207 posts • joined 16 Apr 2011
All-flash arrays need Plan B
This is clearly a major trend, and unstoppable. It doesn't work for all-flash arrays, since the flash modules are proprietary to save cost and space, and to increase performance.
Note that most servers have just a few drive slots. This doesn't work well for today's HDDs, since the idea is to have a lot (60 drives???) of bulk SATA as the secondary storage behind a flash primary tier.
More detail is needed or this whole idea will rebound!
ODMs will enter the fray, too!
Commoditization is already moving at a fast pace, but it will accelerate as the ODMs who service the big CSPs start selling branded product and whiteboxes in the US. The barrier to entry on storage is reaching the Linux level, and we all know what happened to mainframes!
How cheap is that whitebox?
I can imagine the debates at EMC. On the one hand you have guys feeling the hot breath of the ODMs on their necks and worrying the price of hardware will drop like a rock, and on the other you have guys trying to embrace the inevitable, and build software to control it.
It looks like the typical reaction of the status quo. Every salesman in EMC is motivated NOT to sell ViPR, since it's cheaper overall and it loses them most of their account control. Result is dismal sales for the product. ViPR will quietly be buried, and everyone will heave a sigh of relief.
Of course, the ODM issue won't go away, and the ViPR folks will be vindicated, but much good it will do them! Better for EMC that ViPR be transferred to VMWare. Now EMC is without a unified storage approach or a good Software-Defined Storage story....Oh wait!, ViPR isn't dead yet!!!!
Is NetApp talking price not cost?
This seems a bit FUDDY. Internet retail prices of MLC flash drives have already fallen below SAS HDD OEM pricing.I don't know why anyone would use enterprise HDDs.
As to the supply issue. Pricing tells the story. If there were a supply problem, pricing would be going up not down. I agree that flash won't replace bulk disk drives with 5TB capacities any time soon, but enterprise "fast" drives are dinosaurs. There's plenty of flash production capacity for that segment to be satisfied. Note that the replacement rate of flash for enterprise HDD is significantly lower than 1:1 for both drives and total capacity. The SSD are much faster, and don't need short-stroking.
It's nothing to do with rural!
US cities are not much different to say South Korean cities. The rural excuse for lousy broadband doesn't cut it. It's true for Kansas, but not for New York. With most subscribers in cities, the point is even more obvious.
The real driver for the Telcos is getting as much return on their investment...that means no upgrades to the physical plant. Competition is the only force that can drive the issue.
Warrantless search for $500?
In many countries, usage of this tool requires a warrant. If it is being used illegally, will that invalidate convictions?
Britain (and the US ) need smaller cars
There is a partial solution to congestion sitting at hand, and it isn't this crazy electric car thing. Most commuting is done solo, so why not require 1 or 2 person sized vehicles. These would be super-compact, gas efficient, but most importantly, you can park at least 50 percent more cars in the same spaces, and they will effectively reduce traffic by around 30 percent, just by taking up less space on the road. They'd be cheap to manufacture, too.
What is the use case for this?
Apart from giving field engineers a hernia, you have to wonder what this box is for. It can't hold much SSD because there aren't enough interfaces. This means it is a secondary bulk storage system, which begs the question of why SAS redundant interfaces were provided for the drives. SATA would have been the right choice. Front to back depth is going to cause problems, as is the floor load.
This looks a bit like a solution looking for a rapidly vanishing problem. The trend to compact storage boxes with fewer drives, such as FaceBook's 1U 12-drive module, makes much more sense, both use-wise and economically.
Pedalling a bike on the Nurburgring??
You have to
1 Ask why you would need such a fast connection on a snail-paced spinning drive
2 Wonder why SATA isn't enough
3 Ponder the presence of any spinning rust in the primary storage tier
4 Suspect this is yet another ploy to have high-priced (and unnecessary) "enterprise" hard drives
5 Predict that Ethernet drives will replace SATA anyway
Cyber Cold War
There's clearly a cyber battle between the US and China, and it extends to commercial espionage. Why everyone is excited about just network gear escapes me. All of the large US companies, and all the Chinese brands, make their servers, PCs, tablets and phones in China.
