[Most Recent Entries] [Calendar View]
Monday, November 18th, 2013
- 52 percent or more of the power coming into a datacenter is used directly by IT equipment.
- Server utilization rates are typically very low, currently averaging in the 6–12 percent range.
- An idle server that is doing nothing at all can still draw 60 percent of its maximum power.
- One watt of power saved at the server level can generate as much as 2.84 watts of savings along the entire datacenter power chain.
- Encrypt all information that moves between our data centers by the end of Q1 2014;
- Offer users an option to encrypt all data flow to/from Yahoo by the end of Q1 2014;
- Work closely with its international Mail partners to ensure that Yahoo co-branded Mail accounts are https-enabled.
| Time | Event |
| 12:30p | NY Metro Roundup: Xand Completes Expansion of DR Space Last week featured a number of news announcements from data center service providers in the New York/New Jersey market. Here’s a roundup: Xand Completes Expansion of Post-Sandy DR Expansion - Colocation and disaster recovery provider Xand has announced today the completion of a year-long initiative of major new construction, including disaster recovery facilities, raised-floor data center space, and cloud infrastructure to meet an unprecedented surge in demand following Superstorm Sandy. With build-outs in New York, Pennsylvania, and New England, Xand’s 2013 expansion represents a greater than $30 million investment in disaster recovery infrastructure and services. Xand’s post-Sandy expansion agenda began with its Hawthorne, New York facility. To meet demand, Xand quickly added 35,000 square-feet of new disaster recovery seating to Hawthorne in Q1 2013, followed by an additional build-out of 40,000 square feet completed in Q3 2013. In the process of the New York expansion, Xand took occupancy of a second building, transforming raw office space into a state-of-the-art facility dedicated exclusively to disaster recovery seating, suites, and resources. The total 75,000 square-feet of new disaster recovery space joined 11,000 square-feet of premium high-density data center space, enhanced power and network redundancies, and improved cloud infrastructure. United Fiber Establishes Presence at Sabey’s Intergate.Manhattan – Sabey Data Center Properties said this week that United Fiber & Data (UFD), the provider of an alternative fiber network between New York City and the Washington, D.C. region, has signed a licensing agreement to establish two redundant network points of entry at Intergate.Manhattan, Sabey’s new lower Manhattan data center. UFD has built a dark fiber network that connects two large Northeast U.S. Internet hubs – Manhattan and Ashburn, VA — along a route that passes through New Jersey, Pennsylvania and Maryland. UFD’s network bypasses the traditional I-95/Amtrak corridor that most service providers follow. UFD plans to go live at Intergate.Manhattan on December 15, 2013. One of its customers, Datagram, a major web hosting company, is also a colocation tenant at Intergate.Manhattan. NYIIX Expands Connectivity to 165 Halsey – TELEHOUSE America recently announced the expansion of its New York International Internet Exchange (NYIIX) into 165 Halsey Street in Newark, NJ, beginning February 2014. NYIIX is New York’s largest public peering exchange and will now offer six robust locations within the New York City Metro area – 85 10th Avenue, 7 Teleport Drive, 60 Hudson Street, 111 8th Avenue, 32 Avenue of the Americas and 165 Halsey. Customers of NYIIX gain direct access to mainstream traffic, as well as niche, regional markets with powerful network-to-network connections at speeds up to 10 GigE. The exchange’s low latency allows for enhanced network performance and improved connectivity to major providers in a large network space. NYIIX offers optimal primary and secondary IP traffic routing options and is IPV4 and IPV6 capable. “The expansion of NYIIX into 165 Halsey St. allows Telehouse to add another important peering location option for its members,” says Akio Sugeno, Vice President of Internet Engineering, Operations and Business Development at Telehouse. |
