The next step is determine what services are required.
    1. backend - the infrastructure
      • database servers and various network components
    2. middleware - the application layer
      • application, e-mail, etc. servers
    3. frontend - the pieces that talk to the 'net
      • firewall, load balancers, content switches, intrusion detection servers, web servers or portals
  3. Consolidate where possible - eliminate if not needed - improve if required
    • reduce complexity and duplication of services / standardize on a single database platform (if possible) / standardize on a single web architecture (if possible)
  4. Determine what your SLA's *really* are.
    • 99.99% for network and SAN - less than 5 minutes of downtime each month
    • 99.9% for major services - less than 45 minutes of downtime each month
    • 99% for individual servers in redundant or failover groups - less than 8 hours of downtime each month
After you've taken the time to gather that information, and get a real understanding for scope, you should begin to look at your various options.
  1. Partial outsource - retain everything internally except for the actual server room(a)
    • rent or lease rack space from a third party
    • provide your own gear
    • continue to do all administrative tasks - including OS and hardware support
  2. Partial outsource - retain all administration, with all hardware support from vendor(b)
    • rent or lease server and rack space from a third party
    • continue to do all administrative tasks - including OS
  3. Partial outsource - retain all administration except for OS - all hardware support from vendor(c)
    • rent or lease server and rack space from a third party
    • obtain OS support from third party
    • continue to do all administrative tasks - except OS
  4. Piecemeal outsource - each service provided by different vendor - you become contract administrator
    • rent or lease server and rack space from a third party
    • contract out OS support to different third party
    • contract out application / database administrative tasks to other third parties
  5. Inhouse - this options requires a large initial expense, but results in the greatest overall control
    • requires facilities administrator - to take care of power / cooling / space allocation
    • If the facility does not currently exist, or requires expansion...
      • purchase generators / air conditioners / humidifiers or dehumidifiers / racks / network wiring / fire supression components / wiring harnesses and conduits
      • construction - server rooms are considerably more expensive than normal office space
    • OS support and administration / application and database administration / network administration
Only once you have a thorough understanding of the current state, a more robust array of options, and an understanding of cost vs. control will you be able to make the right decision. This is not meant to be a complete template, but should allow you to see the steps that are required more clearly.

Own the facility, hire contractors to set it up (1)

Tjp($)pjT (266360) | more than 7 years ago | (#17384224)

I have planned datacenters in the past and it is a pretty easy choice for a large government agency. You save taxpayer money in the longrun if you own the facility given the size you are needing (assumption). Hire contractors (hey, we are available!) to set you up and get it organized. Create a handoff plan where your own captive staff or long term contractors are brought in and trained by the folks contracted to create the data center. You don't want the high priced talent long term that is needed for setting up the data center in the first place (but they do save you money in setting one up as the skip all the nasty restarts and common errors). Also decide on what your expectations and requirements will be. Do you need 99.999% uptime, e.g., how many hours / minutes a month or year of downtime is acceptable. Plan for the anticipated capacity. And then make sure you can grow over the next several years without moving. Agency a target for terrorism, located in a area where disaster is a consideration (it really is everywhere)? Then plan an ancillary datacenter in your disaster plan. What disaster plan? Well make one! Look for someone who can tell you the advantages and disadvantages of technology choices. Like what UPS systems are out there, and battery backup versus flywheel based UPSes, and fuel cell vs generators. Make sure environmental (as in HVAC, not bunnies, though they count a bit too) concerns are considered especially capacity planning for heat load and growth. What seems like a great deal in air handling equipment may be limited in the future. Decide on what you can tradeoff in personnel to staff the datacenter versus distributed ownership of datacenter issues. Create a trouble ticket and problem escalation plan. When a small problem is not quickly resolved you need to know why and escalate the issue. Who're you going to call at midnight when the staff is on holiday break and the email goes wonky? Plan in advance with a clear set of requirements. I could add more but this is a start. Oh, and do you need to keep your data securely? In house is arguably better, but in the governments case maybe not. So make sure equipment is properly disposed of if it has had sensitive data. Maybe consider a datacenter architecture and choice of products to make that easier.

Best Guarantees? (1)

bill_mcgonigle (4333) | more than 7 years ago | (#17384242)

I bet you're not getting 'Best Guarantees' because you didn't define them.

I also bet if you define them to your satisfaction, none of the vendors will actually agree to signing the contract. Then you can take that to your boss and he'll understand why you need to do it in-house. Don't be afraid to hire contractors to give you a hand in-house too. Pay them to document their work so you have records and little loss of knowledge.