They all have firmware and are easier to compromise. The solution is to treat firmware as software, and load it at the final user from a trusted source.
They missed a wave!
It's interesting Gartner is breaking ranks with the march of the Big Battalions. That's very rare for them to do. Still, they didn't go far enough!
We will see a major move towards white-box storage, driven by both low price points and volume shipped to CSPs. Liken this to the evolution into Linux for systems and I think you get the picture.
The result will be a race to the bottom on prices, which generally is good for the consumer, but not so for the larger traditional providers. Software-Defined Storage reflects that fear, as the realization that value will be in the software not the hardware.
But there are only so many software stacks needed and the open-source community is working on that too. The traditional big vendors are going to need good futurists to guide through this!
Common sense at last!
Now that there's a big chill on Anthropogenic Global Warming, it looks like the professional doomsayers are out finding the "Next Big Thing" (NBT). As with all fortune-tellers, predicting a disaster is better than predicting good news. If the disaster happens, the fortuneteller was right, and if it doesn't, everyone is so relieved they forgive the lie.
The NBT is going to be resources. We are raping the Earth and using them up without recycling - right! Soon we'll have calls for recycling every gram of metal into component elements, and plastics recycling will require us sorting every plastic type into unique bags. I can't wait!
Worse than Fukushima radioactivity in Califonia seawater?
What's worse - 6 pisticles per gallon or 5 nanocuries per liter?
This is environmentalism gone overboard. Doesn't Portland know that California is considering recycled sewage water?
It's this sort of crap that gets real environmental science a bad name!
Re: Synchronous replication over distance !
This is a serious and possibly fatal weakness in VSAN strategies. Vendors seem to extoll the virtues of storage mashups without getting into the detail of achieving no single point of failure. That makes VSANs a retrograde step.
This is consistent with the new instance storage mentality, whereby a local SSD is used to boost performance n an instance. That approach abandons virtualization, since loss of a node can cause actual data loss, given the no repair mentality of CSPs. The result is that jobs are in essence running in a hosted services model rather than a true virtualization model.
VSAN advocates blithely assume that "flexibility" is good, but the concept ignores too many real world constraints. Power efficiency and cooling, efficient networking, and ease of management are all understood in traditional storage, while it's still hard to see any thought of this behind all the VSAN and SDN hype.
AGW isn't doing well
Despite the US President saying it's so, AGW's science is falling apart. Even the much vaunted ice retreat in Antarctica looks to be from natural causes.
It's time to move on to the next scare!
Wait, Wait! AGW is COLLING the Antarctic!
So this morning El Reg has an article on how boffins in Australia think AGW is cooling Antarctica. I guess Aussies see the world upside down, or else the Californians have been on a different planet for a while. You can't have it both ways!
Having delivered servers to the big CSPs, I can tell you this is a very tough market. The CSPs are technically very savvy, and buy in huge volumes, so their cost leverage is strong.
The question here is what HP brings to the table. It isn't support, since the CSPs use limited repai philosophies, and can handle technical issues themselves. It isn't volume buying power, since the CSPs are doing just fine going direct to the ODMs. It isn't technology savvy, since the Googles of the world are so far ahead of "mainstream" design philosophies that they would be teaching HP the business.
This play doesn't make great sense as stated...going after the 20 titans. It makes much more sense for enterprise private clouds and boutique second-tier CSPs or SaaS providers who don't have the skills or leverage to go to the ODMs direct.
I hate to be a bit cynical but I wouldn't expect a committee that has worked so long and hard, with such a strong political leaning toward Warming among the staffers who controlled it, to come up with anything short of a Jeremiad.
As is common in these cases, the baseline is chosen to give a decent increase, but the fact theat Earth has failed to follow the model for the last 21 years seems to have eluded the committee members.
Until there is a real explanation available for the last 22 years of stagnation on the warming front, and not just vague comments about ocean stirring, I won't jump on the AGW bandwagon. I find the model-based science rather dubious, especially as the cheerleaders for AGW regularly tout a symptom such as melting ice as absolute proof AGW is here, only to reverse themselves when the ice rushes back a couple of years late.
