Slashdot Mirror
Low Redundancy Data Centers? Providers Adapt As Tenants Seek Options (datacenterfrontier.com)
1sockchuck writes: Data center providers are offering space with less power infrastructure than traditional mission-critical facilities, citing demand from customers looking to forego extra UPS and generators in return for more affordable pricing. The demand for "variable resiliency" space reflects a growing emphasis on controlling data center costs, along with a focus on application-level requirements like HPC and bitcoin mining. Data center experts differed on whether this trend toward flexible design was a niche, or a long-term trend. "In the next 12 months,data center operators will be challenged to deliver power to support both an HPC environment as well as traditional storage all under one roof," said Tate Cantrell, CTO at Iceland's Verne Global. "HPC will continue the trend to low resiliency options." But some requirements don't change. "Even when they say they're OK with lower reliability, they still want uptime," noted one executive.
Now that variable resiliency data centers are finally available, I can run my sometimes available services in the partially secure cloud space I am building.
Great, now they just have to pick one of those.
Problem solved.
Why? Failover keeps high uptime even if you have less reliable hardware.
I'm reading this as.. "Well, we need to have redundancy, and we're already ponying up this much money, but how can we spend less and still say we're "redundant?" I'm not faulting the datacenters for offering such a service, but the customers should really have a hard look in the mirror.
Bitcoin mining is still an in-demand data center application?
Makes you wonder ... just how much fossil fuels will have been burned by the time we've "created" all of the Bitcoin there is to make?
Breakfast served all day!
is ahead of the curve. They adopted that policy years ago. We've had three power outages so far this year.
As with a lot of other services (in fact, all other services that I can think of) that reach a certain level of maturity and ubiquity in the market place, one of two things seem to happen. Large-scale consolidation reducing the number of competitors until a small number of actors, or a single monolithic entity, remain; or reduced perceived value of aspects of the service leading to a bare-bones offering because customers decide they are less willing to pay for services they are not going to use very often (but when they need them and they are not there... oh boy, will they scream and harass the provider who told them they were sacrificing redundancy and potential uptime for lower costs).
Because a few generators, transfer switches and UPSes are such a large portion of data center costs ... Its not like they are outweighed by just a single months power bill or anything in most cases.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
Maybe it's a new business venture from the same guys who run Jiffy Express?
why would you want to pay for 99.99% uptime when it's rarely provided?
Anons need not reply. Questions end with a question mark.
Why? Failover keeps high uptime even if you have less reliable hardware.
Unless the unreliable hardware is your power source.
"A plan fiendishly clever in its intricacies"- Homer Simpson
Depends on the service. Many things can easily be distributed over a few datacenters. 10 cheap servers with so-so uptime may be better than one with a very high uptime from a cost point of view.
Some people still run Windows on -servers-. This proves that reliability isn't always the most important thing.
The cost is two to four times higher. The basic idea is you can choose:
___Non-redundant A-only power____
1 router (1 amp) + 1 switch (1 amp) = 2 amps.
You need one power plant capable of providing 2 amps.
___Redundant with A/B power___
2 routers (2 amps) + 2 switches (2 amps) = 4 amps
You need two sets of power, each capable of providing the 4 amps, so 8 amps total.
Note you need not twice as much power capacity, but FOUR TIMES as much in order to have full A/B redundancy. Plus the more complex (expensive) design with more transfer switches, etc.
Twice as many PDUs feeding twice as many routers and twice as many switches take up twice as many racks, which means twice as many square feet, and almost twice as much power and cooling cost.
That'll end in tears.
The space provider should provide AC that is guaranteed to be generally up, but may have limited gaps while their gensets come on-line.
If the gaps are predictable with N+1 gensets, then that should provide both cheap and reliable space.
Riding through the gaps is a batteries in my-space issue.
Then power it via the cum dribbles from timothy's micropeen.
Maybe I'm just looking at it wrong, but wouldn't this fit the bill for cloud setups ?
Have X lower cost data centers spread about the regional area, mirror the data over, provide multiple paths to get to it and voila.
Though I'm not sure that X unprotected cloud sites would cost less than 1-2 properly setup redundant sites.
I looked into it a while ago when I had unreliable power and outages lasting more than an hour so well beyond UPS capacity. Once you get beyond just having to fill in for a few minutes the generators get expensive, and then you need a plant mechanic to keep the generators in working order, fuel storage, a test regime heaps of electrical work etc etc. A purpose built large scale data centre can be expected to have such things and absorb the cost but adding it in after the fact with something that is not huge, especially on a site without room to put those generators and fuel storage outside, is not a trivial thing.
Ironically now that the place I work for actually has a plant mechanic, electricians and the space for generators plus storage the power supply is stable enough that I no longer need them.
I comes down to the cost of downtime versus all that messing about - in my case a few days downtime of the site can be coped with and absorbed into moving schedules. There really are not that many businesses that are 24/7/365 no matter how important they think they are. A few days out due to flood or hurricane/cyclone will confirm that for most people.
We spend big bucks at a tier 1 data center to host our data. Our application cluster is fully redundant, and we serve large amounts of high-value data. (think: business intelligence and workflow information, not video streams)
We have a DR cluster, a back-stop for when all else has failed. Although it is a redundant copy of our production cluster, it itself is, by design, non-redundant. Where our production cluster has at least two physical machines (and often more) providing any specific service, the DR cluster has just one.
It's there as insurance against the rare, once-in-a-lifetime event that has been known to shutter companies. I've experienced one such event when the data center I was hosting at went dark after they went insolvent - it took weeks to get all services resolved, because, while I had backups, they were onsite. (doh!)
While I've never since had backups that weren't automatic, daily, and off-site, I see no reason to demand the same power redundancy for the DR cluster that I do at the primary data center. Cost is more of a factor.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
Unless the unreliable hardware is your power source.
That's why you put your failovers in different data centers. If you spread the load over 5 different data centers, with 20% extra capacity, losing one isn't a big deal.
Especially if you go with some of the more modern plans that let you scale within moments.
I don't read AC A human right
This seems like the kind of thing that would benefit some SMBs I've worked with.
I've worked with several that have awkwardly architected homegrown applications or very low-quality vertical market applications that they're highly dependent on. The net result is applications which don't translate well to cloud-hosted scenarios for various reasons, or at least not at cost levels that make any sense.
Colocation would work, but datacenters' relentless focus on being super duper redundant makes them too expensive, so the clients are stuck hosting their own systems on premise, often in facilities not much better than the proverbial mop closet.
Low cost colocation facilities that might not fit some kind of top-tier datacenter definition may not be a great choice for others, but I can see these kinds of clients going for them. It would be a reliability improvement over what they do now and probably at a cost they would afford.
Unless the unreliable hardware is your power source.
That's why you put your failovers in different data centers. If you spread the load over 5 different data centers, with 20% extra capacity, losing one isn't a big deal.
Especially if you go with some of the more modern plans that let you scale within moments.
But then you're paying for network connectivity between the sites which negates any savings.