Bob Cringely wrote recently about the SLA on his Internet connection. The LEC told him that he didn't qualify for the free month of service because while his Internet connection was out for [insert long time here] the Baby Bell circuit was up from his house to the CO. Never mind that he couldn't use it or that the circuit wasn't owned by the company providing the Internet service... long story short - get your lawyers to define the heck out of everything. Say, "how would you wriggle out of this contract?" as a starting point for discussion (more delicately, perhaps, depending on the personalities involved).

Data Center busines, dedication and curve.... (1)

freaker_TuC (7632) | more than 7 years ago | (#17386248)

Introduction, dedication and learning curve

My work is in hosting and administrating a server and client farm. I am both web programmer, system integrator, linux guru, part time designer, programmer, electronicus and artist all-in-one. The list certainly doesn't stop there which makes it a science and me a system engineer. It's a continuesly going learning curve. I've got many books for many years about mail, dns, Apache, Perl, Bash, bibles which have helped me further for about 14 years active unix duty now. It's also a dedicated job where you need to accept overtime, when it's not working you will have to search till it works...

I like to do my work, and that's where the dedication starts to keep your serverfarm also working; liking to do it all, the challenge of getting and keeping it working, whenever it pops to have the stress and guts to get it working again over and over; rinse, leather, repeat; wether you'd be administrating a small or a big serverfarm, the time you put in it would be almost equally; if everything is set up with scripts on the back-end, easy managable or having a accounting officer, getting synched with a local disk through usb/network locally, you could handle from 5 till 100 servers...

All the system administrator has to do is keep his systems updated, tightly controlled; know networking ethics like firewalling with +1 systems, put the databases and websystems on a seperate interface (of the firewall) and stay updates with the newest trends, exploits and technologies to be protected and well from the (evil) internet and stay rendundant for their purpose.

For that purpose only someone on-site would be best for those systems and also someone who knows your internal networking system and resources; which is mostly private company information ;) The system administrator can not be afraid of changes and implementations; as long as the management DOES follow and appreciate his ethics, dedication and corporate decisions needed to keep his systems up and running in the air.

The operating systems and social behavior sets a lot

I've chosen for Freebsd and linux, which are very high maintainable and advanced tweakable OS's (already working with since the early versions). If you got a very good system administrator, together with a team working together for a good system maintainability you got the best and smoothest running server farm on-site.

Wether on or off-site, be sure to make agreements with your ceo and/or employees about system integrity like programming perks, php scripts, perl bugs, injects, etc.. with the web programming team; how passwords get used; redundancy; system overusage, misusage, hacks and dealing with this. Keeping the clients clean keeps the network clean; keep the networks and clients clean and you will keep your servers (and customers/employees) happy too ;)

why on-site is important?

Still, I would like to keep in mind, an on-site maintainer is the best option for a lot of options where you need local access to your systems with; either that or the required hardware and wiring to keep connected with your farm. Having an extra machine to see the entire overage status of the entire network, packet statistic monitoring and machine status can best be done with an external machine only listening with a RX to the network ; let it respond furtheron on a different interface so you can check your network with Nagios, bigbrother or any similar monitoring tool. MRTG has been working like a charm for many years here.

I've had needed several system restarts of not so needed machines which are "experimental" machines, working in development fases for new projects our tests; your system engineer will sure not like to be driving for that, to be on-site to perform such tests (atleast not without having the right equipment there to hardwire reboot)...

Cost effective, faster and easier

  • will be faster to access locally
  • not required cost to have a second dedicated line to maintain and accessyour serverpark
  • easy to push the reset button
  • more easy to maintain hardware/network
  • more secure if handled right
  • image to your own business
  • expandible if done right
  • more cost effective; only if you got the team behind you and sharing the same values of the companies networked assets...
  • same network could be used for the company infrastructure
  • needs own backup and response strategy


  • Needs dedication of atleast one individual with knowledge
  • Needs atleast one dedicated line
  • Can sometimes cost more to rent the same equipment off-site
  • Needs air, monitoring, security and power rendundancy

Off-site notes
  • SLA's going bad could ruin a company if not handled fast/right off-site
  • off-site has their own monitors and response teams, upgrades and fixes mostly in the (more expensive) hosting package

I probably have forgotten some sidenotes, but this will give an idea what you will really need on and offsite and how things can be handled (better) through on-site if everything is maintained right... I've got to work ..
Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?