| 1:30p | Taking a Truly Holistic View of Data Center Efficiency Aaron Rallo is the founder and CEO of TSO Logic. Aaron has spent the last 15 years building and managing large-scale transactional solutions for online retailers, with both hands-on and C-level responsibility in data centers around the world.  AARON RALLOTSO Logic The word “holistic” comes up frequently in the data center industry, with industry leaders and trade publications urging operators to take a more comprehensive view in their journey to improve energy efficiency. This idea has been leading to some significant advances in energy efficiency at the facility level: better cooling systems, power distribution units, universal power supplies, backup power generation, and so on. These kinds of advances have undoubtedly had a positive impact on energy waste at datacenters. In fact, the average Power Usage Effectiveness [PUE] in the industry dropped from 2.7 in 2007 to 1.65 in 2013. Even so, this facilities-centric approach has fostered one very large blind spot, stopping short of a truly holistic view of the data center. The reality of the situation is that there is an equally large opportunity to improve efficiency on the IT side itself, but this has been largely overlooked. Perhaps you’re thinking that IT energy consumption is simply tied to the size of the data center, and is otherwise set in stone. If you can’t imagine that IT equipment actually consumes a lot of energy unnecessarily, take a look at some of these numbers: Relatively speaking, a single server may not use much energy. But multiply this small amount by thousands or tens of thousands of servers, and it’s clear that that a considerable portion of the energy drawn by data centers is simply being wasted on powering and supporting idle servers. That’s a very valuable opportunity to reduce energy consumption and operating costs. Compounding this waste is the fact that individual servers are growing more power hungry. The International Data Corporation (IDC) reports that energy consumption per server is actually rising by an average of nine percent annually. (Of course, the computing power of each server is growing at the same time, but since data demand continues to explode, the absolute number of servers just keeps growing too.) With all of this in mind, it’s imperative that data centers look at both the facility and the IT infrastructure when they track their energy performance—that is, if they want to remain competitive. On the Cusp of ChangeThe good news is that major change is already starting to happen in the industry, in large part due to the introduction and increasing acceptance of software tools that allow for a deeper understanding of datacenter energy consumption. Although these software solutions cover a wide gamut of types and capabilities, the general emphasis is on detailed, real-time measurement and monitoring, opening the door for improved understanding and finer-grained control. Still, the majority of these software tools focus primarily on the facilities side of the equation. To finally start addressing the IT side, a whole new breed of software technology called application-aware power management (AAPM) is gaining traction. AAPM solutions monitor a datacenter’s current and incoming workload at the server and application level, allowing for dynamic control of the power state of individual servers in direct response to application demand. As a result, much less power is wasted on servers that aren’t performing any real work, without any infrastructure changes or negative impact on performance. |
| 2:00p | Cloud News: OnApp Extends its IaaS Platform VMware vCloud Hybrid receives a HIPAA asessment, HyTrust acquires HighCloud Security, OnApp extends its IaaS platform to external hosts, and GenieDB launches an online management console for geo-distributed MySQL databases. OnApp extends IaaS platform. OnApp announced the general availability of the latest version of its Infrastructure-as-a-Service (IaaS) platform for hosts, telcos and MSPs. The new version of OnApp extends the scope of IaaS beyond public and private cloud to include bare metal servers and Smart Servers – a new server type that combines cloud automation with the performance of dedicated hardware. It also introduces Recipes and Blueprints, two new ways for service providers to automate the deployment and customization of servers, applications and complete application stacks. With OnApp, service providers can make more efficient use of their infrastructure as a single pool of resource that can be provisioned at will, to suit a wide range of customer workloads. “What we call ‘cloud’ today is really about abstracting applications away from the infrastructure, and it’s that abstraction that brings the value people look for when they move workloads to the cloud,” said OnApp CEO, Ditlev Bredahl. “That’s what enables automatic scaling, and failover, and self-service provisioning. Bringing dedicated servers into this environment, and extending cloud automation to other types of infrastructure, is the next logical step.” HyTrust acquires HighCloud Security. Cloud automation provider HyTrust announced that it has acquired HighCloud Security, a leader in cloud encryption and key management software. By combining HyTrust’s powerful administrative visibility and control with HighCloud’s strengths in encryption and key management, the acquisition offers customers of both companies an unprecedented level of flexibility in