SAN FRANCISCO — The overarching theme here at ChefConf 2013 this week has been a sense of empowerment for IT folk who are used to shoveling coal in the boiler room of an IT infrastructure.
As cloud computing and scripting languages like Chef and Puppet mature, IT will captain the ship, predicted conference attendees and presenters.
An especially resonant quote from a book called The Phoenix Project was recited from the keynote stage on Thursday morning to drive home this point: “Any COO who doesn’t intimately understand the IT systems that actually run the business is just an empty suit.”
It’s easy to dismiss all this talk as self-serving posturing, but in an increasingly Web-based world, it happens to be true. Consumers now interact with businesses through mobile and Web interfaces much more often than on a sales floor or over the phone. And when was the last time a new startup opened up a storefront without a Web presence? It simply isn’t done these days.
No, now the keys to the castle belong to your company’s computer guy. Let’s hope you’ve been nice to him.
It’s also easy to get bogged down in the intricacies of cookbooks, recipes, the finer points of erlang vs. Ruby, Chef vs. Puppet, DevOps vs. traditional operations, at a conference like this. The bottom line, though, is that it’s all a means to the end of making the current world economy work.
Heady stuff. Let’s hope IT wields its newfound power only for good.
Some analysts estimate that Moore’s Law — the ability of manufacturers to double the number of transistors on a chip every 18 months — will end in 10 years. At a certain point, the electrons of silicon circuits become unstable and can no longer be used to process information. While the end of the road may be coming for silicon-based transistors, there may be a rising contender to the computing throne.
The 2010 Nobel Prize in physics went to two men — Andre Geim and Konstantin Novoselov — for their research on graphene. The graphite-based material is perfect for data center applications — not to mention consumer electronics, high-speed Internet networking and medical equipment — because it is not only highly conductive, but it also handles electron movement and heat better than silicon.
Graphene has been tapped as a possible material to make flexible touchscreen displays for consumer devices, and its single-electron properties makes it an excellent choice for the transistors of the future.
Envision, if you will, the size of a current enterprise data center. Now imagine if each processor could hold 3 billion transistors and take up only the width of a pencil tip. Add in the fact that a processor based on graphene transistors would run cooler than current silicon-based processors, and you’ve got a change in cooling infrastructure as well.
If graphene keeps rolling at its current development pace, we’ll be in the future before you know it.
There’s been a lot of bluster about ARM servers and how they’ll upend the traditional x86 market when 64-bit ARM offerings come to market, but the fact remains that software designed on x86 servers does not readily translate to ARM architecture. To ease this transition, server maker Boston is offering developers a platform to rework x86 code without having to buy an ARM server through its cloud-based ARM as a Service.
Boston’s product — reminiscent of OpenStack’s TryStack.org which encourages the use of OpenStack on ARM servers — aims to coax curious but wary developers into an ARM sandbox by providing cloud servers that come with a pre-installed flavor of Linux along with development platforms and porting tools.
News of Boston’s AaaS (no, really) comes hot on the heels of news that China’s Baidu has decided to go with Boston rival Marvell’s ARM servers to run its behemoth of a search engine.
As intriguing as Baidu’s ARM evolution and Boston’s cloud service are, will they generate enough buzz for mainstream IT ARM server adoption? Only time will tell.
IBM plans to work with Nokia Siemens Networks to make our smartphones more responsive.
What does this mean for the data center? Well, we’ve learned about placing small data center gateways closer to where the information needs to be, which is very similar to what IBM’s partnership spells out.
According to IBM’s press release, the idea is to get mobile data into smaller chunks to be stored and processed closer to its source to reduce latency.
“This enables a huge amount of rich data to be processed in real time that would be prohibitively complex and costly to deliver on a traditional centralized cloud,” said Phil Buckellew, vice president of IBM Mobile Enterprise, in the release.
Will we see more unused industrial spaces turned into data centers or will there be tiny, containerized data centers popping up on street corners? It will be interesting to see what shape these improvements take.
Linux jobs are in high demand for another year, and there are just not enough seasoned Linux admins to fill the spots available, according to The Linux Jobs Report from Dice.com and The Linux Foundation.
That’s good news for experienced IT staff looking for a change. If you have expertise in Linux, then the only trouble you’ll have is narrowing down which positions you find interesting.
According to the report, these job seekers are in a position to settle for nothing but the best. The report cites money, work/life balance and flexible hours as the top of the “wants” list. Luckily for job-seekers, companies are putting their money where their jobs are and are offering higher salaries than many other tech jobs. Plus, those already high salaries increased on average by 9% from 2012, says the report.
