[Most Recent Entries] [Calendar View]
Thursday, July 31st, 2014
- 4 and 8 ARM Cortex-A57 cores
- Up to 4 MB of shared L2 and 8 MB of shared L3 cache
- Configurable dual DDR3 or DDR4 memory channels with ECC at up to 1866 MT/second
- Up to 4 SODIMM, UDIMM or RDIMMs
- 8 lanes of PCI-Express Gen 3 I/O
- 8 Serial ATA 3 ports
- 2 10 Gigabit Ethernet ports
- ARM TrustZone technology for enhanced security
- Crypto and data compression co-processors
| Time | Event |
| 12:00p | Hulu Chooses Cassandra Over HBase and Riak Hulu surpassed the 6-million-paid-subscriber mark last April. The site is now accessible on 400 million internet-connected devices ranging from new consoles, Chromecast, to set-top boxes, such as Amazon’s Fire TV. Andres Rangel, the company’s senior software development lead, spoke of the intricacies of scaling the infrastructure to support this user-base growth. The application has to support a lot of users generating lots of data, as well as maintain sessions for these users across devices. In a three-way competition between Cassandra, HBase and Riak, Hulu turned to Cassandra as the database of choice to ensure high-quality user experience. Hulu depends on the open source distributed database management technology to store and provide real time-access to subscriber watch history. Cassandra keeps Hulu’s content infinitely scalable and always available. “Two years ago, we decided to re-write our service,” Rangel said. The challenge was to make the service scale to keep up with user growth . The experience has to follow a user from device to device, saving the previous session and starting up a show where it was left off. “The problem was we couldn’t scale the writes,” he said. “The boxes couldn’t keep up with the growing database demands, and it wasn’t easy to increase the hardware without a lot of effort. It wouldn’t scale.” Rangel had a small team of people, and they needed something that would scale, but wasn’t overly complicated to manage. Hulu’s primary Cassandra cluster consists of 32 nodes split between two data centers, one on the east coast and the other on the west coast. Its watch history keyspace contains several billion CQL3 rows with approximately 1TB of unreplicated data per data center. It’s tapped into every time someone watches a video or uses Hulu for a recommendation. It’s also used for saving sessions across devices. Path to Cassandra“We looked at HBase and Riak at first,” said Rangel. “Cassandra was an afterthought.” Rangel says his experience with Riak was that it was able to scale, but performance wasn’t as great as with Cassandra. “We also needed it to be able to do range queries, and Riak didn’t do this at the time. Another problem was that no-one in the team was experienced with Erlang [the basis of Riak].” Rangel and his team exemplify a trend among modern IT teams, where a handful of people have immense responsibilities across massive scale. While Riak was powerful, it didn’t quite fit Hulu’s needs because it wasn’t easy enough to use and didn’t fit the real-time needs. The team then looked to HBase as the front-runner. “Hadoop instances take a lot of work to set it up,” said Rangel. “Hbase runs on top of hdfs, and the Name Node was a single point of failure. Also it was more complex to set up and maintain than Cassandra.” There were some concerns with how Hbase handles failures (e.g., the team saw cascading failures take down all region servers). When they turned to Cassandra they found a match. “With Cassandra, it managed to handle the load, it’s very reliable, it allows range queries without limitations, and it’s easy to maintain,” said Rangel. “It’s night and day compared to HBase.” The team had to do some hardware changes because Cassandra specs are different. Cassandra is optimized for SSDs, which improved performance. Rangel also said that Cassandra was better at replication. Hulu does use a Hadoop cluster for long-term storage, but Cassandra was the right choice when it came to maintaining real-time access. “We played around with the new version of HBase for high availability,” said Rangel. “The problem is you still need a lot of babysitting. If you have an already existing Hadoop cluster, than HBase makes sense.” There are now four other services running on Cassandra besides real-time access across devices. It handles social data from the user, some messaging, and you can now use your cell phone to send traffic to your connected device like a remote. “If there are two things I’d like to say about Cassandra, it’s that we haven’t had any bad experiences with it, and it’s been better than expected,” said Rangel. |
