Data Center Knowledge | News and analysis for the data center industry - Industr's Journal
[Most Recent Entries]
[Calendar View]
Friday, August 21st, 2015
| Time |
Event |
| 12:00p |
What Keeps Intel’s CISO Up at Night? In a word, complexity.
Complexity is the number-one thing that keeps Brent Conran, Intel’s chief information security officer, up at night.
Talking heads like to pontificate on the proliferation of mobile devices, rush to the cloud, the Internet of Things, and so on. What those trends mean for the enterprise CISO is a whole new way of thinking about their job. Gone are the days when security chiefs simply worried about “securing the perimeter.” It’s not so easy to see the perimeter anymore, and trying to see it is often a useless task.
“Now, where the perimeter starts and ends, it’s pretty malleable,” Conran said. Employees use their own devices at work, they use outside vendors’ cloud services; they’re no longer confined to technology their company’s IT organization has made available to them, and that’s the reality the IT organization has to live with.
The answer is to have a flexible security framework, Conran said, which makes for a great user experience but a lot more complex and tougher to manage from the security staff’s perspective. To deal with that complexity, “you have to be agile, and you have to have transparency,” Conran said. You also have to always be prepared to move on to the next generation of security technology and, in fact, proactively seek it out.
Conran shared his experience overseeing information security at one of the world’s largest corporations in a security keynote at the Intel Developer Forum in San Francisco earlier this week.
As Chris Young, general manager of the Intel Security Group, put it in the same keynote, the nature of the threats has changed. The most malicious attackers are no longer individual hackers but professional organized cyber criminals and nation states.
Instead of viruses planted via email, the main concern today is theft of personally identifiable data, Young said. And cyber criminals have more tools at their disposal today than they ever had and at a lower cost.
 Chris Young, senior VP, GM, Intel Security Group (Photo: Yevgeniy Sverdlik)
Cloud services give attackers access to unlimited compute and connectivity resources and a wider attack surface. “The cloud is the best place to go if you want to leverage or identify a target for attack,” he said.
The proliferation of connected devices — Intel estimates that the 15 billion connected devices today will turn into 200 billion by 2020 — will create more and more opportunities for attackers. Cyber threats will also have more and more impact on the physical world. “It’s my car driving down the road,” Young said. “What happens if a 100-car pileup is generated by a cyber security attack?”
Within Intel’s own IT organization, one thing that helps address the complexity of securing the infrastructure and stay on top of the latest in security tech is partnering with security vendors instead of simply paying them for their products and services, Conran said. Intel has such a partnership with CyberArk Software, a security company whose market debut last September was one of the year’s hottest tech IPOs.
Instead of worrying about securing the perimeter, CyberArk specializes in dealing with threats that are already inside the enterprise network. The company is an expert in privileged account security, Udi Mokady, its president and CEO, said. Hackers gaining access to privileged accounts is one of the leading threats today. Nearly all of the recent high-profile security breaches covered by the mainstream press involved attacker access to privileged accounts, he said.
Intel and CyberArk collaborate and share data in the continuous process of looking for security threats. Sharing data is crucial to the success of these efforts, Mokady and Conran said. A company like Intel can no longer simply rely on the security vendor to identify potential breaches.
A vendor doesn’t know your infrastructure as well as you do, Conran explained. By combining the expertise on both sides of the relationship, it becomes easier to find that “needle in the haystack,” Mokady said. | | 4:57p |
Report: Google, Amazon May Be Eyeing Tata’s Data Center Assets After the rush by telcos to grow their data center services business through both acquisition and construction a few years ago, some of them are now seeking to get out of the market.
Google and Amazon are said to be looking at the data center assets of the Indian telco Tata Communications, a subsidiary of the Mumbai-based conglomerate Tata Group. Indian business news daily The Economic Times reported that both US internet giants recently had talks with the telco’s management about its data center business, citing anonymous sources.
The paper reported last month that Tata had hired US investment banking firm Jefferies to shop its data center business around, saying the assets were worth about $500 million. Friday’s report, however, said the deal was expected to make $650 million to $700 million.
Another giant telco that’s reportedly hoping to sell off its data center services business is AT&T. Reuters reported in February – also citing anonymous sources – that AT&T was mulling the sale of about $2 billion worth of data center assets to pay down debt.
Tata is also reportedly hoping to reduce the size of its debt by selling a 70-plus-percent stake in its data center services business and focusing on its core network carrier segment.
Buying a fleet of data centers that already house other companies’ equipment would be a big strategic change for a Google or an Amazon. The internet giants design build their own so called “web-scale” data centers and lease smaller bits of data center capacity from commercial data center providers in facilities closer to densely populated metros.
Other than their public cloud infrastructure offerings, these companies don’t provide data center services themselves.
Tata says it has 1 million square feet of colocation space across 44 locations worldwide on its website. But the company doesn’t own all of that footprint. Its latest international expansion push, announced last year, was through partnerships with companies like NextDC in Australia, Interxion in Europe, and Pacific Link Telecom in Malaysia.