Now seems like an appropriate time to mention that Red Hat Enterprise Linux 6.4 just came out with several new features, just in time for Linux pros who wanted to put more on their resume.
Anyone else want to go back to school right now?
We often speak cavalierly about building new data centers and all that goes into them. But unless you’ve actually done it — and I haven’t — it’s easy to take all the effort and talent involved for granted. Just remember, without the data center architects, your smartphone would be decidedly less so.
Once you have the building itself, there are power busways, networking cables and cooling infrastructure to install, not to mention workstations and any creature comforts you want to include for employees. On top of that, you have the cabinets and racks, servers and storage hardware to roll in, all of which is delicate and expensive. Oh, and you have a time limit: Yesterday.
This video from LeaseWeb really brings home just how much goes into setting up the data centers we rely on for computing. It doesn’t mention whether they had to build the facility itself or just needed to move in the equipment and infrastructure.
So even if you’re busy buying chocolates or ties for your Valentine’s Day sweetheart — or picking your zombie flick for Feb. 15’s Singles Awareness Day — take a minute to thank an engineer in your life, and maybe give her a hug.
Chip manufacturing giant Qualcomm has listed job openings for ARMv8 engineers, which makes it the latest vendor to show interest in building 64-bit processors with ARM’s latest design. Though Qualcomm is by no means the first to enter the 64-bit ARM server fray — Calxeda, Marvell, Applied Micro and others have already gotten their feet wet in the ARM market — it’s an interesting development considering the company’s history has been rooted in smartphone and tablet processors, not servers. Since Qualcomm’s project is still in its infancy, many details are still unknown, such as how many cores per processor and what form the chassis will take.
The use of advanced RISC machine (ARM) processor chips is prevalent in mobile devices, but until recently ARM’s designs were limited to 32-bit architecture which is not enough to handle enterprise-sized workloads. Now that ARMv8 is in the hands of chip designers, the energy efficiency of this architecture — just 5 W per chip in HP’s Calxeda-based Redstone Server — may start turning more heads away from Intel’s x86.
However, Intel isn’t taking the ARM threat lightly. Ahead of the low-powered army of ARM chips on the horizon for 2013 and 2014, Intel has put a trio of processors in its S1200 line up for purchase. The power draw ranges from 6.1 to 8.1 W which is still a bit more than ARM chips, but that’s not the only deciding factor for most IT shops.
Intel’s x86 architecture is well-established — 20 years, give or take a few — and for ARM to really take off, it needs server hardware and software companies to get on board. So far the groups making the biggest splash are Red Hat’s Fedora ARM project, which has been vocal in its efforts to raise awareness of the low-powered chip’s usefulness in such markets as hyperscale computing and web development, and Microsoft, which hopes to come up with 64-bit ARM support for Windows — both desktop and server.
Is there room in the market for both low-powered options? We’ll get back to you in a year or two.
It’s hard work discovering the secrets of the universe, and now the Einstein@Home distributed computing project has broken the petaflop – the compute power of approximately 61 million iPad 2 processors – barrier in its quest to find pulsars and gravity waves.
“But,” you might say, “Supercomputers can do circles around that!” True, but if you look at a list of the top 500 supercomputers, only the top 23 run in the petaflop range. With such a large volume of collected scientific data to sift through, having a petaflop of extra computing resources is nothing to laugh at. And overall, distributed computing contributes much more than that.
At the time of publication, the distributed computing projects based on the Berkeley Open Infrastructure for Network Computing (BOINC) software system showed a daily average of 7.2 petaflops across more than 720,000 computers. For perspective, the combined power of BOINC-based projects would rank fifth on the list of top supercomputers.
Distributed and grid computing networks like Einstein@Home, SETI@Home and the World Community Grid provide much-needed, cheap computing power for workload-intensive research projects, and cut down on the costs of building supercomputers. The grid model can also provide scalability and redundancy that a data center would struggle to achieve.
So which will it be: Finding alien intelligence with SETI, proving Einstein’s theories or helping to predict the climate?
Did you enjoy yesterday’s spine-tingler? Then you’ll enjoy today’s terrifying tale of data center despair.
“Total shutdown: A cautionary tale”
Sander van Vugt
Don managed a small data center for a school. He was so confident with the data storage on his servers, that he always urged students to store their data on the network drives alone. “Data that is stored there will be backed up by the IT department, so you’ll always have access to the data!” he told wary students.
A few days before the day that senior theses were due, the last Friday of October, when the IT department’s phone started ringing. Mark, the help desk guy, answered it.
“No access to the data volume,” he called over to Don.
Mark hung up the phone, and it immediately rang again.
“One more,” said Mark.
“Tell them we’re working on it,” Don replied.
