Slashdot Mirror
How Do You Evaluate a Data Center?
mpapet writes to ask about the ins and outs of datacenter evaluation. Beyond the simpler questions of physical access control, connectivity, and power redundancy/capacity and SLA review, what other questions are important to ask when evaluating a data center? What data centers have people been happy with? What horror stories have people lived through with those that didn't make the cut?
Beyond the simpler questions of physical access control, connectivity, and power redundancy/capacity and SLA review
Well first of all, I don't know that I'd write any of those things off as "simple". But some other points worth looking into would be:
Cable Management (over or under floor)
Cooling Capacity and Redundancy
Power Quality (not just redundancy)
Age and Condition of Electrical Hardware (ATSs, STSs, UPSs, Generators)
Outage/Uptime History
Fire Suppression System and Smoke Detection System
Maintenance records
Maintenance records
Maintenance records
you evaluate it by how many cups for frosty piss it can...
wait...
in soviet russia, data centers evaluate you!!!!
Yeah there we go.
Look at a datacenter's history [recent and past], outages, maintenance issues, customer support, management and etc, in conjunction with their listed redundancies and capacities.
Just because they have two electrics going to each server, doesn't mean a random maintenance tech will flip the wrong switch. :)
set it on fire, throw floods at it, generate tornados, then top it off with a nice earthquake.
Make sure you ask about any past outages and how they were handled, I have seen data centre power that has failed 4 times in 1 year due to the same problem that was only found on the fourth outage.
Let's not forget about watts per library of congress.
Conscience is the inner voice which warns us that someone may be looking.
I ran a data center long, long ago. My sales guy knew it wasn't going to pan out and threw me to the wolves. He asked me to start the tour, and then he took a long lunch to miss it.
The guys I gave the tour to seemed very intelligent. They only spent about 60 seconds on our data center. The instant they saw the carpet, their eyebrows were up. When I didn't lie to them that there was no diesel generator on the other side of the (secretly dead) batteries, they did exactly what they should have and stormed out without saying thanks.
slashdot: where everyone yells sarcastic metaphors to themselves to understand the issue
I'd also ask:
Number of years in business.
Involvement of the owner in the current business.
Number of years the current owner has been in this business.
Also do a check with the Better Business Bureau to see what, if any, complaints had been filed.
And, as always, Google is your friend -- definitely do a search for the business you are considering along with the word(s) problem, issue, complaint, praise, etc!
Pull floor tiles and compare the amount of obsolete technology-- Thicknet cables, VAX cluster interconnects, water chiller hookups, FDDI cables, etc. with the amount of space remaining.
Anything less than 4 inches of obsolete crud isn't worth excavating. Leave it a few more years.
--Joe
I co-locate at a data center in Alberta. It is in the basement of a high-rise building. Because of this there is much traffice in/out of the building. The main doors within the building leading to the datacenter itself can be opened with a credit card, or even a set of keys (um, any key). This poses a security risk. Even though you'd need to know exactly where to go and when (so as to not bump into people working there), it is still possible to get what you're after realtively simple with no alarms. They do have cameras though, so wear a mask - and since this is Alberta, no one would question the mask.
I picked up one of my servers a couple days ago, and they didn't ask for ID either. I could have been ANYONE.
... If my direct manager evaluates it as good, then im good. If my direct supervisor evaluates it as fucked, then im fucked.
Find someone you trust who's already a customer. Word of mouth beats any number of white papers or studies or guarantees.
I'm assuming this is evaluating for co-location purposes. Here are some things I'd ask.
1) How quickly can I get a new server deployed into it? How do I do it?
2) Can I get a tour? Now? (Note that this not only lets you see the data centre, but also will give you an idea of security. Look for procedures on getting in, notice if they ask you to sign a release form, etc.)
3) How close to capacity are you? (The answer should include space, floor weight, power, cooling, and network. If it doesn't, why not?)
4) What are your racking/networking/cabling standards? (They should have some, at least where you connect to them, but they shouldn't be onerous).
5) How many people manage the data centre? You don't want to be one car accident away from loss of access or service.
6) How about power management? Is the centre on a UPS, redundant UPSes, or nothing? Can you get charts of the power going to the servers? Can you get DC for telecom servers, or only AC? Is it on a generator for long-term outages? (Note that you may not need this--in which case you shouldn't pay for it. Alternatively, if you need it, make sure it's there!)
7) Is it manned 24/7? (Ditto!)
If you can, ask them to pull a tile so you can see under the raised floor. Underfloor cabling (and suspended ceiling cabling for that matter) should be neat, tied, and labelled. Dead cables should be pulled, not left to rot. There has to be sufficient clearance for unrestricted airflow. Cages are better than lying on the floor.
Most of what makes a good data centre comes down to organization. If it's a rats nest, then even if there's one guy who knows "everything," it will be less reliable, less consistent, and less predictable. Procedures should be written down, printed, filed in labeled binders, and regularly updated. (Note: Online copies should be canonical, but also needs to be accessible offline when shit --> fan.)
Fire suppressant mechanisms (wet vs. dry, live pipes, etc.) need to be considered, as does emergency lighting. If the operators need to start digging around for a flashlight to read what they should be doing, then things aren't happening the way they should.
Be picky. If they're leasing space to you, then their data centre design and maintenance is their BUSINESS, and they had better get it right! Look for a neat, well-organized, well-documented, well-panned data centre. Also make sure that it fits your needs.
"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban
There are basically 3 perspectives from which to evaluate the Datacenter. They're pretty well universal to any IT eval. People, Process and Technology. The datacenter facility itself is only one piece of the puzzle (Facility = Technology, which only accounts for a fraction of the total cost of operating a Datacenter). There are also the people running the datacenter and how they are organized and interact with the technology, one another, and their customers (internal and external). From a people/process standpoint, if you want to give a general "score" to them, you can assess them against the SLM maturity scale. (Read about the Gartner Maturity Model for Infrastructure and Operations) Evaluating a datacenter is going to be a balance between the cost of operating the datacenter and the level of service you require from said datacenter. There really isn't enough information in the question to give you a good answer. Are you looking at evaluating the acquisition of a datacenter to grow into, are you looking for a managed services DC to host your gear with operational support? Are you looking for rack space with pipe and power? If you give more details to your inquiry, I'm sure the community can provide you with some great answers.
based on the Data Center Chicks of course!
Monstar L
Regardless of how well they are decked out, always start with a "pilot project". Start small for a short period to evaluate real world performance of both their equipment and their tech support. We currently have a pilot project in place to evaluate a datacentre for outsourcing our compute requirements. We have learned that while they have exceptionally good equipment in place, their responsiveness and ability to provision is highly questionable.