The highest concentration of Tata data centers is in India, but it has facilities all over the rest of Asia-Pacific, as well as in Europe, South Africa, US, and Canada.
According to The Economic Times, other suitors for Tata’s data center business include Singapore’s Sembcorp Industries and a list of American private equity firms, including the Blackstone Group, Carlyle, KKR, Bain Capital, and Advent International. | | 5:37p |
Weekly DCIM Software News Update: August 21 Altima releases version 15 of its NetZoom DCIM solution, Device42 adds international languages and support for BlueNet PDUs, and Facebook enhances its open source software that handles certain hardware management functions.
Altima releases NetZoom 15. Altima Technologies announced the availability of its NetZoom 15 application, that helps professionals design, diagram and manage IT and facility asset infrastructure. The latest release includes dozens of new asset management and productivity features like auto diagramming, auto generated rack elevations, port-to-port connectivity, powerful device and rack configuration, reporting, and importing and exporting of asset information.
Device42 adds international language support, BlueNet PDU support. Device42 announced it has added Dutch, Japanese, Korean and Portuguese support for Checkpoint Network device support. The company’s DCIM offering can now also auto-discover and monitor BlueNet Monitored PDUs.
Facebook takes over server management control from vendors. Facebook announced that it has enhanced its open source software that handles certain hardware management functions so that it can support one of numerous Facebook server designs. Facebook’s OpenBMC is an open source version of Baseboard Management Controller software used for managing hardware temperature, energy usage and other functions. | | 6:47p |
Verizon Buys Into Mesosphere’s Data Center OS Add Verizon to the list of well-known companies that use a version of Apache Mesos in their data centers. That list also includes Twitter, Apple, Netflix, and Airbnb, among others.
Verizon announced this morning it will implement the commercial distribution of the open source server cluster management software by Mesosphere, the startup driving commercialization of the technology that was born at University of California at Berkeley.
Today is the second and final day of the Linux Foundation’s #MesosCon event in Seattle.
Verizon said Mesosphere’s Datacenter Operating System software will help it build and roll out new services faster and improve scalability and efficiency of those applications.
“Mesosphere DCOS gives Verizon the foundation to create far-reaching benefits by increasing our ability to quickly launch new products and services while reducing the IT requirements in our data centers,” Kumar Vishwanathan, VP and chief technologist for Verizon Labs, said in a statement.
The data center OS abstracts disparate data center resources, such as physical servers and on-premise and cloud VMs, and presents them to applications as a unified pool of resources. It lends itself particularly well to the use of application containers, such as the popular Docker containers.
Technologies and companies like Docker and Mesosphere represent the rising wave of software development and data center operations pioneered by web-scale companies like Google, where new web services and features are created, deployed, and refined quickly and on a constant basis. Used by the web-scale companies for years, this method of development requires a degree of agility on the data center infrastructure side traditional enterprise IT shops aren’t used to, and the likes of Mesosphere hope to help them make the transition and make money in the process.
Here’s more #MesosCon coverage by DCK:
Mesosphere and Cisco Partner on Turnkey Data Center Platform for Open Source Tech
Linux and Windows Servers to Be Cogs in One Data Center OS Wheel | | 7:22p |
Cray Supercomputer to Power Weather Research in Iceland Supercomputer vendor Cray announced it has been awarded a $6 million contract to provide the Danish Meteorological Institute with a Cray XC supercomputer and Sonexion 2000 storage system. As a joint venture between DMI and the Icelandic Meteorological Office, the new system will be installed at the IMO data center in Reykjavik, Iceland.
The Cray XC line of supercomputers has done well in a variety of industries, crunching numbers for government, science, finance, and others, and being used at several leading Top500 sites.
Lately, Cray systems have helped power a number of new installations at worldwide weather and climate centers. Within the last year Cray announced that it has helped the US National Oceanic and Atmospheric Administration boost its supercomputer capacity by a factor of ten and landed its largest contract outside of the US with a $128 million deal with the UK Meteorological office.
“It was paramount to us to find a business partner with the organization and skills to install a supercomputing solution in Iceland that could operate remotely, 1,300 miles away from Denmark, and Cray filled all the criteria,” Marianne Thyrring, director general of the Danish Meteorological Institute, said in a statement.
When in its final configuration, the supercomputer will be ten times more powerful than the current DMI system and will allow the two meteorological services to collaborate on producing high quality numerical weather modeling and predictions, according to Cray.
Speaking to DMI’s involvement with the Danish Ministry of Climate, Thyrring added that they “feel obligated to use 100 percent green energy supply and cooling. Iceland supplies this through geothermal and hydropower.”
