Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror

Submission Summary: 0 pending, 224 declined, 88 accepted (312 total, 28.21% accepted)

×

Submission + - How the Leap Second Bug Led Facebook to Build DCIM Tools (datacenterknowledge.com)

miller60 writes: On July 1, 2012 the leap second time-handling bug caused many Linux servers to get stuck in a loop. Large data centers saw power usage spike, sometimes by megawatts. The resulting "server storm” prompted Facebook to develop new software for data center infrastructure management (DCIM) to manage its infrastructure, providing real-time data on everything from the servers to the generators. The incident also offered insights into the value of flexible power design in its server farmss, which kept the status updates flowing as the company nearly maxed out its power capacity.

Submission + - U.S. Government Data Center Count Rises to 7,000

miller60 writes: The U.S. government keeps finding more data centers. Federal agencies have about 7,000 data centers, according to the latest stats from the ongoing IT consolidation process. The number started at 432 in 1999, but soon began to rise as agencies found more facilities, and exploded once the Obama administration decided to include server closets as well as dedicated data centers. The latest estimate is more than double the 3,300 facilities the government thought it had last year. The process has led to the closure of 484 data centers thus far, with another 855 planned over the next year. The GAO continues to call for the process to look beyond the number of facilities and focus on savings.

Submission + - Fidelity Now Selling Mutual Funds, Data Centers (datacenterknowledge.com)

miller60 writes: Mutual fund giant Fidelity Investments has developed its own factory-built data centers and will begin selling them to other companies. The company liked the benefits of modular data center design, including faster deployment and the ability to adapt to new technology, but was unsatisfied with the leading offerings, so it built its own. After using the design in its own facilities, Fidelity is commercializing its pre-fab units as Centercore. Fidelity's move follows the recent decision by another giant US brand, the retailer Sears Holdings, to enter the data center real estate market.

Submission + - Sears is Turning Shuttered Stores Into Data Centers (datacenterknowledge.com)

miller60 writes: Servers may soon fill the aisles where shoppers once roamed. Sears Holdings is seeking to convert former Sears and Kmart stores into Internet data hubs. Some stand-alone stores and distribution centers may be repurposed as data centers, while mall-based stores can be converted into disaster recovery sites, the company says, offering access to stores and eateries for displaced workers who may be on site for weeks. Then there's the wireless tower opportunity. Seventy percent of the U.S. population lives within 10 miles of a Sears or Kmart store, and these rooftops can be leased to fill gaps in cell coverage. It's not the first effort to convert stores into IT infrastructure, as Rackspace is headquartered in an old mall, and companies have built data centers in malls in Indiana and Maryland. But Sears, which operates 25 million square feet of real estate, hopes to make this strategy work at scale.
Hardware

Submission + - 'Data Center in a Box' Brings Colo to the Customer (datacenterknowledge.com)

miller60 writes: What if your company could have a data center delivered to its doorstep in less than 120 days? That's what IO has done for LexisNexis, using modular data centers to create a Tier III data center just minutes from the company's global headquarters in Dayton, Ohio. LexisNexis, which provides database and disaster recovery services for law firms, is the prototype customer for the on-site offering from IO, whose "data center in a box" offering is being adopted by Goldman Sachs and the Securities & Exchange Commission, which will use modules to house its EDGAR database. The concept has come a long way since Sun introduced the Blackbox container in 2006.
Hardware

Submission + - Data Centers Springing Up in Old Chip Fabs (datacenterknowledge.com)

miller60 writes: A growing number of former semiconductor fabs are becoming server farms. A data center company will convert a former fab near Dallas into a massive data center that will offer between 700,000 and 1.4 million square feet of server space. The company, QTS, followed a similar path with a huge former Qimonda facility in Richmond. In Silicon Valley, Facebook's servers live in an old Seagate plant converted by Fortune Data Centers. Up in Oregon, a former Hynix chip plant is becoming a data center. These facilities offer several attributes that make them good covnersion prospects: lots of existing power and cooling capacity, with raised floor already available in many faciltiies. Data center companies says this existing infrastructure saves them money, leaving less to retrofit.
Supercomputing

Submission + - Titan is New Champ in Supercomputing's Top500 (top500.org)

miller60 writes: The new Top500 list of the world's most powerful supercomputers is out, and the new champion is Titan, the new and improved system that previously ruled the Top500 as Jaguar. Oak Ridge Labs' Titan knocked Livermore Labs' Sequoia system out of the top spot, with a Linpack benchmark of more than 17 petaflops a second. Check out the full list, or an illustrated guide to the top 10.
Hardware