Feed the need: Digitaladdiction.net
Read their disaster preparedness plan. If you can get through it without your BS alarm ringing off the wall or laughing hysterically, there is hope. You do have a disaster preparedness plan, right?
What does your company _NEED_? How much bandwidth do you need? What kind of servers do you need? Are you looking for Co-Lo or Dedicated? If you're doing Co-Lo, how much power and space do you need? If you're doing dedicated, do you need managed or unmanaged? PCI compliance? HIPAA compliance? Do you want to pay for certain redundancies? Do you need an Uptime Institute Tier certified facility? I could go on and on. The one thing that you need consistently is good customer service. The rest depends on what you need. Full Disclosure: I work for one of the biggest privately held dedicated hosting companies on the planet.
Since the odds are I'm going to be spending the night there at some point, good vending machines or a cafeteria are a must.
"Have you ever thought about just turning off the TV, sitting down with your kids, and hitting them?"
Power from the ceiling, data under the floor.
The reason is data centre floods don't occur very often but when they do the d.c can tolerate the data cable being in water but when the power gets in contact with water circuit breakers trip and they don't work again until they are dry.
I encountered it when the AC water feed burst and co-incidentally the drain for the data centre had been blocked. If your power and data are through the floor then I would suggest that you invest in a good wet and dry vacuum cleaner. I do have other suggestions but this seems such a basic thing to me.
My ism, it's full of beliefs.
1) Hookers
2) Beer
3) Illicit drugs
Seriously, my top three are as follows ....
1) Bandwidth Available / Oversubscription rate
2) Geographically different alternative location.
3) Disaster Planning directives.
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
After redundant power, adequate cooling, 24/7 ops...etc. Neatness is the most important thing. Making each rack very neat with redundant power bars and individual patch panels is key for the long term. Its when you allow cables to run under the floor and between racks that things get out of control. We have an old timer that will yell at anyone who does anything messy and its works out very well.
Rods/hogsheads...
sleepy now
Linux, you magnificent bastard, I read the fucking manual!
Such as street access. Is there more than one way in, if the access road was closed off (police incident, subsidence, civil unrest - depending where it's sited), what would happen. Could staff get to work, or leave for home?
Ease of recruiting / retaining sufficiently qualified staff in the locale, or persuading your to commute or relocate
Is the on-site restaurant / canteen or local eateries likely to give everyone food poisoning (this could be a single point of failure)
Local crime rate - number of times the facility has been broken in to - even the amount of graffiti on the walls could be a negative indicator
politicians are like babies' nappies: they should both be changed regularly and for the same reasons
Everyone posts the "questions" to ask; no one posts the acceptable "answers" to those questions. Way to regurgitate uselessness, but I guess posting (even if it's useless) makes us all feel helpful/smart/educated/witty.
- Ask them how many Amps their electrical cords are tested for.
- Make sure you ask there EMI rating in Ohms / squared Newton times their rack cabling coefficient.
Yes, I'm mocking you...
You forgot a few:
- Enough qualified *on site* staff 24x7 to deal with all clients including yourself
- 24x7 phone support, with people who understand English and have immediate access to the techies
- Company financial records and history (You don't want someone almost broke or a new startup with no backing)
- These days availability of virtualisation solution and supporting hardware (depending on your application, if virtualisation is an option)
Oh and your emphasis on maintenance records may be a little misplaced. They can be faked. They also may not be available due to security concerns (of their other clients). *IF* you can get hold of them they should be complete. Hardware service level should be part of the agreement and service schedule should be part of that.
These posts express my own personal views, not those of my employer
When I worked at a corporate office in Maryland, they used the building's air conditioning to cool the server room.
This worked well until the outside temperature got down to about 15 degrees Fahrenheit, but then it failed miserably: the outdoor condensers no longer functioned, the AC shut down, and the entire IT department went into a panic.
The first time this happened, I (a lowly Help Desk tech) suggested to the CIO that he run a duct into the room from the outside: a simple fan would bring in enough sub-freezing air to cool the servers.
The second time it happened, the look on his face told me he hadn't taken my suggestion seriously enough.
The third time, he flipped a switch and the fan cooled his server room just fine.
More important than the technology is the policies and training of the personnel running the operation. It will fail, eventually: It always does, no matter how well its designed or what with promises of infinite uptime. So walk into the data center and count the number of people wearing hiking boots, divide by the number of racks, and there you go. The most grizzly looking guy wearing hiking boots usually knows everything. He also usually has a lighter and a screwdriver if you ask.
I don't know why this is...
#fuckbeta #iamslashdot #dicemustdie
Security:
manned or unmanned 24x7
Closed Circuit cameras
How people are granted access to the facility
Do you get a cage, a rack, or space in a rack?
If a rack or space in a rack, how is your equipment secured?
How do you grant access to vendors if they come to swap a hard drive or motherboard for you?
Facility:
How close are they to their maximum power consumption?
How do they propose to scale once they reach maximum?
KW per Rack limits
How many power grids is the facility tied to?
Localized fire suppressant or general "drench them all with water"?
What kind of backup power generators and do they have access to fuel?
Do they have a priority relationship with fuel suppliers in the event of an emergency?
Do they have contiguous space available? If so, how much?
What is their plan for growth if they fill all available floor space
Network:
How many carriers are present in the facility?
Can you bring your own in, or do you have to share?
what is the cost for cross-connects?
Acct Mgmt & Processes:
Billing process
Do you have a dedicated account team?
What is the process for dealing with SLA violations?
Hands-on:
How many hours and what skillsets come with the contract?
What is the hourly rate to do simple tasks like swap tapes?
mmmmmmmmm raw
I used to have a large cage in an Exodus colocation facility. Turns out that if we wanted to put in an EMC Symm5 (these are three tiles wide), we would have to rent a fork lift and put it through an open rollup door on the second floor. Their "freight elevator" was barely big enough for two people and a dolly.
One of my other cages was housed in a Global Crossing facility; when they started to run out of out cooling, they would hook up huge external A/C units in the parking lot and run 2ft diameter ducting to a hole in the wall. If you happened to walk near one of these openings you'd be greeted by freezing 50mph winds.
Anybody find it odd that Exodus bought Global Crossing, who then went out of business?
I am the Director of Operations for our DC. When we give tours, I explain the following (pseudo order of the tour):
- Begin with the history of the building, when it was built (1995), why it was build (result of Andrew in 1992), and how it is constructed (twin T, poured tilt wall).