addressing security, compliance and data privacy requirements in in all cloud environments—private, public and hybrid. The combined offering from HyTrust and HighCloud enables ‘cloaked’ private, hybrid and public clouds and helps address three of the primary security concerns in cloud environments – a broad level of access available to privileged users with malicious intent, breaches, and challenges involved in maintaining the security and privacy of the data itself. “With the increasing prevalence of data breaches, leaks of classified information by insiders, and surveillance in the cloud, data security and privacy are more important than perhaps ever before,” said HyTrust CEO John De Santis. “Cloud computing in all its forms has become the top technology priority for every enterprise, and that’s why we’ve quadrupled our growth at HyTrust in just the past year. By bringing HighCloud Security’s market-leading technologies into the HyTrust family of solutions, we can take to market the strongest protection for virtualized cloud infrastructure available anywhere.” VMware vCloud Hybrid to help meet HIPAA requirements. VMware (VMW) announced Health Insurance Portability and Accountability Act (HIPAA) assessments and attestation for VMware vCloud Hybrid Service, and VMware is now providing HIPAA Business Associate Agreements (BAAs) to customers. BrightLine CPAs & Associates recently completed VMware’s assessment against the relevant HIPAA / Health Information Technology for Economic and Clinical Health Act (HITECH). The auditor also issued an attestation report stating that the information security program for vCloud Hybrid Service meets the applicable HIPAA security and HITECH breach notification requirements. “Healthcare-focused customers and partners can now obtain a HIPAA BAA from VMware for vCloud Hybrid Service and take advantage of an independent third party assessment of vCloud Hybrid Service against HIPAA security controls and breach notification requirements,” said Bill Fathers, senior vice president and general manager, Hybrid Cloud Services Business Unit, VMware. “This significant milestone enables our healthcare customers to not only take advantage of the agility and flexibility of vCloud Hybrid Service, but collaborate, share, store and centralize Protected Health Information (PHI) so they can operate more efficiently, while protecting patient privacy.” GenieDB launches online management to provision geo-distributed MySQL. GenieDB announced the availability of its simple and easy to use Management Console for automated database administration, including backup processes, tuning and updates. Now companies can experience the benefits of GenieDB’s geo-distributed, continuously available MySQL with just a few mouse clicks. With its Globally Distributed MySQL-as-a-Service, companies can launch MySQL database clusters that remain continuously available during regional outages, and provide fast application response time anywhere in the world. “Our management console makes it easy to take advantage of the benefits of GenieDB’s MySQL-as-a-Service,” said Cary Breese, CEO, GenieDB. “By simplifying database creation and administration, companies can stay focused on their core businesses, rather than worrying about outages, availability, and application response time.” |
| 3:00p | Streaming Music Service Rhapsody Expands With Internap Last week we saw a number of customer wins and expansion announcements by leading service providers. Here are some of the noteworthy links : Rhapsody Pursues Global Expansion with Internap – Internap Network Services announced that Rhapsody International, the parent company of leading streaming music services Rhapsody and Napster, is expanding its use of Internap’s high-performance IT infrastructure services. An Internap customer since 2010, Rhapsody International signed a multi-year renewal for colocation and route-optimized IP connectivity services in Internap’s Seattle data center and added managed hosting services in London. “As we continue to expand into new markets across Europe and Latin America, we need to address a new set of customer demands,” said Brian Ringer, chief technology officer at Rhapsody. “Internap provides the right mix of critical IT infrastructure and network optimization to ensure our subscribers consistently enjoy a premium music streaming experience.” 