Friday, August 8th, 2014

Time	Event
12:00p	Apple’s Reno Data Center Expansion Marches On In preparation for the next phase of data center construction at its Reno, Nevada, campus, Apple has applied for planning permits to stand up more data center buildings there. The company is asking for permission to build two more buildings at the site in Reno Technology Park, Reno Gazette-Journal reported, citing documents the company filed with Washoe County. Apple currently has six “data cluster” buildings in RTP and an administrative facility, four of which have been completed and two still under construction, Trevor Lloyd, a senior planner with the county, told Data Center Knowledge in an interview. Apple appears to be building the facility out in modules it refers to as data clusters. However, the buildings are interconnected, Lloyd said. Phase II of construction at the Apple data center site is currently underway, slated for completion before summer’s end. The company owns 345 acres in RTP, currently building at the park’s southern end but planning to build at its northeast end as well, according to the Gazette-Journal. RTP was envisioned by a team of developers called Unique Infrastructure Group. The property provides access to a wide variety of energy sources, including hydroelectric, wind and geothermal power. The developers also envisioned solar farms at the site – a vision Apple is working to realize with a project to build a 20-megawatt solar installation. Building the solar farm together with a local utility, the company plans to finish the project in early 2015. Apple has a standing public commitment to making 100 percent of the power it consumes renewable, and has been buying geothermal energy to power the entirety of its Reno operations, which it said it would continue doing until the solar array is completed. In the environmental responsibility report the company released last month, it said its Reno data center facilities had consumed 3,000 megawatt-hours of electricity in fiscal 2013. This is one of the newer location in the company’s data center fleet and therefore consumes less energy than the others. Its other data centers are in Prineville, Oregon, Maiden, North Carolina, and Newark, California. It also has some capacity deployed at colocation data centers, but says that most of its workload is served out of its own facilities. Fiscal 2013, which ended in September of last year, was the first year Apple’s data centers consumed more energy than any of the company’s other corporate facilities. Cumulatively, its data centers used more than 300,000 megawatt-hours that fiscal year. (Comment on this)
2:00p	Data Center Jobs: Allstate Insurance Company At the Data Center Jobs Board, we have a new job listing from Allstate Insurance Company, which is seeking a Data Center Facilities Leader in Rochelle, Illinois. The Data Center Facilities Leader is responsible for managing human resources, hiring, training, coaching, salary administration and various management administrative tasks, continuous operation of the facility, optimizing equipment performance, ensuring the building interior and exterior are kept clean and in a like-new condition, ensuring the surrounding grounds are maintained and kept neat in appearance, and recommending and implementing changes where necessary in keeping with current Allstate facility and work place standards. To view full details and apply, see job listing details. Are you hiring for your data center? You can list your company’s job openings on the Data Center Jobs Board, and also track new openings via our jobs RSS feed. (Comment on this)
4:15p	Five Texas Towns Consolidate 911 Dispatch in CyrusOne Data Center CyrusOne will house what it says will be the first ever colocated 911 dispatch center. Dispatch centers for Texas cities of Carrollton, Coppell, Farmers Branch and Addison will share 12,000 square feet of the provider’s Carrollton facility. Carrollton is just outside of Dallas. The communities are consolidating 911 services from the North Texas Emergency Communications Center (NTECC). The consolidation is expected to reduce costs for 911 services while maintaining strict uptime. The project is expected to be complete by early 2015. There’s very little that is more mission critical than 911 dispatch. Prior to this, dispatch centers would buy a facility and retrofit it according to their specifications. ““CyrusOne offers us a facility that would have taken us years to build,” said Gary Greer, NTECC board president. It is a win not only for CyrusOne but for outsourcing in general. It validates that colocation is a viable alternative to in-house for one of the most important functions in the country. CyrusOne will house call center workstations and meeting rooms and provide a technology space. 