Infastructure:
- Take you through the gen room, show you it is internal to the building, show you the roofing structure from the inside, explain the N+1 redundancy, the hours on the gens, when they are ready for maintenance, how they are maintained, by whom (the vendor), how the diesel is stored, supplied, duration of fuel at max and current loads. Explain conduct before a hurricane or lockdown, how we go off grid 24hours ahead of a storm, mention our various contracts for after storm refill and our straining / refill schedule.
- Take you to the switch gear room, explain the dual feeds from the power company, how the switch gear works, show you the three main bus breakers, show you the numerous other breakers for various sub panels, etc. Explain and show you the spare breakers we have in case replacement is needed.
- Take you to the cooling tower area, explain the piping, the amount of water flowing, the number of pumps, how many are needed, the switching schedule, explain the N+1 capacity and overall capability of the towers, explain maintenance, show you the replacement pumps in stock, explain the concept of condensed water cooling if needed.
- Take you through the UPS and battery rooms, explain the needed KW capacity, what the UPSs back up and what they do not. Show the various distribution breakers out to floor, their capacity, the static switches, bypass, explain the battery capacity, type of cells, number of cells, number of strings, last time the jars were replaced and how they are maintained. Explain max capacity of the load vs time. Answer questions relevant to switching from utility->UPS->generator and back.
Raised floor:
- Take walk on raised floor, explain connectivity, vendors, path diversity we have, how the circuits are protected. Show them network gear, dual everything, how we protect from a LAN or WAN outage, and specific network devices we have for DDoS, Load Balancing, Distribution, Aggregation. Explain how telco and others deliver DS0 to OC-12 capacity, offer information on cross connections regarding copper, fiber, coax. Explain our offerings (dedicated servers up to 5K sq ft cages) and ask what they are interested in.
- Explain below the floor, size of raise, that power and network is delivered under, what are on level one trays, level two trays, and the piping for cooling. Show the PDU units and how they related to the breakers in the previous rooms. Show them the cooling panel and leads out to CRAC units, explain the cooling capacity, plans for future cooling, explain hot/cold aisle fundamentals, and temperature goals. At this point, there are usually more questions about vented tiles, power types available and overall floor density in watts/sq ft.
- Explain the fire detection / mitigation system, monitoring of PDU's, CRAC units, and FM200. Explain the maintenance of the fire system, show them the fire marshal inspection logs and the panels that alert the police and fire departments (both on floor and in our security office in front).
- While finishing the walk on the floor, show cameras, explain process to bring in and remove equipment, tell them the retention on the video, explain the rounds the guards make, the access list updates and changes.
NOC:
- At this point we're back to the front of the building, go into the NOC, explain what we are monitoring (connectivity, weather, scheduled jobs, etc). Introduce NOC and security staff, explain they will always get a person if they call, submit a test ticket from a e-mail on my phone, they will see the alerts light up and the pager for the NOC will signal. The final steps are to introduce them to security and then I'll lead the customer(s) to the conference room so they can continue the conversation
So there.
I had a machine colocated in a very nice, secured facility right in the middle of a major city where all the telco wiring runs. It was awful for these reasons:
- they advertised 24 access to your equipment on the web site, then the smarmy salesperson explained how that's actually not going to happen. That should have been it right there, but I was dumb.
- later, they had a brief power outage due to a contractor f-ing up one day, and I was never notified. This in turn disabled my traffic shaping configs, which I intentionally do not have running upon startup. I didn't know anything was amiss until I got a huge bill for bandwidth overages. I had to fight heavily with them to overturn the charges because I had a contract with them that said 100% uptime in regards to power. They disabled my controls by not upholding their deal, and were trying to pin the results on me. Afterward I put in a simple script to email me when the maching mysteriouly rebooted. The whole time they were acting like jackasses about it, then acting like they were doing me a huge favor when they gave in.
- Then, the worst offense. I took off for a three day weekend right when they cold-cocked my machine during a maintenance operation. I didn't notice until the following sunday that things had been down all weekend. I had missed my opportunity to visit my machine since it was just after 6:00pm, and I'd have to wait until 9:00 the following morning to see what the hell was wrong. Or they offered to let me PAY THEM to look at it for me sooner. I declined.
When I got there, they had unplugged my machine, moved it to a new location, failed to power it back up, and had the network cable in the wrong port. All of those things were in total violation of my contract. When giving me excuses, they were saying that because the network lights were on they thought the machine was powered on.
Then it was like all hell to get them to come through on their contractual provisions when they don't provide the guaranteed uptime and exhibited severe negligence.
I eventually got paid back what I had paid for the service that month, but not any of the reimbursement specified in the contract for exceeding downtime. And it took two months for them to return my money.
Anyone have worse stories?
Is it near a hazardous materials facility? Under an airport approach or departure path? A rail line? A major highway possibly having traffic hauling hazardous goods? Gas, oil or chemical pipelines? Flood plane? etc.
- data redundancy, offsite specifically
- ability to cut over? ie what happens if there's an earthquake, are your services to the world down until everything is replaced and backups are restored?
- what do you have on hand for hot spares in the event of equipment failure?
- when you are in failover mode for whatever reason, how does it impact your performance? ie does webmail just crawl until the mirror finishes rebuilding?
- how are your external resources? got a plan to truck in gas for the genny if a tornado levels the local substation? got a hotline or multiple points of expert contact available 24/7 for every critical piece of hardware that you can't fix yourself regardless of how it breaks? same for software.
- do you have a forensics plan in place? ie if you get hacked, (and don't answer that with "that can't happen") do you have any idea what you will do and in what order, to preserve forensic information, stop additional damage, and orderly cleanup? What are your legal obligations for notification, who is your contact with the press? (and there better only be ONE) Do you have a specific partner waiting in the wing all picked out if needed? after the fact is not the time to be choosing one.
- if you have a failure that affects multiple services or clients, what is your priority order? who gets their service back first?
- do you have a set "fire schedule" that people know specific additional hours they will be required to work in the event of an emergency situation? Are you going to run short on manpower in a specific area because you're already overextended by day to day operations in some aspect?
- are there any people that are single points of failure? What if Bob gets hit by a bus? what if Dave is the only one that knows the firewall and gets hospitalized when it explodes while he's working on it? crosstrain crosstrain crosstrain.
- not sure if it was covered above but documentation, documentation, and more documentation. How consistent is it? Does every network map look like it was written with a different drafting app by a different person? Is all of your documentation collected together and well organized? multiple copies in various places? are some things much better documented than others?
- server rebuild lists. do you have a step by step set of instructions for EVERY critical box that will take it from a freshly formatted HD to back in production, that any of a dozen of your monkeys can follow, with no "well wasn't that obvious?" missing steps? And how often do you test these? Walk in one morning and drop a new box on a desk and say "WEB15 just got STOLEN. Rebuild it. Fast. Starting NOW." and hit your stopwatch and see what you get. You do this from time to time, right?