We do need to stop using oil, because there's a limited amount of it, and I don't want my grandchildren fighting a war to get the last drops. But dubious science is not the way to do this. We need some serious backbone in the debate, some common-sense and an acceptance that we can build safe nuclear reactors - we have thorium fuel for 5 millennia for those and pollution is much much better than uranium power!
Not much boost from Win8
It doesn't look like there's a wave of desktop replacement due to the demise of Windows XP, so either new desktops are going to SSD, or Microsoft has a dud on their hands again.
NVDIMMs can be much faster
There are two types of NVDIMM. One is a flash block-IO device using the DIMM memory bus, and the other is a DRAM memory with a flash backup on power failure.
The latter runs at DRAM speed, and can be directly addressed by register-memory instructions. That leaves FusionIO, and PCIe generally, in the dust, but using the capability needs some software changes.
So the speed argument only applies to the other type of NVDIMM, which is still tied to the storage software stack in the OS and the moving of 4KB blocks of data.
Did the NSA write this bug?
The hole is so elegant and so widespread that you would wonder if NSA wrote it.
If indeed they failed to act, that implies they didn't see this as a threat to any national data, which suggest they knew it wasn't originated overseas or by black hat types. That means they are likely perps!
Those pesky Excel spreadsheets.
Jay's spreadsheet needs some work on it. S3 is indeed 2.5 cents per gigabyte per month, but that is for 3 replicas of the data, so it's really only 0.8 cents per drive gigabyte. At 3 months that nets 2.4 cents per drive gigabyte, not the 90 cents Jay talks up. This is $32/terabyte/year
At $32 per year, I suspect that AWS is way lower than the purchase price of a terabyte from NetApp, amortized over 4 years, NetApp has to include the drive, caddy, and a piece of the filer hardware and software to get a fair comparison, and also factor the cost of money.
What happens if a God Particle meets anti-matter?
Is the Higgs particle its own anti-particle. If not, the interaction of an anti-proton with one could be spectacular!
Assuming the tablet has a standard browser, server-based education and services are possible. This would easily overcome many of the deficiencies in the tablets. India is already developing education content.
We Westerners are spoiled from running our own copy of Word on our PCs. The server-based model is cheaper, especially at the discount level the Indian government could command.
Compress, deduplicate - What price per Terabyte??
At $35 per terabyte, which is Google's price for a 2TB drive, amortized over 4 years, that works out at a drive cost of $0.75 per month.
But it's much better than that! Compression and deduplication reduce stored data size by 85 percent in a business environment, so we are down to around 10 cents cost of the drive per month.
In the consumer space, compression isn't as good, due to the large number of images stored. Deduplication across users, however, is fantastic, if implemented. Reductions of 99 percent are possible.
There are more costs than drives of course, but bottom line, these prices could still go down further!
At Last - Real Science
For years I've watched articles nibbling away at the AGW issue, and many end with an obligatory sentence that, "Notwithstanding the above, AGW is warming the Earth", or some similar platitude.
It was career suicide for a scientist to open break with the Warmists. Now we find what we suspected. The models are subjective, and worse, look like they've been played up by unscrupulous or gullible people.
This is good science, insofar as it shows that much of the Warming story is POOR science. Heck, any model that ignore solar activity is broken! It's time we got real. We aren't going to roast. We aren't going to drown!
Of course, we could be like China and choke on our own pollution, and we are using up gas and oil faster than we should, so moving to alternatives makes some sense. Wind power is on the edge of a scam though, and solar may already be there. We need serious alternatives such as thorium reactors.
You must be a Heinlein fan! That was "Starship Troopers"
The article misses a couple of points. It's predicated on Block-IO and SANs being the future of storage, which is a bit out of touch.
Look at Object Storage and the shingled drive makes sense. In Object stores, data is written sequentially, and if a roll-back to previous versions is needed, any updates create a new block. Shingle drives work really well in this mode, since each written block just tags on the end of the last written block.