| 12:30p | Hurricane Electric PoPs into Telx Data Center in San Francisco Hurricane Electric, a Silicon Valley-based colocation and network connectivity services provider, has installed a Point of Presence (PoP) in a Telx data center at the Digital Realty Trust-owned facility at 200 Paul Avenue in San Francisco. The PoP gives Telx customers at the site access to an additional network provider and the ability to peer with Hurricane’s global network. The provider offers both IPv4 and IPv6 connectivity (it is one of the biggest players in the IPv6 market), and its network consists of four redundant paths across north America, two separate paths between the U.S. and Europe and rings in Europe and Asia. Headquartered in Fremont, California, the company claims to operate the world’s largest IPv6 network, which it started building in 2001. The estimate is based on the number of networks it is connected to. Each of its customers gets both IPv6 and IPv4 services over the same connection. It offers 100 Gigabit Ethernet, 10 GbE, 1 GbE and 100BaseT connectivity. Commenting on the expansion into 200 Paul, Hurricane President Mike Leber said, “The new PoP will give customers of [the data center] additional connectivity options, while expanding access to next-generation IP service.” Telx is a tenant at 200 Paul – a 145 megawatt facility south of downtown San Francisco. The company has positioned itself early on to capture demand for IPv6 services as the number of available IPv4 addresses continued to shrink. In April, American Registry for Internet Numbers reported that the Internet had entered the final phase of the countdown to the last IPv4 address. ARIN was down to the final 16 million addresses. Hurricane provides colocation services out of two large data centers in Fremont and one smaller facility in nearby San Jose. Its Fremont data centers are 45,000 square feet and 200,000 square feet, and the San Jose facility measures about 3,000 square feet. |
| 1:00p | AMD Starts Shipping 64-bit ARM Server Developer Kit AMD has launched a developer kit that includes its first 64-bit ARM processor codenamed “Seattle.” The chip maker started sampling the product earlier this year, according to an announcement it made in January. Its biggest competitor in the space, Applied Micro, has already started shipping production units of its 64-bit ARM System-on-Chip (SoC) called X-Gene. “We are very pleased to report that we have shipped initial production X-Gene units,” Applied Micro President and CEO Paramesh Gopi said in a statement announcing the company’s second-quarter results Tuesday. “Purchase orders continue to grow and backlog is bulding.” Applied Micro is a newcomer to the server-chip market, and AMD has been positioning itself as a company with a lot more experience in building server parts than its competitors in the ARM space. AMD recently hired Karl Freund, a former vice president of marketing for Calxeda, a startup that was also building ARM chips for servers but went bankrupt late last year. Freund is now vice president of server marketing at AMD. AMD’s new Opteron A1100-series developer kit, which features Seattle, is now available upon request to software and hardware developers, as well as operators of large data centers, according to the company’s announcement. Suresh Gopalakrishnan, general manager and vice president of AMD’s server business unit, said the goal of the kit’s release was to grow an ecosystem around the architecture. “After successfully sampling to major ecosystem partners such as firmware, OS and tools providers, we are taking the next step in what will be a collaborative effort across the industry to reimagine the datacenter based on the open business model of ARM innovation,” he said. AMD’s Opteron A1100 supports: AMD also recently announced an effort to design a server where x86 and ARM chips were interchangeable. |
| 2:00p | Data Center Jobs: Amazon.com At the Data Center Jobs Board, we have a new job listing from Amazon.com, which is seeking a Data Center Field Engineer in Ashburn, Virginia. The Data Center Field Engineer is responsible for understanding the design and functionality of the data centers within your assigned region, providing direct influence over all designs for construction of new data centers as well as legacy data center upgrades, driving the design engineering teams standards and specifications, working with internal teams to trouble shoot problems and conduct Root Cause Analysis (RCA) and Corrective Action (CA) for design related problems, and working with local utility companies to understand and coordinate site utility requirements. 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. |
| 4:21p | Canada’s Shaw Communications buys ViaWest for $1.2B Canadian telco Shaw Communications has agreed to buy Colorado-based data center provider ViaWest for $1.2 billion in a massive expansion of its data center play. The current ViaWest management team will stay on board to run the company. Shaw offers cable television and Internet services in addition to operating Canada’s Global Television network. Shaw is acquiring ViaWest from private equity firm Oak Hill Capital partners and other shareholders. Oak Hill acquired ViaWest in 2010. The firm has invested in a number of companies in the data center sector, including Savvis, TelecityGroup and Cincinnati Bell. Shaw is based in Calgary, which is located in the midwest of Canada, and ViaWest gives it a strong position in the U.S. ViaWest has more than 350 employees and 27 data centers in eight key markets. It has been undergoing aggressive expansion. The company recently opened a huge data center in Denver and another one in Chaska, Minnesota. It also opened