- do you have a structured command that avoids differing opinions in a crisis slowing things down? when it comes right down to it there needs to be one clear person or command structure that has final say in a crisis.
I'm sure I'm missing some things but that's a good start for ya.
I work for the Department of Redundancy Department.
Many good comments, but nobody is asking what PUE a datacenter gets. Bad PUE turns into lower rack deliverable power and more expensive power when you do get it. I would have a hard time picking a datacenter that didn't have tight closed loop hot isle cooling.
The quality of a datacenter has less to do with the equipment (although thats important), and more to do with who designed and is running the equipment.
Most of the datacenter outages I have been a part of in one way or another (Customer, or Provider) have been caused by:
Poor planning
Human Error
Poor design
As a normal customer, there is no way to know if any of these problems exist. The solution? Ask for references that utilize that datacenter. Make sure they don't give you a customer that utilizes another data center from the same provider. Data center design varies greatly, even across the same provider. Ask that reference how long they have been there, how many problems they have had, and the companies response to those issues. Look for a customer with a long history in that data center (3+ years, 5 would be better).
Don't rule out a data center because they had an outage. Outages will happen, no matter how redundant their systems are. Their response to it is very important. If you find out about a previous outage, ask to see the root cause analysis they provided their customers. If they can't or won't produce it, even under NDA, then walk away.
Nuke it from orbit and then see how soon their backup site with your backup data has you back online.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
I prefer calling them Congress-Hertz
Some thoughts, having been a shift supervision in a DC long ago...
(a) don't have the emergency poweroff & off halon dump switch uncovered right by the DC room's light switch. Someday someone will *hit* it...
(b) where does the water go when the 5 inch chill water main fracture?
(c) who is the cleaning staff and do they know not to turn the "computers" off when they dust? Also why are they dusting unsupervised
in the data center anyway?
(d) Oh sure you have multiple telco feeds that service the building from that *sole* pole over there, which that dump truck just took out.
(e) And your procedure for the phone in fire alarm is what, wait for the fire dept to axe the entry door? This was my favourite. What's that chopping noise?
(f) Lastly, if the city bomb squad runs a trial, let the staff know ahead of time because that ensures they might be able to focus after the squad leaves...
Some additional things I looked for recently when evaluating a colo were:
- Ease of access to the building :) ) ... this is subjective but important. talk to the NOC guys, are they nice, or do they hate their jobs, how long have they been working there, etc etc.
- logistics including:
-- how is the parking situation? 24/7 parking?
-- if i need to drop something off from my car, is it close to a loading dock?
-- how are deliveries handled, where are they stored until i come in to use new said equipment
- how many years remaining on the lease (assuming the building is leased)
- when was the last time they checked their backup systems (power, cooling)
- do they have multiple paths of entry for fiber into the building? (or will 1 backhaul bring down the interwebs?)
- do they have crash carts? (monitor / key boards)
- do they have spare cables or power tools I can borrow?
- how sturdy are the racks bolted into the ground? (push them, see if they are telling the truth
- power density per rack that is available
- cross connects to various providers
- do i "like" the place
just a few things I evaluated when choosing a colo.
Unless yours servers absolutely must be local, one of the most important factors should be local climate and environmental risk. I've worked in a couple datacenters in Michigan and it's really ideal:
* No state-wide forest fires
* No flooding if you're above the flood plain
* No hurricanes
* Very few tornadoes
On top of that, if the AC units should spontaneously fail all at once, 99% of the time you can just open up all the doors and run a couple of large fans to keep things cool enough to run.
All line up and evacuate slowly.
How dont you evaluate a data center
You're focusing too much on the data aspects. How would you measure how central it is?
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Newer datacenters don't have raised floors because it is more energy efficient to have concrete floors.
Hogwash.
Why? You can't put as much weight on a raise floor as you can on plain concrete.
Further, if you're doing cold- or hot-aisle containment, you can do it without the need to do a raised floor. There are plenty of in-row cooling options so that you can put the cooling in the places you need it (either beside the racks, or on top):
http://www.42u.com/cooling/in-row-cooling/in-row-cooling.htm
There's even in-rack cooling:
http://www.42u.com/cooling/in-rack-cooling/in-rack-cooling.htm
Or, pipe the water (or refrigerant) straight to the aisle and put it in door-based cooling system which is rated to 35 kW per rack:
http://www.sun.com/servers/cooling/
The above systems also don't have fans, so you don't have to worry about maintenance on it--just the overall circulation system.
Not quite sure what the GP mean by the efficiency of concrete floors, but there are certainly better systems than general circulation air via raised floors.
Why re-invent the wheel?
Yes it's People - Process - Technology... there are audit / checklist standards (multiple) for each area.
Search for SysAdmin SA-BOK-0500.pdf and you'll find a good overall checklist. SANS Institute also has a physical security checklist as well.
Beyond that you have the super serious audit things like SAS-70 or ISO-17799 / ISO-27000 or ISO-20000 or COBIT or ITIL... things that actually have outside audit standards / agencies.
If you want the informal check lists for your own review, SA-BOK-0500 or SANS Institute is good. If you really want to do a proper / formal thing, inquire as to their SAS-70 and ISO-20000 (ITIL) compliance. Ask to see copies of latest audit under NDA.
Full Disclosure: I work for a (Great!) data center provider (ViaWest ).
Infrastructure:
- What is the UPS run-time?
- What is the generator startup time?
- What is the genset capacity in relation to UPS demand? (i.e. is the UPS demand larger than the genset capacity - you'd be supprised!)
- Does the provider have multiple refueling contracts?
- Are the refueling contracts high priority?
- Can the provider detail out green initiatives to improve PUE?
- Does the provider have sufficient capital resources to expand the data center?
- How much investment has the company made - this year - into the data center?
- Is the data center in a flood plain? Check http://msc.fema.gov/
Compliance:
- Is the data center SAS-70 type II audited? Type II means they're serious about it.
- Are the results of the audit available for review?
- Are a list of control objectives available?
- How does the provider assist with customer audits? (i.e. PCI auditor requests for info)
- Can the provider demonstrate servicing other companies where compliance is a requirement?
- Will there be additional charges for audit related work or requests?
Network Remote Hands
- Does the provider offer managed hosting / hybrid hosting options
- What is the expertise level of the NOC staff?
- How are remote hands charged?
- What is the response time for a remote hands event?
- What monitoring options are available?
Corporate
- Does the company have a business continuity plan documented?
- Are the company financials available for review?