365 Main Signs Managed Service Provider NzingaNet – Data center developer and operator 365 Main Inc. has announced that NZingaNet, a Philadelphia managed cloud service provider, has selected 365 Main’s Philadelphia data center to grow its business. NZingaNet offers nationwide services, focusing on the Pennsylvania, Delaware and the tri-state area, New York, New Jersey and Connecticut. “Serving local businesses is a foundation of our company,” said Chris Dolan, 365 Main CEO. “We’re pleased that NZingaNet recognized our dedication to our customers and theirs, and we look forward to working with them in our other data centers as their business grows.” NZingaNet provides co-location, managed, and wireless services for small to medium-sized businesses and residential customers. DataBank Completes $4 Million Minneapolis Expansion – DataBank, a leading custom data center and multi-tenant colocation provider with facilities in Texas and Minnesota, has released details on the company’s $4 million upgrade of their Minneapolis data center. The data center, part of DataBank’s March 2013 acquisition of VeriSpace, will now encompass more than 18,000 square feet of customer datacenter space. DataBank also developed new resources for customer use which include conference rooms, lounge areas and technical equipment staging facilities. “DataBank’s entrance into Minneapolis through acquisition has put us in a great position to facilitate growth in the market,” said Tim Moore, DataBank’s CEO. Datapipe Helps Lionsgate Manage AWS Deployment – Datapipe, a global provider of managed services and infrastructure for outsourced IT, said this week that Lonsgate Entertainment is leveraging Datapipe’s Managed SharePoint as a Service on Amazon Web Services (AWS) as the cornerstone of its IT collaboration and workflow strategy. Working with Datapipe and AWS, Lionsgate is maximizing the efficiency and reliability of its Microsoft SharePoint environment. “Datapipe provides complementary managed service offerings that enterprise customers need to migrate and manage core IT workloads on AWS,” said Terry Wise, Director, Worldwide Partner Ecosystem, AWS. “Combining extensive Microsoft SharePoint experience, managed services and deep AWS skills, Datapipe is enabling enterprises, like Lionsgate, to quickly deliver innovation and costs savings via AWS.” |
| 3:30p | Diablo, 60East Partner to Build an Advanced Electronic Trading Platform Diablo and 60 East partner to build an advanced electronic trading platform, Scality helps a German Aerospace company archive high-res satellite images, and Radware launches its next generation ADC. Diablo and 60 East build trading platform. Diablo Technologies and 60East Technologies announced that they have entered into a strategic alliance to deliver an advanced, low-latency electronic trading platform. Both companies will partner closely to create a joint solution and align product roadmaps designed to leverage Diablo’s Memory Channel Storage(MCS) architecture and the capabilities of the AMPS messaging platform from 60East Technologies. Diablo’s MCS architecture connects NAND flash directly to the CPU through the server’s memory bus, enabling linear scalability in performance at extremely low-latencies for the most demanding enterprise applications. By placing high-capacity flash directly onto these massively parallel structures, MCS breaks the final latency and IOPS paradigm inherent to existing NAND-based storage solutions. When combined with 60East’s AMPS messaging platform, the two technologies are able to deliver more than four times as much throughput and an order of magnitude of lower latency in the 99.9th percentile. ”We chose to work with 60East because of their reputation for success in mission-critical, high performance computing. The synergy between Diablo and 60East is amplified by our combined commitment to providing innovative, leading-edge solutions that can be leveraged across the entire enterprise,” said Riccardo Badalone, Co-founder and CEO of Diablo Technologies. “Our dynamic partnership integrates the two-best-in-class technologies to deliver a joint platform built for the most demanding, low latency applications.” Scality selected by German Aerospace Center. Scality announced that its RING petabyte scale storage solution has been selected by DLR / German Aerospace Center to host its active archive for large amounts of their high definition satellite images. DLR’s Earth Observation Center uses remote sensing to perform environmental and climate studies. Based on Scality’s success in meeting the stringent demands of DLR’s high performance computing environment, EOC is planning on doubling the capacity of their Scality RING production system. “Scality’s RING provides us with a proven solution unlike any other we have found,” said Professor Peter Reinartz, Head of the Photogrammetry and Image Analysis department. “Scality provides for unlimited growth and high IOPS across many information system formats, including access to traditional file systems and cloud storage resource via CDMI. This is very unusual, and as we grow, it is proving to be an essential element of our workflow and infrastructure.” Radware launches next generation ADC. Radware (RDWR) announced it has launched its next generation