System deliveries are expected in 2015 and 2017. | | 7:45p |
Seagate Shuts Down its Cloud Storage Service Wuala 
This article originally appeared at The WHIR
Wuala cloud storage service has been closed down by LaCie, the Seagate subsidiary has announced. The service is closed to purchases and renewals effective immediately, and it will transition to read-only on Sept. 30, 2015 before being completely shuttered on Nov. 15.
A post on Wuala’s website announced the closure, while recommending subscribers migrate to Tresorit, and offering a discount for the service of its former competitor. The post also includes a 15 question FAQ to assist customers in migrating their data.
Seagate told The Inquirer in a statement that is was discontinuing the service to focus on its consumer and SMB cloud partnerships. Wuala will issue discounts and at the end of the process delete all customer data, but the company did not provide details of the deletion process it will follow.
“Wuala’s closure is emblematic of the commoditisation of cloud storage,” Dr Chenxi Wang, VP of cloud security and strategy at CipherCloud told The Inquirer. “We’re likely to see more closures and consolidations, which will mean more end users asking the same questions about their data and business continuity.”
Cloud storage provider Bitcasa was nearly driven out of business by its unlimited storage offering last fall, but as it discontinued that service Hive launched as a free, unlimited cloud storage service. More recently Liquid Web launched a cloud storage product specifically for high volume customers, while security broker Adallom integrated with Dropbox for Business to deliver cloud storage with governance and security controls. Kim Dotcom also recently revealed plans to launch a new cloud storage service, focused on user privacy.
This first ran at http://www.thewhir.com/web-hosting-news/cloud-storage-service-wuala-shutting-down | | 8:44p |
UPDATE: Explosion in Downtown Los Angeles Disrupts Data Center Operations Updated with information about Level 3 service disruption to data center customers, including Internap and Equinix (within the same building), and with a statement from CoreSite.
Call this August the month of explosion-related data center outages.
At least one data center has experienced disruptions following an explosion in a basement of a high-rise in downtown Los Angeles Thursday evening.
While the cause of the explosion is still being investigated, the Los Angeles Fire Department suspects it had something to do with a generator in the basement. The blast at 811 West Wilshire Blvd. took out an on-site power station, leaving 12 buildings in the area without electricity, according to the local utility.
The explosion interrupted connectivity on network infrastructure operated by Level 3 Communications, which serves a lot of data center users in the area, Craig VerColen, a spokesman for LogMeIn, a company whose data center went dark as a result of the incident, said via email. Level 3 issued a statement saying its technicians were working to restore services.
“The Level 3 network is currently experiencing a service disruption due to a commercial power outage in Los Angeles,” the statement read. “Ensuring the stability of our network is our primary concern, and we are dedicated to minimizing customer impact during an outage.”
LogMeIn uses data center services by Internap, which in turn uses an Equinix facility downtown, VerColen said. The Equinix data center is one block away from the site of the explosion. The data center “never lost power,” he said.
Other Internap or Equinix customers in the facility were also affected by the incident, VerColen said.
Equinix and Internap representatives did not respond to requests for comment in time for publication. We’ll update the post once we hear from them.
The incident also affected One Wilshire, one of the most important carrier hotels and data center hubs on the West Coast.
At least a portion of One Wilshire switched to generator power around midnight as a result of the incident, according to a source who works in one of the data centers in the building who wished to stay anonymous. The building switched to generator power “for a short period of time,” they said, but was back on utility power Friday morning.
One of the data center operators at One Wilshire is CoreSite. In a statement, the company’s senior VP of operations Dominic Tobin said the facility failed over to generator power without experiencing a data center outage.
“As with many other buildings in the area, CoreSite’s LA1 data center at One Wilshire experienced a utility outage,” Tobin said. “CoreSite’s LA1 data center was not affected by the loss of utility power due to the seamless transfer we made to emergency power systems in response to this outage.”
LogMeIn, whose data center went offline as a result of connectivity loss, is a Boston-based Software-as-a-Service company that provides remote connectivity for collaboration, IT management, and customer engagement.
“In this area was Level 3, and Level 3 provides a lot of the fiber and connections for many ISPs,” VerColen said. “They were the ones most impacted by the explosion, so it was really the connections between things in the cloud to our data center that were interrupted.”
LogMeIn first tweeted about the outage early Friday morning PDT. In an update about one hour later, the company tweeted that full connectivity was restored.
While the company’s systems failed over to other data centers (it has nine sites total), various aspects of the service were interrupted for about two yours, VerColen said. “Given the time of day, it impacted mostly UK or European customers.”
Two people were hospitalized with minor health issues following the explosion, according to the fire department.
The blast also briefly interrupted a Shania Twain concert that was taking place in the nearby Staples Center.
Another much bigger explosion caused some prolonged data center downtime in Tianjin, China, where a warehouse complex used by a chemicals transportation company erupted multiple times and killed more than 100 people last week.
Data center operators at the nearby National Supercomputing Center shut down the government’s Tianhe-1 supercomputer as a precautionary measure following the blast. China’s official government media reported earlier this week that the supercomputer had been brought back online. |
|