Depending on your uptime needs and size, also consider weather you want your Internet Access included as part of your Data Center or whether you want a carrier neutral facility. Many places just lump data access in with the colocation space and you get an ip.
Other places, sometimes called a "hotel operator" simply rent you space and power, after which you can connect to one of usually a couple hundred ISPs that are cross-connecting in their meet-me room.
Also, don't know if you are starting or moving. If you are just starting, be sure you look very closely into simply renting some cloud space. I admit I have been skeptical of it for a long time, but I am now a convert. Sure it is more expensive than buying your own servers and hosting them, but redundancy and capacity planning are almost eliminated.
Just to get it out of the way, yes, IAADRS (I am a Disaster Recovery Specialist - the "speaks bit and byte" and "cosfi" datacenter visiting type...)
Concrete ? Well, yes. Under the raised floor. What did you want ? Marble ?
No cooling from the raised floor ? why not ?
Overhead network cables and a "newish" cooling solution like here :
http://www.datacenterknowledge.com/
front page. Think...giant heatsink...(first overlord joke gets the boot 8p)
As for concrete, let me introduce you to this wonderful answer : it depends. Mostly on the concrete.
"Ramazan Demirboa
Civil Engineering Department, Engineering Faculty, Atatürk University, 25240 Erzurum, Turkey
Abstract
In this study, the effect of silica fume (SF), class C fly ash (FA), blast furnace slag (BFS), SF+FA, SF+BFS, and FA+BFS on the thermal conductivity (TC) and compressive strength of concrete were investigated. Density decreased with the replacement of mineral admixtures at all levels of replacements. The maximum TC of 1.233 W/mK was observed with the samples containing plain cement. It decreased with the increase of SF, FA, BFS, SF+FA, SF+BFS, and FA+BFS. The maximum reduction was, 23%, observed at 30% FA. Compressive strength decreased with 3-day curing period for all mineral admixtures and at all levels of replacements. However, with increasing of curing period reductions decreased and for 7.5% SF, 15% SF, 15% BFS, 7.5% SF+7.5% FA, 7.5% SF+7.5% BFS replacement levels compressive strength increased at 28 days, 7- and 28-days, 120 days, 28- and 120 days, 28 days curing periods, respectively. Maximum compressive strength was observed at 15% BFS replacement at curing period of 120 days."
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V23-4KPFKFY-1&_user=10&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_searchStrId=1085368183&_rerunOrigin=google&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=b8e7b8ce7b5e23b07db5805bf9ad9740
It takes 40+ muscles to frown, but only four to extend your arm and bitchslap the motherfucker
* If you move your infrastructure in with them and later lose confidence in their abilities for whatever reason, how quickly and easily can you terminate and move out w/o all the hassle of pre-paying to end of term? This should be your biggest requirement. No matter how much do diligence you do on technical matters prior to moving in, things change...the datacenter provider could have financial problems, sudden facility problems, lose key critical staff, etc. Your contract needs to allow you the option to terminate when you reasonably lose confidence in their abilities w/o a billion hoops or outlandish costs.
* Power pricing and limitations on how much they are allowed to pass on additional costs from utilities during term of contract.
Seriously. Use AWS for your custom apps. Outsource your email and other G&A aspects of running your company. Data centers are for dinosaurs. All of the cool kids are in the cloud.
My only qualification for a datacenter: Will it blend?
See if you can eat for 2 days out of the vending machines!
A successful data center has no single points of failure. Make sure everything is triple redundant with a by without dropping the load.
Libraries of Congress per second.
Maybe not this figure, but it is definitely good to get statistics on load, figure out what all the systems are doing, etc.
In the datacenter that I work in, there is a lot of deadweight, and even systems that are powered on, and doing nothing!
Other systems are being hammered, and several of our tasks could *probably* have their load redistributed across some of the dead weight systems.
Forget the presentations and the tours with all the impressive blinking lights tapering off to the horizon. It's all about what happens when things go wrong! We've had Alex answer the phone on a weekend and refuse to do anything with a windows server because he only likes Linux, wait for the Windows guys on Mondays. No help, no escalation. End of contract there. Failed mirrored disk? - don't be surprised if they replace the good one. Be aware that the fancy network status websites only include a sample of the actual errors, power outages and faults. Rebuild a simple server? three shifts to do it and the last guy left without leaving the root password. Ignore the data center review sites. They insist you provide a URL that they can track back to the hosting company's network. A brilliant way to ensure happy reviews only because no current customer is going to let the world know their servers are at a horrible hosting site. Make sure the escalation process works. Use fire drills on them to figure out if it really works. Another great question to ask - is their phone system VOIP? If they get a DDOS attack then you can't call them because it takes out their phone system too. The above mentioned analysis of power and cooling is wonderful but if the NOC manager decides not to pay overtime then maintenance gets done in business hours. These are not bottom-feeder companies, these are mid-range or higher price ranges. Either you have to deal with these problems and set the right expectation in their minds, or be prepared to have plenty of backups and be able to move to another company when required.
All the technical points folks are making here are very important.
But the most important thing is the people managing the datacenter.
At least in Paris, about 4/5 of the catastrophic failures experienced in the last several years have been due to:
- the management being a bunch of slimy cheapos and not doing maintenance on time or cutting corners when they do get around to doing maintenance
- some cro-magnon "technician" from a maintenance contracting company doing something stupid because he was completely unsupervised
In all cases, these datacenters had everything that folks above have described: dual theoretically diverse utility power feeds, dual generators with big fuel tanks, big battery systems, dual theoretically diverse chiller circuits, etc etc etc.
The only thing you can do to protect yourself against this sort of thing is to treat the datacenter selection process, especially the salesbeasts, as a job interview. If you say something like "I don't care how big your generators are; show me proof that they've had an oil change sometime in the last two years, that you test them regulary, and that your emergency fuel delivery contract is paid up" and they bullshit you, it's time to look elsewhere.
The single most important thing for me is to find out what procedures they follow when a 3rd party contractor is on-site doing maintenance on their critical equipment, especially the power transfer systems. Power control master switches seem to have some sort of special attraction for morons. Outages experienced recently (all the fault of the unsupervised 3rd party maintenance technician):
- removing both utility feeds from the master control switch along with both of its own internal battery backups at the same time, so that it defaults into the fail-safe mode of "off"
- needing to transfer a PDU feed from one source to another and being alone, so he shuts off the primary feed before he walks over to the backup to enable it
- doing some kind of "test" of the redundancy settings and screwing things up enough that the datacenter power is running off house batteries, the generators do not kick on as they detect that the utility power is working, and the batteries are disconnected from utility power. 15 minutes later, all the batteries are flat and the datafloor power is dead. The house lights still work as they run off "dirty" power direct from the utility, and the cleaning people are running the vacuum.