a data center in Phoenix last February. Canadian rival Telus has also been expanding its data center business. Data centers are an attractive proposition to communications conglomerates, as outsourcing continues to grow. The rational was similar for another Canadian cable provider Cogeco, which acquired Peer 1 Hosting for $482.5 million. Like Cogeco, Shaw doesn’t offer a mobile service and sees data centers as a big revenue opportunity. Beyond Canada, examples of telcos getting into the data center business include Time Warner cable’s acquisition of NaviSite, Verizon’s acquisition of Terremark, TDS Communications’ OneNeck IT deal and NTT’s purchase of RagingWire. The growing data center industry represents an attractive new business line and source of growth for cable conglomerates who operate in commodity industries. “With the acquisition of ViaWest, Shaw gains significant capabilities, scale and immediate expertise in the growing marketplace for enterprise data services,” Brad Shaw, CEO of the Canadian telco, said. Shaw acquired Enmax, which operates fiber-optic networks in Calgary, for about C$225 million last April. |
| 7:21p | Cavium Buys SDN-Focused Switch Silicon Firm Xpliant for $90M In an effort to boost its position in the switch silicon market for software defined networking, chipmaker Cavium has acquired switch silicon company Xpliant, which specializes in processors for SDN applications. Cavium was an prior investor in Xpliant, and the $15 million it invested in the company is included in the total $90 million price tag. The rest consists of $40 million in cash and $35 million in stock. Cavium estimates the Ethernet switch silicon market’s size to be more than $1 billion. Anticipating growth in SDN adoption, the company expects an accelerated refresh cycle for Ethernet switches, which will make the opportunity even greater. The company said current switch solutions, built on a “legacy fixed-function architecture,” would not be able to address requirements of the next-generation data center, enterprise and service-provider networks. Xpliant’s current product portfolio includes 10 Gigabit Etherneet, 40 GbE and 100 GbE switching silicon. The company has designed its products specifically with SDN capabilities in mind, according to Cavium. Cavium said Xpliant’s products would be complementary to its own portfolio, which consists of specialized silicon solutions for networking, wireless, storage and security products. Cavium is also working on bringing to market a 64-bit ARM processor for servers called ThunderX. Cavium President and CEO Syed Ali said Xpliant’s switching silicon solutions were disruptive. “The Xpliant line of products is highly synergistic with Cavium’s existing infrastructure products and will enable Cavium to deliver highly optimized end to end solutions to our customers,” Ali said. Cavium customers include Alcatel-Lucent, Cisco, IBM, Juniper and Qualcomm, among others. |
| 11:05p | Cloud Host Iomart Turns Down Acquisition Offers from Host Europe
This article originally appeared at The WHIR Scottish web hosting and cloud services provider Iomart rejected an offer to buy the company for more than half a billion dollars (or £300 million) earlier this month. German web host Host Europe, according to a statement from Iomart, made two cash offers in June which would pay 2.75 GBP and then 2.85 GBP per Iomart share. When news of this broke last week, Iomart’s stock prices jumped dramatically, closing 7.45 percent higher than the day before. The Host Europe Group consists of several mass-market hosting and domain brands including 123-reg, Webfusion, Heart Internet and Host Europe. This group of companies was acquired by private equity firm Cinven in July 2013 for approximately $667 million. Cinven has since backed Host Europe Group’s acquisition of managed hosting provider Telefónica Germany Online Services GmbH, and domainFACTORY, one of Europe’s largest web hosting companies. Iomart’s rejection of this offer is not an indication that the company is not for sale, but perhaps more likely that the offer was too low. In a statement, Iomart said Host Europe’s offers were rejected because they “undervalued iomart.” In a research note reported by a Scottish newspaper, SP Angel Research suggested £3 per share as a target price, noting Iomart’s “high margins, recurring revenue and strong organic growth,” which should demand a premium for any possible buyer. Host Europe’s attempt to buy Iomart speaks to a consolidation trend in mass market hosting which has seen a great deal of investment from private and public equity. Mass-market web host GoDaddy, which is owned by venture capital firms, was able to make a string of purchases over the past few years including the acquisition of web host Media Temple and domain marketplace Afternic. And Endurance International Group, a publicly-traded web hosting company, bought Indian web host Directi earlier this year. With many potential buyers in the marketplace, it seems that Iomart is in no hurry to sell. This article originally appeared at: http://www.thewhir.com/web-hosting-news/c |
| << Previous Day |
2014/07/31 [Calendar] |
Next Day >> |