Archiving works the same way,and if the runes are read correctly, this is the future for hard drive storage anyway, with Tier 1 being SSD only.
Seagate's new interface makes a lot of sense. It's a lot more efficient than the existing software stack. It's time to revise the stack for all disks.
As for Helium, this isn't a magic substance. All it means is that HGST has figured out how to do a 6TB drive. They've an edge over their competitor, but that's only temporary!
Is this one a bit slow
The performance numbers seem a bit on the low side. 12 minutes to boot my desktop is a long time to wait, given it takes 30 seconds off my (inexpensive) SSD.
Garbage In Garbage Out
Most of the dire predictions of AGW are based on IPCC's model. This failed to take account of the sun's output variation, and of the fact that most sensors are near cities, which biases the average global temperature upward. Add to that a very unrealistic model of Greenland ice cap melt, and the Warmists had a story of gloom and doom to tell.
We are fixing the model, but it still can't acount for the 'hiatus', so it still is clearly very inadequate. Add to that all the 'warning signs' are turning into red herrings. The Arctic ice is roaring back, the PIG ice pack in Antarctica is growing like crazy, and winters seem to be getting much colder.
The reality is that the models don't work yet, and we should ignore them, but that would shatter the ricebowls of a lot of climate scientists, consultants and alternate energy companies. And a few politicians too!
An IBMer told me a story....
I heard from a director-level IBMer in the PC Division at the time that CPM actually was IBM's first choice, but that it was difficult working with Kildall.
He said, and I have not verified it, that they had arranged to call Kildall with a final decision to go or not, and asked he be by the phone. When they called with the good news, Gary was playing golf, and a call to the clubhouse resulted in a caddy driving the "mobile" phone out to Kildall. He was heard making a comment to "Tell those ******** I'll call them back when I finish my golf." That triggered Plan B at IBM!.
Might be true, might not. But the teller would have been on that call without question.
That could make Stephen fry close to accurate.
Contradict's your own recent article
It seems that you are contradicting your own article of a couple of weeks ago on this subject where the British Antarctic Survey found the PIG had retreated because of the loss of a submerged ridge which allowed warm water in.
Article is at http://www.theregister.co.uk/2014/01/03/antarctic_ice_shelf_melt_lowest_ever_recorded_just_not_much_affected_by_global_warming/
The irreversibility also is in question, since the thermocline has now dropped 750 ft (cold water in this area overlays warm water). This effectively blocks the warm flow, and ice is growing back rapidly. I'd note in passing that the quoted irreversibility is in fact the output of a computer model and follows the "GIGO" rule. This was in the original BBC article.
A job well done!
Thank you, Bill and Melissa!
Where is the Globe Warming?
Gosh, Lewis, when I last wrote on climate, the commentards jammed the Antarctic melt issue down my throat. You mean it's a myth, and those guys stuck in the ice aren't there because of AGW. I suppose the really nasty winters we are having are due to AGW, but I'm straining to find the connection anymore.
CNN ran "March of the Penguins" twice last night, so I'm sensing the Press is preparing us for bad news about Ice Ages. Al Gore may be the first recipient of both a Nobel and an IgNobel for the same work of fiction if this goes on!
Aussie is the only place that's getting hotter!
Following the pack
In fiddling the model, Mr Stern was just following IPCC precedents. After all, the basic computer model for Global Warming has been revised downward drastically in the last few years, in part because none of the predicted warming has occurred. So just plain guessing of the model parameters by Mr Stern, with judicious insertion of biased assumptions, would be nothing new.
I suppose those sea-based wind turbines could be used to anchor cruise ships and fishing vessels
Re: Why not just use SSD or flash
Sorry to deflate you, Been there, done that!
Re: Why not just use SSD or flash
These all seem like things you do to get around having slow storage. Why not just put in fast storage in the first place?
Why not just use SSD or flash
Starting the article with "Providing an Oracle database with 10,000 IOPS could mean aggregating dozens of 15,000 RPM drives, and unless the database is several terabytes in size that is a lot of wasted space" begs the issue of why you should just go to SSD.