- during a cooling system purge, leaving the drain valve for the cooling system open, with the fill valve for the reserve water tank shut, and the reservoir level alarm disabled, broken, or ignored. It took almost 24 hours to get the datafloor temp back to normal as the entire cooling system circuit was dry.
- some construction jackass lucky to still be alive drilling directly into the master B-feed power riser cables and getting God knows how many amps directly into his concrete drill. Every single individual breaker in all the B-feed PDUs on every floor popped. The worst bit was that the jackasses that run the place didn't have a master record of the breakers in each PDU (the data was just kept in the individual client records), so they started digging through all the client records one by one (also note the lack of someone that knows about SELECT FROM) to figure out who to turn back on until about 10 people ran into the DC manager's office screaming at them to turn everything on and sort out which ones they should turn back off later.
- a maintenance by a utility power technician causing the datacenter power system and the utility power system to have a somewhat different idea of what constitutes neutral voltage on a ground, again leading to the generator system thinking that the utility power was just fine but the battery system detecting a ground fault and refusing to us
Good thing to check is what is the average time after submission until a reboot is completed. Along with this do they have a SLA for reboots and what the price is for the reboot.
Also if you might need remote hands(if you are not within driving distance to the DC) how long do they take to attach the equipment and at what cost for what length.
After u are done with all these rigorous questions
print them out and ball them up and throw them away.
Because either you (and/or the money people) will decide based on completely unrelated criteria or the data centers won't give you half of the info you requested (maintainance records? 9/11, total power in vs current used? 9/11 and trade secret, etc) or if by some miracle they do give you the info then it means basically jack.
Take the pompous asses at 365 Main for example, if you go by all this crap that the other people here are telling you then they would be near the top of the list - N+fucking 1, etc but guess what? They had a huge outage due to, as they tell the story "commands hiding like scared bunnies in the command queue" of their uber N+1 flywheel generator systems who came scurrying out when they went to fire up the generators. (Yeah, right - we bought a $125 million colo for $2.6 million, waaaahoooo!! and the vendors and contractors were left holding the bag for the difference - but we charge you like we spend $125 million - suckers)
Meanwhile, you have a colo nearby (now web.com, previously Verio) that had not had an outage in 4 or 5 years but didn't have all the buzzwords.
Based on a recent video on slashdot, the datacenter needs to be located other than in a flood plain.
I'm an individual! Just like everyone else!
Besides the obvious backup power, cooling, environmental stuff mentioned above....
1. Expandability.. how much rack/cage space is available nearby? Get a right of refusal on any empty space near your stuff if you can.
2. Power max per rack, can you get enough for SAN's, blades etc
3. Remote hands availability/skills/costs (really want to make a trip to the datacenter to replace dead hard drives? No. Do the employees know enough to do *limited* work for you)
4. 24/7 access, near employees if a physical presence is required
5. What sort of racks do they have? Can you buy your own?
6. Storage space. Need 100 servers shipped but don't have racks yet? Need to keep spare hardware on hand? Do they have room keep that for you?
7. How many/what carriers do they offer? How is access delivered? What does their network look like?
8. Do they have spare tools/network cables/misc parts. Can they order stuff for you, or is there some place nearby you can pick things up if needed in a hurry?
9. How many employees can you get access for?
10. Do they have a crash cart? Comfortable place to work? Wifi or other forms of internet access available?
I'm looking for a datacenter. You got any of those for sale?
Dick.
if utility power goes out, batteries and Gen kick in.... but if the center is on gen power as part of peak load sharing with the utility... and the fuel runs out, can it auto kick over to the grid? Happened at an internal telco data net ops site... took two weeks to get everything right... :-)
I'd guess 90% of projects fail at step #1: Define your needs. What's the objective here? Why are we doing this, and what are the benchmarks required for success. Does this sound familiar?
First, define your needs, then evaluate possible solutions to what might meets your needs.
If you don't know what you need, you don't know what the hell you are doing. Hire someone who does, like a consultant.
If you have commercial information that you absolutely cannot allow to fall into the wrong hands (or accidentally deleted, corrupted, not backed up, whatever), is storing that data in a data center ever really acceptable? I would think not, but I'd like to hear someone else's opinion. Has anyone here done things DIY for this very reason?
If I have seen further it is by stealing the Intellectual Property of giants.
I wrote an extensive article on choosing a datacenter/colocation facility several months back. The full post can be found on my blog, but I will paste it below for your Slashdot reading convenience:
http://www.bitplumber.net/2009/04/how-to-choose-a-colocation-facility/
How to choose a colocation facility
Choosing a colocation facility is one of the most important decisions an IT professional can make. It will have repercussions for years down the road, as there is generally a contract term associated, and it becomes difficult/costly to move. At the same time, unless you are a facilities professional, it is hard to tell the difference between the quality of one facility vs. that of another without knowing the right questions to ask. I have developed this list in the hopes that it will be a reference to folks evaluating datacenter options. This has been written using the assumption that you need a local datacenter rather than a DR facility (which can have very different needs), however, many of the same concepts will apply.
Location
When it comes right down to it, there are still certain things you have to do physically in person. You can’t run a network cable through SSH or RDP. Having a datacenter close by makes a huge difference, especially when you lose remote connectivity and must go push a button in an emergency (we all have done this once or twice). In general, the newer, more high-end, and redundant your equipment is, the less you should have to touch it in person. Things are getting much better with out of band remote access controllers, but sometimes being there is worth a lot. You can’t hear that fan making funny noises from your office.
Does the facility have good access to transportation such as freeways and airports? Are their hotels nearby if you will have out-of-town contractors visiting? How close to logistics depots are you for your vendor-of-choices parts, i.e. Cisco, Dell, HP, etc
Does the facility have adequate parking that is close to the building, does it cost money? Is it somewhere you want to leave your car in the middle of the night while you are inside working?
Do you have line-of-sight to the datacenter? If you can manage to get a wireless link to your datacenter this can be an extremely cost-effective option for high speed connectivity. There is something to be said for controlling your own destiny when it comes to your connectivity rather than being at the mercy of a telecom provider. Will the facility allow you to put a wireless antenna on the roof and how much will they charge?
Staffing
Do they have on-site staff 24×7 to respond to emergency situations, to secure the facility, and to provide access when you forget/loose your badge (or have to stop by on your way home from the gym).