After half an hour the volume still wasn’t available. With a growing sense of dread, Don decided to reboot the server, backed by Alex, the external specialist. Alex couldn’t find a way to make the volume accessible again either.
When they walked out of the office toward the data center, they met a group of very nervous senior students.
“No comment,” Don said nervously as he pushed past.
Once in the data center, Alex and Don found the server and rebooted it. While booting it gave at least 50 different error messages when it completely crashed and hung.
“What do we do now?” said Don, starting to sweat. The specialist tried to reassure him. “Don’t worry I’ve got it all under control.”
Beyond the sealed, secured doors of the data center, the small crowd had become a shambling, moaning mass of sleep-deprived students.
“You’ve got them under control too, right?” Don asked.
Very late that evening Don and Alex cautiously ventured from the data center to take the train home. The crowd, not to be deterred, shuffled behind, finally caught up to the two technicians at the train station.
Alex’s train arrived before Don’s, so they separated at the railway station. Just as Don’s train arrived, the students shuffled into view. He boarded the train, hoping the doors would close quickly. Don called out to the dead-eyed students. “Wait until tomorrow morning when manual recovery is finished.”
Don and Alex had tried everything, including pulling from the backups that the IT department was supposed to make every day. Unfortunately, they hadn’t verified the tapes so the backups were empty. The manual restoration had not been successful.
Alone in the train car the next morning, Don wondered what he would tell the students. Once the train pulled into the station, he knew it was too late. The shambling horde of students came forward as the train doors opened.
When Don didn’t show up, Alex called him several times but all he got was voicemail. Don, it seems, hadn’t been seen since he got off his train that morning. The only thing left of Don was a broken backup tape at the railway station and his school ID badge. He was never seen again.
Grab a flashlight, head to your nearest lights-out data center, and hunker down for some creepy data center horror stories courtesy of some data center experts. Are these true tales? Stay tuned for tomorrow’s part two.
Data center fright night, part one
“Night of the Living Data Center”
All was quiet and dark – well, it was a lights-out data center after all. The lone admin had been called in. Something had gone wrong in the facility. He went to push the door and – nothing. He remembered he had to use his two-factor secure key, finger print and retina recognition to get inside.
Creeping past the reception desk, a low moan sent shivers down his spine, but it was just the murmurings of the under-paid security guard catching up on his sleep after a 12-hour shift at the local Walmart. Moving silently along, the admin heard “zip, click, zip, click, zip, click” as he walked along. Glancing behind him, he saw the lights slowly clicking off as he moved along. In his befuddled state, it took some time to realize this was the movement-sensitive lights turning on and switching off along the corridor he was in.
At the final interior door, he waited, listening in case something was beyond the door that shouldn’t be. Mind you, as the data center had been built to LEED quality, the amount of insulation in the walls would have deadened the sounds of the demons of hell escaping from the Solaris box he had in the far corner. Screwing up his courage, he opened the door and tip-toed in. There was enormous “CRACK!”
No, sorry, there was an enormous CRAC – it had been there for 20 years, so it shouldn’t have surprised him. Another low moan, far different to the security guard’s sleeping sounds, went past him in a gust of warm, fetid air. He made a mental note to check the raised floor for holes where cooling might be leaking and get the drains fixed.
He made it to the console and logged in. Strange images appeared in front of his eyes – ones he had never seen before. Immediately switching to CLI mode, the easier, more functional and useful graphical user interface was instantly hidden. His fingers flew over the keyboard. Eventually, he calmed down and started typing. The strange incantations he typed brought up steams of glowing data – log files, audit trails, details of pizza delivery companies.
Eventually, tracing the root cause of the issue caused him to take a sharp intake of breath. “How could this happen?” he wondered. Well, it was a waste of time asking anyone else, since they were all in bed at their homes. With a rising sense of panic, he gathered up the tools of his trade – the old Bell telephone modem he might need where he was going, the book on troubleshooting systems and the 1 million candle flashlight. For good luck, he also took a silver thumb drive. Moving away from the console, he muttered some charms under his breath “agility, flexibility, SaaSability” and “I wish we had outsourced this years ago.”
In the corner was a massive black monolith – the admin had always assumed that this was a throwback to the film “2001: A Space Odyssey” and that it was something facilities looked after – nothing for a tech guru like him to worry about. However, the console had provided directions that said this was the source of the trouble. Reading the book, the admin identified the special way to use the runes on the monolith to access secret bits. Finally, he folded down a screen and pushed a switch. The green and black screen flickered, showing a talismanic figure, which changed over into a single word and sent the admin screaming back into the night.
In dripping, blood-red letters, it said “ABEND.”