Submission + - New York Data Centers Battle Floods, Utility Outages (datacenterknowledge.com)

miller60 writes: At least three data center buildings in lower Manhattan are struggling with power problems amid widespread flooding and utility outages caused by Hurricane Sandy. Flooded basements at two sites took out diesel fuel pumps, leaving them unable to refuel generators on higher levels. One of these was Datagram, which knocked out Buzzfeed and the Gawker network of sites. At 111 8th Avenue, some tenants lost power when Equinix briefly experienced generator problems.
Hardware

Submission + - Rack Falls Down, Goes Boom - From 18 Stories (datacenterknowledge.com)

miller60 writes: In the name of failover science, Symantec dropped a rack of expensive IT gear off the roof of an 18-story building in San Jose. This experiment confirmed the results of 2007 research by HP, which blew up racks of IT gear to see whether auto-failover features would work. Perhaps some IT staff simply long to see their servers explode or fly through the air.
Hardware

Submission + - Newspaper Publisher Enters the Data Center Business (datacenterknowledge.com)

miller60 writes: Want some servers with your news? While some newspaper companies are struggling with the digital transition, a Missouri publisher is getting into the data center business. The News-Press & Gazette will invest $20 million in Online Tech, a data center service provider. The News-Press, a 100-year old family-owned company, runs newspapers and TV stations in secondary markets. "The News-Press has seen that journalism is more and more becoming digital, and at the heart of everything that's digital is data centers," said Mike Klein, CEO of Online Tech.
Security

Submission + - Go Daddy: Network Issues, Not Hacks or DDoS, Caused Downtime (datacenterknowledge.com) 1

miller60 writes: GoDaddy says yesterday's downtime was caused by internal network problems that corrupted data in router tables. "The service outage was not caused by external influences,” said Scott Wagner, Go Daddy’s Interim CEO. “It was not a ‘hack’ and it was not a denial of service attack (DDoS)." The outage lasted for at least 6 hours, and affected web sites and email for customers of the huge domain registrar.
Hardware

Submission + - Server Farms Step Up Efforts to Reduce Water Waste (datacenterknowledge.com)

miller60 writes: How much water does your data center use? Is it more or less than last year? The largest data centers are working to slash their water use, and the industry has developed metrics and best practices in hopes of reducing the impact of server farms on local potable water supplies and sewer capacity. Facebook sees data center design as the key to reduced water impact, and last week published data on its water use and efficiency, Google and Microsoft have focused their efforts on using recycled "gray water" in their cooling systems, rather than potable water.
Cloud

Submission + - Power Outage Causes Downtime for Salesforce.com (datacenterknowledge.com)

miller60 writes: It's already been a tough summer for uptime. Salesforce.com experienced a lengthy service outage this morning, attributed to a power outage at an Equinix data center in Silicon Valley. Meanwhile, a UPS failure knocked out power at a Level 3 facility in London. The outages come less than two weeks after major downtime for some cloud computing customers of Amazon Web Services. So far 2012 is looking like a replay of the summer of 2009, when major data centers suffered a series of power failures in late June and early July.
Facebook

Submission + - Facebook Revamps Designs for Servers, Storage (datacenterknowledge.com)

miller60 writes: Just a year after open sourcing the plans for its custom servers, Facebook is updating its hardware designs to adapt them for its Open Rack enclosure standard. That's led to changes in the form factor for servers and a new power system with "outboard" power supplies that are housed on a power shelf at the base of the rack rather than the servers. Facebook's hardware designers provide hands-on video overviews of the storage prototype, as well as changes to their racks and servers. Facebook has also overhauled its network topology to run its entire network at 10 gigabits per second.
Cloud

Submission + - Defective Fan Triggered Amazon Cloud Outage (datacenterknowledge.com)

miller60 writes: A defective generator cooling fan wound up triggering a lengthy outage last week that affected customers of Amazon Web Servvices, including sites like Pinterest, Quora and HootSuite. The issues began when an Amazon data center in Virginia lost utility power. The backup generators started as designed, but the bad fan caused one unit to overheat and shut down. The system was then supposed to switch to a seperate power distribution circuit, but an improperly set breaker caused that process to fail as well.

Slashdot Top Deals

New York... when civilization falls apart, remember, we were way ahead of you. - David Letterman

Working...