If they do not have staff on site 24×7, what is their on-call policy? How long would it take them to respond to a power failure, a UPS exploding, a transformer catching fire in the parking lot, an Internet outage, an FM-200 fire suppression system going off, an HVAC system failing, or any other major malady (yes I have had all of these things happen to me in facilities I have worked in, and I am still waiting for the day a fire sprinkler goes off or there is a real fire in a datacenter).
What level of professional services can they provide? Basic remote hands (please press the power button)? More advanced troubleshooting (help diagnose a failed network switch)? Or even managed services (i.e. they take care of backups).
How competent are their NOC engineers, facilities folks, etc What quality of vendors do they use to do electrical work, HVAC maintenance, network cabling? This can be hard to tell, but there are lots of small clues you can pick up on.
Does their staff speak English fluently and without heavy accent? It is extremely difficult to communicate on the phone with someone in a loud datacenter environment about complex technical issues when both of you are having a hard time understanding each other. This dramatically slo
if they care enough for you to have an extraordinary experience when you have to use the restroom, that means they really care.
Is there a good desk working area? Is there a landline/PBX for you to make calls from? Is there decent mobile phone reception in the work area and by your cabinet? Can you eat food or bring drinks into the work area or around your cabinet? Is it in a shady neighborhood, where you might feel a little intimidated bringing in tens of thousands of dollars of emergency IT equipment @ 3 AM? In the event that your credentials aren't working (i.e. hand scanner, ID card swipe), can they let you in remotely, or is it manned 24/7? Is it carrier neutral and are there other backbone providers that you can connect with? Do they charge for running cables between cabinets, especially in cases where the cabinets are not adjacent? What is the max amperage that they'll provide per cabinet? Do the rack cabinet doors remove easily? Are there chairs available, and damn it, are they comfortable?
Your question is a little ambiguous. Are you looking to buy a data center of your own or are you renting rackspace?
If you are buying the Data Center
1.) Normal title , lien, Structural due diligence as for any RE purchase
2.) Is it on a flood plain
3.) Seismically active site?
4.) Serviced by multiple communication providers from multiple CO's
5.) Power available from two different substations.
6.) Physical security / susceptibility to civil unrest
7.) Physical access driveways, parking, loading docks, hallway widths elevators ramps
8.) Floor / raised floor design loads. I have seen more than one raised floor rippled by rolling overweight gear on it.
9.) On site power generation / fuel storage. Mech. condition, age, availability, reliability, repair-ability
10.) Sufficient Chiller Capacity
11.) Sufficient UPS / Power Conditioning
12.) Sufficient space both for current needs and growth for planned lifetime
13.) Sufficient office / command center space
Those should be adequate to get you started.
For rented rackspace
I would say you at least need to glance at items 2 through 11 above. Beyond that
1.) Per rack power limits
2.) Physical security
3.) If you are using "hands on" services it's skill set and response time.
4.) Whatever value add services you will be using.
Sorry it is late and a long day and this is all I can think of.
If you need a data center for your business, and you are into one of those Ponzi-like schemes, the first thing to look at is the speed of low-level disk formatting, just in case...
White Mountain datacenter in downtown Stockholm, Sweden. It is located in a bunker 30 meters under solid bedrock. It was a cold war bunker that was converted into this datacenter and is said to be able to withstand a Hydrogen bomb blast. http://www.youtube.com/watch?v=qwlATf9xse4
All the normal stuff is important, but it is the things they won't tell you that is really important.
I visited a data center that was really easy to commute to. See, it was above a subway train. Further, it was below a parking garage. Nice.
I've seen data centers at the end of airplane runways. I've seen data centers in flood zones, hurricane areas, earthquake prone locations and with roof leaks. Nobody is going to point these things out to you.
I really like how one DC manager showed off the huge springs that the turbine generators were sitting on to prevent vibration translating to other parts of the building, but they were behind on their battery maintenance schedule. Doing the simple things, well and consistently are key.
In the end, you can't trust anyone and the safest answer is to spend your money in 2-3 locations. Don't make 1 a "primary." They all need to be "primary." Mix production, DR, Dev and Test apps across all locations, split them evenly, and keep them 500mi or more apart. Don't have a single supplier for the DCs either. Consistency is good, unless it isn't. Being different in each location and having different management is a hassle, but both probably teams won't make the same critical mistake that takes your systems down.
Don't hire, Wesley Carver! http://www.amazon.com/review/R3VO962Z99T6KZ
What kind of contract must you sign? Is it in legalese or in understandable English? Will the data center only be liable for broken hardware in case of a failure or will they also compensate for your loss of revenue and blemished reputation?
1. water based fire suppression (sprinkler heads over server racks)
2. security fences that only go up to a false/lowered ceiling
3. metallic dust on surfaces (from cutting metal in DC)
4. external/temporary/emergency cooling ducts on DC floor blowing cool air from AC unit parked in loading dock
5. generator running for 2 weeks
6. tropical feeling to air in DC
7. 45 minute wait to get access to customer colo area due to no onsite staff
8. power conduit under raised floor obstructing airflow
9. fire suppression override always set to off
10. mis-configured power ATS
11. NOC staff that are active crackers / black hat
12. hill billies doing rack installs / construction
13. windows server datacenter edition
(1) any history of letting police walk in and wreck stuff
(2) the décor
I am sorry, but all these people going on and on and on about what you want in a data center are missing the unforeseeable. And the only way to do that is redundancy. What you want is two or more different data centers, in two or more distinct regions (ideally of the World).
Ask yourself, what would you want running if the entire city or region was nuked, had an earthquake, was hit by a Tsunami, or an asteroid dropped on it? How about if an airplane flies in to it?
What happens if the regional power grid or network is out for days, weeks, or months? Ask all the guys recently in Asia or the middle east about underwater data cables being cut, and what they wish they had. All the above have happened, and will likely happen again. Do you really want all your eggs in one basket?
The only way to be sure is to diversify your data over as large a geographical area as possible.
I will still take larger numbers of lower quality but diversified data centers in numbers, over a single high quality data center any day. Of course, having both helps.
By the way I keep my data on four different machines, located in four different physical locations in two countries, on two continents. I figure any disaster that will take all four down, will be one sufficiently big that I just don't give a dam anymore.
Living in Chile
Data center draws a lot of power. One should consider whether there are Green technology implemented to reduce operating expense. Also spelling out stringent resiliency requirement is good, but one must be pragmatic when doing the specification. What are we trying to protect, what is the business impact, how much downtime can the business tolerate, and what level of risk can the business take. When carefully thought through, the added resiliency might not be necessary.
This summer Equinix Paris had a major failure of their cooling system. Of course, they had a backup, but as was expected, the backup was identical to the primary system, and therefore failed identically. Temperature raised over 55C AFAICT. We didn't experience hardware failure since all our servers shut down automagically at 45C. We also had all our systems clustered over a gigabit MAN to another DC, so we suffered only a minor outage.