911 dispatch gains peace of mind that infrastructure will remain up and on highly reliable electrical and mechanical systems. “Essentially, the dispatch center will operate within our Carrollton data center environment,” said Dottie Spruce, executive vice president of sales at CyrusOne. “The recent consolidation will allow the cities to be much more efficient, improving response times and no doubt saving lives. We’re thrilled to work with the NTECC, and we hope to serve as a model to other dispatch centers across the country.” NTECC board president Greer said consolidation would improve the service and cut cost at the same time. “The closest unit available will now be able to respond to the scene of a crime or accident, no matter the jurisdiction, saving precious time. We won’t have instances where someone needing assistance would have to wait for a unit in their jurisdiction that could actually be further away because they sit right on a city line.” CyrusOne has been expanding at rapid pace across its footprint, particularly in Texas. The company landed another blockbuster in Carrollton earlier this year. Wikimedia Foundation chose Carrollton to house infrastructure for one of the world’s busiest sites, Wikipedia. (Comment on this)
5:00p	Friday Funny: Colorful Rack Mount Power Strips We’re adding extra color to our Friday fun with this week’s caption contest. Help us complete our Kip and Gary cartoon – submit your caption below! Diane Alber, the Arizona artist who created Kip and Gary, has a new cartoon for Data Center Knowledge’s cartoon caption contest. We challenge you to submit a humorous and clever caption that fits the comedic situation. Please add your entry in the comments below. Then, next week, our readers will vote for the best submission. Here’s what Diane had to say about this week’s cartoon, “I’ve started to see a new trend in data centers – color coding A and B feed Rack Mount Power strips as well as power cords so there’s no confusion which side you’re working on. Well . . . as usual, Kip got a little carried away!” Congratulations to the last cartoon winner, Ed, who won with, “We never thought about the noise and smell when we decided to move the farm into the cloud!” For more cartoons on DCK, see our Humor Channel. For more of Diane’s work, visit Kip and Gary’s website. (Comment on this)
5:30p	OneNeck IT Expands in Growing Minnesota Data Center Market Data center, managed services and cloud provider OneNeck IT opened a $12 million expansion in Eden Prairie, Minnesota. An additional 6,000 square feet of raised floor has been added to the Tier III Uptime Institute Certified facility. The building consists of 18,000 square feet of raised floor space. In addition to the new space, biometric security functionality has been added, as well as a general update to the layout. “For nearby customers, their next visit to our data center should also prove to be more pleasing and efficient,” said Hank Koch, vice president of mission critical facilities at OneNeck. “Shipping, receiving and storage areas have been improved, building navigation is easier, the facility command center has been enlarged and the entire entrance along with meeting room areas have been updated.” OneNeck offers managed services and generates high revenue per square foot. It targets compliance requirements facing businesses today, such as SSAE 16, PCI-DSS and HIPAA. It currently operates six data centers with one on the way in Englewood, Colorado. The expansion is yet another indication of a growing Minnesota data center market. Minnesota has emerged as one of the most promising second-tier markets in the country, with a flurry of service providers deploying new mission-critical space. In the past few years, ViaWest opened a 9 megawatt facility in the market; DataBank is planning a 20 megawatt data center; Cologix has been expanding rapidly; Stream Data Centers is building a 75,000-square-foot data center; and Compass Datacenters partnered with CenturyLink on a new facility, which recently received Uptime Tier III Certification across design, construction and commissioning.   (Comment on this)
6:00p	DataSite Upgrades Critical Infrastructure in Orlando Data Center DataSite completed a $12 million critical systems upgrade in its Orlando data center, adding power and cooling capacity. The company provides wholesale and retail colocation space for medium and high power density computing environments. The single story, free-standing building is 130,000 square feet. It was built to withstand hurricanes, one of the most potent natural disasters in the state. The facility is Category 5 hurricane rated and is 60 miles inland, outside of the 500-year floodplain. The exterior is windowless and has 8-inch-thick steel-reinforced concrete walls built from the underground up. Power is delivered from two separate utility substations through two compartmentalized and isolated rooms. The upgrades include: New UPS system delivering 6.25 megawatts total Standby power plant of 13.4 megawatts and utility service capacity of 12 megawatts 2N electrical support infrastructure for redundancy New air-cooled chiller plant with total system cooling capacity of 2,300 tons A/B header system providing 2N system level redundancy. “We are thrilled with the quality and efficiency