Application Delivery Controller (ADC), Alteon NG, built from the ground up to ensure application SLA. The Alteon NG can provide full application SLA assurance by offering dedicated ADC virtual instance (vADC) per app/service, complete isolation between instances which will not impact neighboring application performance, “on-demand” scaling up with no hardware modifications and best-in-class application acceleration features and an innovative security offering. “As a leading fashion retailer that relies on a steady stream of web traffic, the performance on our website is of paramount importance,” says Frank De Sa, IT General Manger of The Just Group. “Since deploying each of our brand’s websites on separate vADCs using Radware’s FastView™ technology, the Alteon NG not only improved the performance of our websites, it has reduced the average page load time by over 30 percent.” |
| 7:49p | Yahoo Will Encrypt All Data Moving Between its Data Centers  Several Yahoo Computing Coop data centers in Lockport, N.Y. In a move to reassure customers about the privacy and security of their data, Yahoo today said it will encrypt all data moving between its data centers. The announcement, made in a Tumblr by Yahoo CEO Marissa Mayer, follows a similar effort by Google. Documents leaked by former NSA sysadmin Edward Snowden suggest that the National Security Agency has attempted to intercept data moving across private intra-data center networks operated by both Google and Yahoo. “We’ve worked hard over the years to earn our users’ trust and we fight hard to preserve it.,” Mayer wrote. “To that end, we recently announced that we will make Yahoo Mail even more secure by introducing https (SSL – Secure Sockets Layer) encryption with a 2048-bit key across our network by January 8, 2014. Today we are announcing that we will extend that effort across all Yahoo products.” Mayer said Yahoo would: Google recently said an existing program to encrypt more data in transit was stepped up in June as Google sought to reassure consumers about the U.S. government’s access to data traveling across its network. |
| 8:00p | A Visual Guide to The World’s Most Powerful Supercomputers  Here’s our visual guide to the November 2013 Top500 list of the world’s most powerful supercomputers. The twice-a-year list of the Top 500 supercomputers documents the most powerful systems on the planet. Many of these supercomputers are striking not just for their processing power, but for their design and appearance as well. For a look at the top finishers in the latest Top 500 list, which was released earlier today at the SC13 supercomputing conference in Denver, Germany, see our photo feature, The Top 10 Supercomputers, Illustrated. |
| 8:05p | China’s Milkyway-2 Remains the World’s Top Supercomputer  A look at the rows of processors inside the Milkway-2 supercomputer in China. China’s Milkyway-2 supercomputer remains the most powerful supercomputer in the world. Milkyway 2, also known as Tianhe-2, has retained the number one ranking on the twice-yearly Top500 list of the most powerful supercomputers in the world. Much of the top 10 rankings remain the same as the June 2013 list, with Milkyway-2 maintaining a performance of 33.86 petaflop/s (quadrillions of calculations per second) on the Linpack benchmark, and the DOE’s Titan performing at 17.59 perflop/s. The DOE’s Sequoia IBM BlueGene/Q, Fujitsu’s K computer and the IBM-powered Mira supercomputer at the DOE’s Argonne Lab round out the top five. The Top500 was announced as the SuperComputing conference gets started in Denver this week. A new entry comes in at number six: Piz Daint, a Cray XC30 system installed at the Swiss National Supercomputing Centre (CSCS) in Lugano, Switzerland and now the most powerful system in Europe. Piz Daint achieved 6.27 Pflop/s on the Linpack benchmark. Piz Daint is also the most energy efficient system in the top 10, consuming a total of 2.33 MW and delivering 2.7 Gflops. Cray announced the Swiss contract earlier in the year, noting that the supercomputer would be transformed to include NVIDIA Kepler K20X GPU accelerators. This YouTube video shows installation of the Cray XC30. A total of 31 systems performed greater than a petaflop/s of performance on the November 2013 list, with the top 17 systems achieving half of the total 250 Petflop/s combined performance. Milkyway-2 and Stampede are using Intel Xeon Phi processors, while Titan and Piz Daint are using NVIDIA GPUs. A total of 53 supercomputers are using accelerator or co-processor technology. Ninety-four percent of the systems use processors with six or more cores and 75 percent have processors with eight or more cores. Intel continues to provide the processors for the largest share (82.4 percent) of TOP500 systems. |
| << Previous Day |
2013/11/18 [Calendar] |
Next Day >> |