Shit happens. You always have to keep that in mind. But two things could have made the whole situation better.
- They didn't alert us of the failure. We could have avoided the outage completely had we been called when the cooling system failed, instead of finding out an hour later. We could have supervised the switch over, and there would have been no loss of service at all.
- Datacenters should not get that hot, ever. They should be built so that they don't get much hotter than outside temp; it's not that hard, thanks to a high technology device called a fucking WINDOW. Too hot? AC down? You open the god damn windows! I believe the latests DCs are built that way, with passive cooling doing most of the job.
You don't need high end AC with temperature controlled down to the degree. It's a complete waste of money. Instead you should have reliable AC, with graceful failure modes.
Just wanted to comment that the data center I hosted at for 7 years (Chicago Equinix via Internap) had all the right stuff. But when the power to the facility went out when a fire broke out at local power station, one of the UPS generators failed when the wires shorted out. The other generator (it was an N+1 facility) couldn't handle the load and the whole facility was down. Good thing for us it was on a Friday night which was a slow time for our ECommerce site (ArtSelect.com, now shutdown by art.com who bought us).
CTO using Perl and MySQL (and proud of it).
If "pro" is the opposite of "con", what's the opposite of progress?
Anyway, I lol'd at the Libraries of Congress per second thing... shame the downmodder didn't "get" it.
I live in the DC area and drive by the National Archives every once in a while, and I still hate it when it's used as an analogy.
Read Urs Hölzle and Luiz Barroso's (both Google) book on datacenters:
http://books.google.com/books?id=Y3hhzdSSK58C&dq=h%C3%B6lzle+barroso+warehouse&printsec=frontcover&source=bl&ots=3GAME9c91o&sig=SFKN5MEnDyxy_Zk_bOctH776cjU&hl=de&ei=xaz5Su-zB8fm-Qbur_ngBw&sa=X&oi=book_result&ct=result&resnum=1&ved=0CAsQ6AEwAA#v=onepage&q=h%C3%B6lzle%20barroso%20warehouse&f=false
Also available as PDF somewhere.
i did not read all the comments but i had done cage wrestling with servers as a week long event as well as those mid-night cravings that i need to touch the servers just because they are not purring at the right frequency. in addition to all the power, connections, server survival conditions, server security, assisted monitoring (don't rely solely on anyone), etc etc. i also find the need for good nearby parking, large clean bathrooms, available stools or chair, tool set(sockets, big screw drivers, pilers), a place for a coffee break, scream at the VAR and making excuses to the spouse, ULTRA HELPFUL. for small setups and fuzzy architects, these are something as a tech look for in my experience.
Comment removed based on user account deletion
While everybody else is suggesting technical things to look at, I'm going to suggest that you look at the legal things instead.
Read the contract with a magnifying glass. In fact, have a professional with experience look at it for you. The contracts often have traps in them that it takes a trained professional to spot.
For example: one managed hosting provider had it in their contract that they would replace any dead servers within 2 hours of determining what the problem is. Catch is, all they had to do was say "we haven't determined what the problem is yet". Kept our main server off line for nearly a month while the night crew and the day crew argued over what needed to be done.
Contract said that if you so much as upgraded a disk drive, that re-upped the contract for eighteen months. Think about that for a bit. We couldn't afford to extend our contract for another 18 months and we couldn't do a thing about our overloaded server situation. You also had to give 60 days notice before the end of the contract or it was automatically renewed.
Also: never go with just one provider. Diversify. If all your data is in one data center, they can hold it hostage. Happened to my dad in the mainframe days, happened to me last year. Some things never change.
I could go on for hours, but I need to watch my blood pressure. Or my cholesterol. Something like that anyway.
In down-time per megaton-kilometer.
Check out http://www.usdatacenterlist.com/ - from there you can search and compare data centers on their technical and operational processes to ensure they are a good fit before you engage sales staff!!
There are many good points above about infrastructure and convenience, all should be considered when choosing a data center.
However, there is another point which should be considered; how well is the company run from an operation/core values standpoint. You can have the coolest features in the industry, but if the culture of the company doesn't invoke 100% accountability throughout the entire organization it doesn't mean much. Make sure that whatever company you choose has the correct value from all aspects of the business, not just data center functionality.
http://www.onlinetech.com/company/core_values/
Geographic location is also a necessary component of your choice. The Midwest is typically the best choice for a number of reasons. Here are a few:
http://www.onlinetech.com/company/company_overview/
Jason Yaeger
Online Tech, Operations Manager
John Savageau, who has been an executive with a number of hosting and colo companies, recently blogged about Questions Data Center Operators Don't Want You to Ask, which some colo shoppers may find useful. It looks at issues for colo centers in mixed-use buildings and the merits of SAS70 certifications, which are often a key marketing point for facility operators.
My auditor is really nice, but some things in a data center really bug them. And, that makes my job difficult.
Can you tell me if my auditor will like visiting you? My auditor doesn't like:
- external door hinges they can pop the pins out and get directly into the "secure" server rooms from the parking lot. Especially the ones on the roof and fire exits.
- outsourcing without asking if I'm OK with that, and ensuring we all know how we will manage any added risks.
- security personnel vendors that can't provide the last date they verified all their temporary contractors and terminated employees had their access shut off.
- same for data center personnel; and saying it is not the data center's responsibility to oversee the data center's vendors.
- propped doors for smoke breaks with no alarms.
- battery-powered wireless security features that don't tell you when the battery is dead.
- preventive maintenance personnel with 24x7 access cards.
- if I get all I need via email to verify, but that our auditor will have to provide 2 weeks notice of all questions, be accompanied by me, make no audit records (i.e., photocopy, pen
and paper, laptop typing, etc.), bring no equipment, not write anything down, not ask any questions involving HR or outsourced personnel, etc.
- the ability to plug in a USB drive or USB modem in a server without detection
- not allowing me, my auditors, or an independent auditor data center information for "security reasons" to validate our GLBA, HIPAA, FISMA, etc. compliance
- not having an online data center ticket request and clearing system that I can watch for progress and compliance status.
I assume you know the rest like uptime, capacity, insurance, etc.
I work for FORTRUST, a data center services provider in Denver, CO. FORTRUST published a comprehensive white paper and workbook as a data center site selection resource. The white paper, Evaluating Data Center High-Availability Service Delivery, examines several key criteria such as operational process controls, service assurance, maintenance and lifecycle strategies, critical infrastructure management, and capacity planning. Download the free white paper and workbook at http://tiny.cc/yqn4C.