delivered by TSS with this expansion,” said Jeff Burges, president of DataSite said, commenting on the upgrade work done by contractor TSS. “TSS is our go-to partner for design and construction of all our data center facilities. They are uniquely able to complete these types of upgrades seamlessly.” Chief technicaloOfficer of TSS Jerry Gallagher noted the significant upgrades to critical mechanical and electrical infrastructure were performed without interruption or impact to normal data center operation. DataSite is owned by BURGES Property + Company and operates facilities in Orlando, Atlanta and Boise, Idaho. It entered Boise earlier this year by purchasing a former Bank of Idaho building. Last year, the company undertook an expansion in Marietta, an Atlanta suburb. (Comment on this)
7:00p	Facebook’s New Cluster Management Software Cuts Power Consumption Every time Facebook’s data center engineers figure out a way to reduce server consumption by a single watt, the improvement, at Facebook’s scale, has the potential to add millions of dollars to the company’s bottom line. This is why its infrastructure team never stops looking for new ways to win back some power. The company’s product consists of bits and bytes, and its data centers are its manufacturing plants. The more efficiently Facebook servers can convert power into text, pictures and video, the wider its profit margin becomes. The company has optimized hardware and supporting data center infrastructure across the board, and the next step is using software management tools to utilize the infrastructure more efficiently. The latest such tool it has come up with is Autoscale. It was designed to work in tandem with load balancers that distribute workload among servers in each cluster to make sure the load balancers take energy efficiency into consideration when doing their job. Qiang Wu, infrastructure software engineer at Facebook, described Autoscale in a post on the Facebook Engineering blog Friday. The system is already running on production clusters in Facebook data centers and has been effective in substantially reducing energy consumption. ‘Round-robin’ load balancing is inefficient Workload on Facebook servers fluctuates throughout the day, peaking around noon and dropping substantially around midnight. Until now, a load balancer would distribute the load evenly among servers in a cluster it manages, regardless of the size of the workload. Thus, CPU utilization would drop to very low levels across the board during low-demand periods and increase across the board as demand grew. The team learned, however, that this “round-robin” approach to load ballancing is not really the most efficient way to go. “As a result, during low-workload hours, especially around midnight, overall CPU utilization is not as efficient as we’d like,” Wu wrote. One type of a web server Facebook uses, for example, draws 60 watts when idle, 130 watts at low-level CPU utilization and 150 watts at medium-level utilization. The difference in power consumption between low-level and medium-level utilization is small, while the difference in the number of requests processed at the two levels is quite large. Therefore, to increase overall output-per-watt ratio, it is better to avoid running servers at low-level utilization, sticking to either idle or medium level. Fighting the evil of low CPU utilization This is where Autoscale comes in. An Autoscale controller collects request volume and CPU utilization data from a cluster and dictates to the load balancer responsible for that cluster how many servers should process the traffic at hand. By throttling the amount of active servers in a cluster, the software increases utilization of each active server, leaving the others idle. Autoscale adjusts the size of the active server pool dynamically. To decide how big the optimal active server pool should be, Facebook engineers model the relationship between CPU utilization and requests per second, among other factors. While energy efficiency is important, it is also important not to load a subset of hardware to a point where performance starts suffering. “We conduct experiments to understand how they correlate and then estimate the model based on experimental data,” Wu wrote. Autoscale works as intended Results of implementing Autoscale in production are impressive. The team described normalized energy consumption data from one of its production web clusters managed by Autoscale over a 24-hour period, showing a 27-percent power-draw reduction around midnight: Normalized power consumption for one of Facebook’s production web clusters with and without Autoscale. The savings decrease gradually as the workload grows, eventually reaching ‘zero’ at peak demand. Average power savings over 24 hours from a number of different web clusters is between 10 percent and 15 percent. “In a system with a large number of web clusters, Autoscale can save a significant amount of energy,” the engineers wrote. (Comment on this)

<< Previous Day

2014/08/08 [Calendar]

Next Day >>