No flash cache, no SSD tier...just use SSD. Heck, a Terabyte SSD is just $500 retail, which is about the same as a 15,000 RPM hard-drive. And you'll only need a couple to get 10K IOPS!
I'm surprised this discussion about how to get more from hard drives still comes up. It's over, and SSD won by several laps!
Old SIGs never die they just fade away!
Why bother? Ethernet has won!
This is a big plus for Intel, even if very belated. They are finally moving from their view that Wintel is the center of the universe to a much more sensible position. Despite the jokes about Chrometel and the rest, their direction is coming together, but they have a lot of ground to make up in some areas.
This will help enliven a computer industry that was rapidly becoming white box and commoditized. In the end, the losers are the proprietary CPU chip makers, and Microsoft.
Re: Took a long time...
There's now room for US hardware manufacturing as well as design companies. On-shoring, especially in niche markets, makes a lot of sense for smaller startups.
IPCC blaming heretics again?
It looks like climate change disaster scenarios are moving from the horror section to the comedy section of the store.
No heat since 1995, no melting glaciers, no ice falling off Greenland, and a really quiet Sun. All those spenders of billions for green energy got the story wrong. Now we'll need a new reason to wean ourselves off oil use!
HMC by another name!
When Intel pulled out of the Hybrid Memory Cube Consortium, the reckoning was that they intended to roll their own version, and this is it! It's probably put together with Micron as a partner, and is close to, but not exactly the same as, HMC.
The high end will be interesting. NVidia is in to HMC, AMD is likely working up the idea, too.
And the idea of memory stacking fits the mobe market as well, so ARM is in the game!.
Roll on persistent carbon nanotube memory. That will upset the applecart again!
HMC talks to Terabyte/second speeds, so it will be a big impactor on performance.
InfiniBand is not an escape route
Mellanox is trying to convert itself to an Ethernet provider with RoCE. I think the writing is on the wall for both FC and IB, as Ethernet surges, and as blockIO gives way to object access modes. Supporting object storage is the only way Brocade could move into a growth area.
IPCC doesn't help
I think Lewis hit the nail on the head. We need nuclear power, though I'd like a push for Thorium-cycle molten salt reactors, because they are cheaper, safer and create almost no long term waste. Thorium supplies run to thousands of years already, which gives us a bit of extra time to do fusion!
Of course, IPCC's fatuous insistence that we are heading toward climate disaster misses the point. We'll run out of oil and gas first. Their insistence on orthodoxy in thinking is constipating the debate. We need to talk to the fuel supply issues and not to Gaia-centric irrelevances.
I'll end with the standing disclaimer, cos I don't want to lose my grants. "Irrespective of any facts, and based on a model that is known to be totally inaccurate, Anthropogenic Global Warming will possibly happen sometime on a planet near Earth."
MTBF is a fiction?
With an 11.8 percent AFR in any of the four years, never mind earlier failures, the MTBF cannot be even close to the 2 million hours claimed by vendors. The number would be worse than 0.2 failures in 35,000 hours, which is 150,000 MTBF.
BlackBlaze may be using consumer drives in an enterprise mode, which increases duty cycle, but decreases start-stop. You have to read the small print. Even so, the drives don't seem to meet spec.
I'd believe SSD MTBFs, though. They should meet or exceed 2 million hours without a problem, except for early life failures due to manufacturing.
IPCC is really good at headlining doom and pushing the missing parts of the story under the table. Having done that for decades, they are now hoist on their own petard, when they can't explain why the model that had such dire prognostications no longer ties to reality..
This is so clearly a case of crying wolf!
- Product round-up Six of the best gaming keyboard and mouse combos
- China building SUPERSONIC SUBMARINE that travels in a BUBBLE
- Boffins attempt to prove the UNIVERSE IS JUST A HOLOGRAM
- Review Raspberry Pi B+: PHWOAR, get a load of those pins
- 6 Obvious Reasons Why Facebook Will Ban This Article (Thank God)